No real change here. Instead of using an internal structure to the action rule,
the http return rules are now stored as an http reply. The main change is about
the action type. It is now always set to ACT_CUSTOM. The http reply type is used
to know how to evaluate the rule.
The structure owns an error message, most of time loaded from a file, and
converted to HTX. It is created when an errorfile or errorloc directive is
parsed. It is renamed to avoid ambiguities with http_reply structure.
The http_reply structure is added. It represents a generic HTTP message used as
internal response by HAProxy. It is based on the structure used to store http
return rules. The aim is to store all error messages using this structure, as
well as http return and http deny rules.
TX_CLDENY, TX_CLALLOW, TX_SVDENY and TX_SVALLOW flags are unused. Only
TX_CLTARPIT is used to make the difference between an http deny rule and an http
tarpit rule. So these unused flags are removed.
In the CLI command 'show ssl crt-list', the ssl-min-ver and the
ssl-min-max arguments were always displayed because the dumped versions
were the actual version computed and used by haproxy, instead of the
version found in the configuration.
To fix the problem, this patch separates the variables to have one with
the configured version, and one with the actual version used. The dump
only shows the configured version.
This reverts commit 957ec59571.
As discussed with Emeric, the current syntax is not extensible enough,
this will be turned to a section instead in a forthcoming patch.
The ring to applet communication was only made to deal with CLI functions
but it's generic. Let's have generic appctx functions and have the CLI
rely on these instead. This patch introduces ring_attach_appctx() and
ring_detach_appctx().
A few fields, including a generic list entry, were added to the CLI context
by commit 300decc8d9 ("MINOR: cli: extend the CLI context with a list and
two offsets"). It turns out that the list entry (l0) is solely used to
consult rings and that the generic ring_write() code is restricted to a
consumer on the CLI due to this, which was not the initial intent. Let's
make it a general purpose wait_entry field that is properly initialized
during appctx_init(). This will allow any applet to wait on a ring, not
just the CLI.
Instead of using malloc/free to allocate an HPACK table, let's declare
a pool. However the HPACK size is configured by the H2 mux, so it's
also this one which allocates it after post_check.
This patch adds the new global statement:
ring <name> [desc <desc>] [format <format>] [size <size>] [maxlen <length>]
Creates a named ring buffer which could be used on log line for instance.
<desc> is an optionnal description string of the ring. It will appear on
CLI. By default, <name> is reused to fill this field.
<format> is the log format used when generating syslog messages. It may be
one of the following :
iso A message containing only the ISO date, followed by the text.
The PID, process name and system name are omitted. This is
designed to be used with a local log server.
raw A message containing only the text. The level, PID, date, time,
process name and system name are omitted. This is designed to be
used in containers or during development, where the severity only
depends on the file descriptor used (stdout/stderr). This is
the default.
rfc3164 The RFC3164 syslog message format. This is the default.
(https://tools.ietf.org/html/rfc3164)
rfc5424 The RFC5424 syslog message format.
(https://tools.ietf.org/html/rfc5424)
short A message containing only a level between angle brackets such as
'<3>', followed by the text. The PID, date, time, process name
and system name are omitted. This is designed to be used with a
local log server. This format is compatible with what the systemd
logger consumes.
timed A message containing only a level between angle brackets such as
'<3>', followed by ISO date and by the text. The PID, process
name and system name are omitted. This is designed to be
used with a local log server.
<length> is the maximum length of event message stored into the ring,
including formatted header. If the event message is longer
than <length>, it would be truncated to this length.
<name> is the ring identifier, which follows the same naming convention as
proxies and servers.
<size> is the optionnal size in bytes. Default value is set to BUFSIZE.
Note: Historically sink's name and desc were refs on const strings. But with new
configurable rings a dynamic allocation is needed.
Before this path, they rely directly on ring_write bypassing
a part of the sink API.
Now the maxlen parameter of the log will apply only on the text
message part (and not the header, for this you woud prefer
to use the maxlen parameter on the sink/ring).
sink_write prototype was also reviewed to return the number of Bytes
written to be compliant with the other write functions.
This patch extends the sink_write prototype and code to
handle the rfc5424 and rfc3164 header.
It uses header building tools from log.c. Doing this some
functions/vars have been externalized.
facility and minlevel have been removed from the struct sink
and passed to args at sink_write because they depends of the log
and not of the sink (they remained unused by rest of the code
until now).
since commit c0cdaffaa3 ("REORG: ssl: move ssl_sock_ctx and fix
cross-dependencies issues"), `struct ssl_sock_ctx` was moved in
ssl_sock.h. As it contains a `struct buffer`, including
`common/buffer.h` is now mandatory. I encountered an issue while
including ssl_sock.h on another patch:
include/types/ssl_sock.h:240:16: error: field ‘early_buf’ has incomplete type
240 | struct buffer early_buf; /* buffer to store the early data received */
no backport needed.
Fixes: c0cdaffaa3 ("REORG: ssl: move ssl_sock_ctx and fix
cross-dependencies issues")
Signed-off-by: William Dauchy <w.dauchy@criteo.com>
Add swrate_add_dynamic function which is similar to swrate_add, but more
accurate when calculating moving averages when not enough samples have
been processed yet.
In order to move all SSL sample fetches in another file, moving the
ssl_sock_ctx definition in a .h file is required.
Unfortunately it became a cross dependencies hell to solve, because of
the struct wait_event field, so <types/connection.h> is needed which
created other problems.
Add forward declarations in types/ssl_crtlist.h in order to avoid
circular dependencies. Also remove the listener.h include which is not
needed anymore.
The ssl_sock.c file contains a lot of macros and structure definitions
that should be in a .h. Move them to the more appropriate
types/ssl_sock.h file.
This patch adds the ability to register callbacks for SSL/TLS protocol
messages by using the function ssl_sock_register_msg_callback().
All registered callback functions will be called when observing received
or sent SSL/TLS protocol messages.
The EVP_MD_CTX_create() and EVP_MD_CTX_destroy() functions were renamed to
EVP_MD_CTX_new() and EVP_MD_CTX_free() in OpenSSL 1.1.0, respectively. These
functions are used by the digest converter, introduced by the commit 8e36651ed
("MINOR: sample: Add digest and hmac converters"). So for prior versions of
openssl, macros are used to fallback on old functions.
This patch must only be backported if the commit 8e36651ed is backported too.
This one really ought to be defined in hathreads.h like all other thread
definitions, which is what this patch does. As expected, all files but
one (regex.h) were already including hathreads.h when using THREAD_LOCAL;
regex.h was fixed for this.
This was the last entry in config.h which is now useless.
The setting of CONFIG_HAP_LOCKLESS_POOLS depending on threads and
compat was done in config.h for use only in memory.h and memory.c
where other settings are dealt with. Further, the default pool cache
size was set there from a fixed value instead of being set from
defaults.h
Let's move the decision to enable lockless pools via
CONFIG_HAP_LOCKLESS_POOLS to memory.h, and set the default pool
cache size in defaults.h like other default settings.
This was the next-to-last setting in config.h.
CONFIG_HAP_MEM_OPTIM was introduced with memory pools in 1.3 and dropped
in 1.6 when pools became the only way to allocate memory. Still the
option remained present in config.h. Let's kill it.
It is now possible to use log-format string (or hexadecimal string for the
binary version) to match a content in tcp-check based expect rules. For
hexadecimal log-format string, the conversion in binary is performed after the
string evaluation, during health check execution. The pattern keywords to use
are "string-lf" for the log-format string and "binary-lf" for the hexadecimal
log-format string.
Just like in previous patch, it happens that HA_ATOMIC_UPDATE_MIN() and
HA_ATOMIC_UPDATE_MAX() would evaluate the (val) argument up to 3 times.
However this time it affects both thread and non-thread versions. It's
strange because the copy was properly performed for the (new) argument
in order to avoid this. Anyway it was done for the "val" one as well.
A quick code inspection showed that this currently has no effect as
these macros are fairly limited in usage.
It would be best to backport this for long-term stability (till 1.8)
but it will not fix an existing bug.
When threads are disabled, HA_ATOMIC_CAS() becomes a simple compound
expression. However this expression presents a problem, which is that
its arguments are evaluated multiple times, once for the comparison
and once again for the assignement. This presents a risk of performing
some side-effect operations twice in the non-threaded case (e.g. in
case of auto-increment or function return).
The macro was rewritten using local copies for arguments like the
other macros do.
Fortunately a complete inspection of the code indicates that this case
currently never happens. It was however responsible for the strict-aliasing
warning emitted when building fd.c without threads but with 64-bit CAS.
This may be backported as far as 1.8 though it will not fix any existing
bug and is more of a long-term safety measure in case a future fix would
depend on this behavior.
It is now possible to add http-check expect rules matching HTTP header names and
values. Here is the format of these rules:
http-check expect header name [ -m <meth> ] <name> [log-format] \
[ value [ -m <meth> ] <value> [log-format] [full] ]
the name pattern (name ...) is mandatory but the value pattern (value ...) is
optionnal. If not specified, only the header presence is verified. <meth> is the
matching method, applied on the header name or the header value. Supported
matching methods are:
* "str" (exact match)
* "beg" (prefix match)
* "end" (suffix match)
* "sub" (substring match)
* "reg" (regex match)
If not specified, exact matching method is used. If the "log-format" option is
used, the pattern (<name> or <value>) is evaluated as a log-format string. This
option cannot be used with the regex matching method. Finally, by default, the
header value is considered as comma-separated list. Each part may be tested. The
"full" option may be used to test the full header line. Note that matchings are
case insensitive on the header names.
It is now possible to use different matching methods to look for header names in
an HTTP message:
* The exact match. It is the default method. http_find_header() uses this
method. http_find_str_header() is an alias.
* The prefix match. It evals the header names starting by a prefix.
http_find_pfx_header() must be called to use this method.
* The suffix match. It evals the header names ending by a suffix.
http_find_sfx_header() must be called to use this method.
* The substring match. It evals the header names containing a string.
http_find_sub_header() must be called to use this method.
* The regex match. It evals the header names matching a regular expression.
http_match_header() must be called to use this method.
Some HTTP sample fetches will be accessible from the context of a http-check
health check. Thus, the prefetch function responsible to return the HTX message
has been update to handle a check, in addition to a channel. Both cannot be used
at the same time. So there is no ambiguity.
Given that a "count" value of 32M was seen in _shctx_wait4lock(), it
is very important to prevent this from happening again. It's absolutely
essential to prevent the value from growing unbounded because with an
increase of the number of threads, the number of successive failed
attempts will necessarily grow.
Instead now we're scanning all 2^p-1 values from 3 to 255 and are
bounding to count to 255 so that in the worst case each thread tries an
xchg every 255 failed read attempts. That's one every 4 on average per
thread when there are 64 threads, which corresponds to the initial count
of 4 for the first attempt so it seems like a reasonable value to keep a
low latency.
The bug was introduced with the shctx entries in 1.5 so the fix must
be backported to all versions. Before 1.8 the function was called
_shared_context_wait4lock() and was in shctx.c.
Jérôme reported an amazing crash in the spinlock version of
_shctx_wait4lock() with an extremely high <count> value of 32M! The
root cause is that the function cannot deal with contention on the lock
at all because it forgets to check if the lock's value has changed! As
such, every time it's called due to a contention, it waits twice as
long before trying again and lets the caller check for the contention
by itself.
The correct thing to do is to compare the value again at each loop.
This way it makes sure to mostly perform read accesses on the shared
cache line without writing too often, and to be ready fast enough to
try to grab the lock. And we must not increase the count on success
either!
Unfortunately I'd have expected to see a performance boost on the cache
with this but there was absolutely no change, so it's very likely that
these issues only happen once in a while and are sufficient to derail
the process when they strike, but not to have a permanent performance
impact.
The bug was introduced with the shctx entries in 1.5 so the fix must
be backported to all versions. Before 1.8 the function was called
_shared_context_wait4lock() and was in shctx.c.
I changed my mind twice on this one and pushed after the last test with
threads disabled, without re-enabling long long, causing this rightful
build warning.
This needs to be backported if the previous commit ff64d3b027 ("MINOR:
threads: export the POSIX thread ID in panic dumps") is backported as
well.
It is very difficult to map a panic dump against a gdb thread dump
because the thread numbers do not match. However gdb provides the
pthread ID but this one is supposed to be opaque and not to be cast
to a scalar.
This patch provides a fnuction, ha_get_pthread_id() which retrieves
the pthread ID of the indicated thread and casts it to an unsigned
long long so as to lose the least possible amount of information from
it. This is done cleanly using a union to maintain alignment so as
long as these IDs are stored on 1..8 bytes they will be properly
reported. This ID is now presented in the panic dumps so it now
becomes possible to map these threads. When threads are disabled,
zero is returned. For example, this is a panic dump:
Thread 1 is about to kill the process.
*>Thread 1 : id=0x7fe92b825180 act=0 glob=0 wq=1 rq=0 tl=0 tlsz=0 rqsz=0
stuck=1 prof=0 harmless=0 wantrdv=0
cpu_ns: poll=5119122 now=2009446995 diff=2004327873
curr_task=0xc99bf0 (task) calls=4 last=0
fct=0x592440(task_run_applet) ctx=0xca9c50(<CLI>)
strm=0xc996a0 src=unix fe=GLOBAL be=GLOBAL dst=<CLI>
rqf=848202 rqa=0 rpf=80048202 rpa=0 sif=EST,200008 sib=EST,204018
af=(nil),0 csf=0xc9ba40,8200
ab=0xca9c50,4 csb=(nil),0
cof=0xbf0e50,1300:PASS(0xc9cee0)/RAW((nil))/unix_stream(20)
cob=(nil),0:NONE((nil))/NONE((nil))/NONE(0)
call trace(20):
| 0x59e4cf [48 83 c4 10 5b 5d 41 5c]: wdt_handler+0xff/0x10c
| 0x7fe92c170690 [48 c7 c0 0f 00 00 00 0f]: libpthread:+0x13690
| 0x7ffce29519d9 [48 c1 e2 20 48 09 d0 48]: linux-vdso:+0x9d9
| 0x7ffce2951d54 [eb d9 f3 90 e9 1c ff ff]: linux-vdso:__vdso_gettimeofday+0x104/0x133
| 0x57b484 [48 89 e6 48 8d 7c 24 10]: main+0x157114
| 0x50ee6a [85 c0 75 76 48 8b 55 38]: main+0xeaafa
| 0x50f69c [48 63 54 24 20 85 c0 0f]: main+0xeb32c
| 0x59252c [48 c7 c6 d8 ff ff ff 44]: task_run_applet+0xec/0x88c
Thread 2 : id=0x7fe92b6e6700 act=0 glob=0 wq=0 rq=0 tl=0 tlsz=0 rqsz=0
stuck=0 prof=0 harmless=1 wantrdv=0
cpu_ns: poll=786738 now=1086955 diff=300217
curr_task=0
Thread 3 : id=0x7fe92aee5700 act=0 glob=0 wq=0 rq=0 tl=0 tlsz=0 rqsz=0
stuck=0 prof=0 harmless=1 wantrdv=0
cpu_ns: poll=828056 now=1129738 diff=301682
curr_task=0
Thread 4 : id=0x7fe92a6e4700 act=0 glob=0 wq=0 rq=0 tl=0 tlsz=0 rqsz=0
stuck=0 prof=0 harmless=1 wantrdv=0
cpu_ns: poll=818900 now=1153551 diff=334651
curr_task=0
And this is the gdb output:
(gdb) info thr
Id Target Id Frame
* 1 Thread 0x7fe92b825180 (LWP 15234) 0x00007fe92ba81d6b in raise () from /lib64/libc.so.6
2 Thread 0x7fe92b6e6700 (LWP 15235) 0x00007fe92bb56a56 in epoll_wait () from /lib64/libc.so.6
3 Thread 0x7fe92a6e4700 (LWP 15237) 0x00007fe92bb56a56 in epoll_wait () from /lib64/libc.so.6
4 Thread 0x7fe92aee5700 (LWP 15236) 0x00007fe92bb56a56 in epoll_wait () from /lib64/libc.so.6
We can clearly see that while threads 1 and 2 are the same, gdb's
threads 3 and 4 respectively are haproxy's threads 4 and 3.
This may be backported to 2.0 as it removes some confusion in github issues.
It can be sometimes useful to measure total time of a request as seen
from an end user, including TCP/TLS negotiation, server response time
and transfer time. "Tt" currently provides something close to that, but
it also takes client idle time into account, which is problematic for
keep-alive requests as idle time can be very long. "Ta" is also not
sufficient as it hides TCP/TLS negotiationtime. To improve that, introduce
a "Tu" timer, without idle time and everything else. It roughly estimates
time spent time spent from user point of view (without DNS resolution
time), assuming network latency is the same in both directions.
This bug was introduced by the commit 2444aa5b ("MEDIUM: sessions: Don't be
responsible for connections anymore."). In session_check_idle_conn(), when the
mux is destroyed, its context must be passed as argument instead of the
connection.
It is de 2.2-dev bug. No need to backport.
It is now possible to match on a comma-separated list of status codes or range
of codes. In addtion, instead of a string comparison to match the response's
status code, a integer comparison is performed. Here is an example:
http-check expect status 200,201,300-310
This reverts commit 1979943c30ef285ed04f07ecf829514de971d9b2.
Captures in comment was only used when a tcp-check expect based on a negative
regex matching failed to eventually report what was captured while it was not
expected. It is a bit far-fetched to be useable IMHO. on-error and on-success
log-format strings are far more usable. For now there is few check sample
fetches (in fact only one...). But it could be really powerful to report info in
logs.
Since all tcp-check rulesets are globally stored, it is a problem to use
list. For configuration with many backends, the lookups in list may be costly
and slow downs HAProxy startup. To solve this problem, tcp-check rulesets are
now stored in a tree.
The patch is not obvious at the first glance. But it is just a reorg. Functions
have been grouped and ordered in a more logical way. Some structures and flags
are now private to the checks module (so moved from the .h to the .c file).
Defaut health-checks, without any option, doing only a connection check, are now
based on tcp-checks. An implicit default tcp-check connect rule is used. A
shared tcp-check ruleset, name "*tcp-check" is created to support these checks.
This function is unused for now. But it will have be used to install a mux for
an outgoing connection openned in a health-check context. In this case, the
session's origin is the check itself, and it is used to know the mode, HTTP or
TCP, depending on the tcp-check type and not the proxy mode. The check is also
used to get the mux protocol if configured.
It is not set and not used for now, but it will be possible to force the mux
protocol thanks to this patch. A mux proto field is added to the checks and to
tcp-check connect rules.
HTTP health-checks are now internally based on tcp-checks. Of course all the
configuration parsing of the "http-check" keyword and the httpchk option has
been rewritten. But the main changes is that now, as for tcp-check ruleset, it
is possible to perform several send/expect sequences into the same
health-checks. Thus the connect rule is now also available from HTTP checks, jst
like set-var, unset-var and comment rules.
Because the request defined by the "option httpchk" line is used for the first
request only, it is now possible to set the method, the uri and the version on a
"http-check send" line.
Instead of having 2 independent integers, used as boolean values, to know if the
expect rule is invered and to know if the matching regexp has captures, we know
use a 32-bits bitfield.
All tcp-check rules are now stored in the globla shared list. The ones created
to parse a specific protocol, for instance redis, are already stored in this
list. Now pure tcp-check rules are also stored in it. The ruleset name is
created using the proxy name and its config file and line. tcp-check rules
declared in a defaults section are also stored this way using "defaults" as
proxy name.
For now, all tcp-check ruleset are stored in a list. But it could be a bit slow
to looks for a specific ruleset with a huge number of backends. So, it could be
a good idea to use a tree instead.
It is now possible to specified the healthcheck status to use on success of a
tcp-check rule, if it is the last evaluated rule. The option "ok-status"
supports "L4OK", "L6OK", "L7OK" and "L7OKC" status.
A shared tcp-check ruleset is now created to support agent checks. The following
sequence is used :
tcp-check send "%[var(check.agent_string)] log-format
tcp-check expect custom
The custom function to evaluate the expect rule does the same that it was done
to handle agent response when a custom check was used.
A share tcp-check ruleset is now created to support SPOP checks. This way no
extra memory is used if several backends use a SPOP check.
The following sequence is used :
tcp-check send-binary SPOP_REQ
tcp-check expect custom min-recv 4
The spop request is the result of the function
spoe_prepare_healthcheck_request() and the expect rule relies on a custom
function calling spoe_handle_healthcheck_response().
A shared tcp-check ruleset is now created to support LDAP check. This way no
extra memory is used if several backends use a LDAP check.
The following sequance is used :
tcp-check send-binary "300C020101600702010304008000"
tcp-check expect rbinary "^30" min-recv 14 \
on-error "Not LDAPv3 protocol"
tcp-check expect custom
The last expect rule relies on a custom function to check the LDAP server reply.
A share tcp-check ruleset is now created to support MySQL checks. This way no
extra memory is used if several backends use a MySQL check.
One for the following sequence is used :
## If no extra params are set
tcp-check connect default linger
tcp-check expect custom ## will test the initial handshake
## If the username is defined
tcp-check connect default linger
tcp-check send-binary MYSQL_REQ log-format
tcp-check expect custom ## will test the initial handshake
tcp-check expect custom ## will test the reply to the client message
The log-format hexa string MYSQL_REQ depends on 2 preset variables, the packet
header containing the packet length and the sequence ID (check.header) and the
username (check.username). If is also different if the "post-41" option is set
or not. Expect rules relies on custom functions to check MySQL server packets.
A shared tcp-check ruleset is now created to support postgres check. This way no
extra memory is used if several backends use a pgsql check.
The following sequence is used :
tcp-check connect default linger
tcp-check send-binary PGSQL_REQ log-format
tcp-check expect !rstring "^E" min-recv 5 \
error-status "L7RSP" on-error "%[check.payload(6,0)]"
tcp-check expect rbinary "^520000000800000000 min-recv "9" \
error-status "L7STS" \
on-success "PostgreSQL server is ok" \
on-error "PostgreSQL unknown error"
The log-format hexa string PGSQL_REQ depends on 2 preset variables, the packet
length (check.plen) and the username (check.username).
A share tcp-check ruleset is now created to support smtp checks. This way no
extra memory is used if several backends use a smtp check.
The following sequence is used :
tcp-check connect default linger
tcp-check expect rstring "^[0-9]{3}[ \r]" min-recv 4 \
error-status "L7RSP" on-error "%[check.payload(),cut_crlf]"
tcp-check expect rstring "^2[0-9]{2}[ \r]" min-recv 4 \
error-status "L7STS" \
on-error %[check.payload(4,0),ltrim(' '),cut_crlf] \
status-code "check.payload(0,3)"
tcp-echeck send "%[var(check.smtp_cmd)]\r\n" log-format
tcp-check expect rstring "^2[0-9]{2}[- \r]" min-recv 4 \
error-status "L7STS" \
on-error %[check.payload(4,0),ltrim(' '),cut_crlf] \
on-success "%[check.payload(4,0),ltrim(' '),cut_crlf]" \
status-code "check.payload(0,3)"
The variable check.smtp_cmd is by default the string "HELO localhost" by may be
customized setting <helo> and <domain> parameters on the option smtpchk
line. Note there is a difference with the old smtp check. The server gretting
message is checked before send the HELO/EHLO comand.
A shared tcp-check ruleset is now created to support ssl-hello check. This way
no extra memory is used if several backends use a ssl-hello check.
The following sequence is used :
tcp-check send-binary SSLV3_CLIENT_HELLO log-format
tcp-check expect rbinary "^1[56]" min-recv 5 \
error-status "L6RSP" tout-status "L6TOUT"
SSLV3_CLIENT_HELLO is a log-format hexa string representing a SSLv3 CLIENT HELLO
packet. It is the same than the one used by the old ssl-hello except the sample
expression "%[date(),htonl,hex]" is used to set the date field.
A share tcp-check ruleset is now created to support redis checks. This way no
extra memory is used if several backends use a redis check.
The following sequence is used :
tcp-check send "*1\r\n$4\r\nPING\r\n"
tcp-check expect string "+PONG\r\n" error-status "L7STS" \
on-error "%[check.payload(),cut_crlf]" on-success "Redis server is ok"
It is now possible to set a custom function to evaluate a tcp-check expect
rule. It is an internal and not documentd option because the right pointer of
function must be set and it is not possible to express it in the
configuration. It will be used to convert some protocol healthchecks to
tcp-checks.
Custom functions must have the following signature:
enum tcpcheck_eval_ret (*custom)(struct check *, struct tcpcheck_rule *, int);
A list of variables is now associated to each tcp-check ruleset. It is more a
less a list of set-var expressions. This list may be filled during the
configuration parsing. The listed variables will then be set during each
execution of the tcp-check healthcheck, at the begining, before execution of the
the first tcp-check rule.
This patch is mandatory to convert all protocol checks to tcp-checks. It is a
way to customize shared tcp-check rulesets.
This option defines a sample expression, evaluated as an integer, to set the
status code (check->code) if a tcp-check healthcheck ends on the corresponding
expect rule.
These options define log-format strings used to produce the info message if a
tcp-check expect rule fails (on-error option) or succeeds (on-success
option). For this last option, it must be the ending rule, otherwise the
parameter is ignored.
It is now possible to specified the healthcheck status to use on error or on
timeout for tcp-check expect rules. First, to define the error status, the
option "error-status" must be used followed by "L4CON", "L6RSP", "L7RSP" or
"L7STS". Then, to define the timeout status, the option "tout-status" must be
used followed by "L4TOUT", "L6TOUT" or "L7TOUT".
These options will be used to convert specific protocol healthchecks (redis,
pgsql...) to tcp-check ones.
x
A global list to tcp-check ruleset can now be used to share common rulesets with
all backends without any duplication. It is mandatory to convert all specific
protocol checks (redis, pgsql...) to tcp-check healthchecks.
To do so, a flag is now attached to each tcp-check ruleset to know if it is a
shared ruleset or not. tcp-check rules defined in a backend are still directly
attached to the proxy and not shared. In addition a second flag is used to know
if the ruleset is inherited from the defaults section.
An extra parameter for tcp-check send rules can be specified to handle the
string or the hexa string as a log-format one. Using "log-format" option,
instead of considering the data to send as raw data, it is parsed as a
log-format string. Thus it is possible to call sample fetches to customize data
sent to a server. Of course, because we have no stream attached to healthchecks,
not all sample fetches are available. So be careful.
tcp-check set-var(check.port) int(8000)
tcp-check set-var(check.uri) str(/status)
tcp-check connect port var(check.port)
tcp-check send "GET %[check.uri] HTTP/1.0\r\n" log-format
tcp-check send "Host: %[srv_name]\r\n" log-format
tcp-check send "\r\n"
Since we have a session attached to tcp-check healthchecks, It is possible use
sample expression and variables. In addition, it is possible to add tcp-check
set-var rules to define custom variables. So, now, a sample expression can be
used to define the port to use to establish a connection for a tcp-check connect
rule. For instance:
tcp-check set-var(check.port) int(8888)
tcp-check connect port var(check.port)
With this option, it is now possible to use a specific address to open the
connection for a tcp-check connect rule. If the port option is also specified,
it is used in priority.
With this option, it is possible to establish the connection opened by a
tcp-check connect rule using upstream socks4 proxy. Info from the socks4
parameter on the server are used.
Register the custom action rules "set-var" and "unset-var", that will
call the parse_store() command upon parsing.
These rules are thus built and integrated to the tcp-check ruleset, but
have no further effect for the moment.
Add a dedicated vars scope for checks. This scope is considered as part of the
session scope for accounting purposes.
The scope can be addressed by a valid session, even embryonic. The stream is not
necessary.
The scope is initialized after the check session is created. All variables are
then pruned before the session is destroyed.
Create a session for each healthcheck relying on a tcp-check ruleset. When such
check is started, a session is allocated, which will be freed when the check
finishes. A dummy static frontend is used to create these sessions. This will be
useful to support variables and sample expression. This will also be used,
later, by HTTP healthchecks to rely on HTTP muxes.
The loop in tcpcheck_main() function is quite hard to understand. Depending
where we are in the loop, The current_step is the currentely executed rule or
the one to execute on the next call to tcpcheck_main(). When the check result is
reported, we rely on the rule pointed by last_started_step or the one pointed by
current_step. In addition, the loop does not use the common list_for_each_entry
macro and it is thus quite confusing.
So the loop has been totally rewritten and splitted to several functions to
simplify its reading and its understanding. Tcp-check rules are evaluated in
dedicated functions. And a common for_each loop is used and only one rule is
referenced, the current one.
After the configuration parsing, when its validity check, an implicit tcp-check
connect rule is added in front of the tcp-check ruleset if the first non-comment
rule is not a connect one. This implicit rule is flagged to use the default
check parameter.
This means now, all tcp-check rulesets begin with a connect and are never
empty. When tcp-check healthchecks are used, all connections are thus handled by
tcpcheck_main() function.
To allow reusing these blocks without consuming more memory, their list
should be static and share-able accross uses. The head of the list will
be shared as well.
It is thus necessary to extract the head of the rule list from the proxy
itself. Transform it into a pointer instead, that can be easily set to
an external dynamically allocated head.
Parse back-references in comments of tcp-check expect rules. If references are
made, capture groups in the match and replace references to it within the
comment when logging the error. Both text and binary regex can caputre groups
and reference them in the expect rule comment.
[Cf: I slightly updated the patch. exp_replace() function is used instead of a
custom one. And if the trash buffer is too small to contain the comment during
the substitution, the comment is ignored.]
The rbinary match works similarly to the rstring match type, however the
received data is rewritten as hex-string before the match operation is
done.
This allows using regexes on binary content even with the POSIX regex
engine.
[Cf: I slightly updated the patch. mem2hex function was removed and dump_binary
is used instead.]
Allow declaring tcpcheck connect commands with a new parameter,
"linger". This option will configure the connection to avoid using an
RST segment to close, instead following the four-way termination
handshake. Some servers would otherwise log each healthcheck as
an error.
Some expect rules cannot be satisfied due to inherent ambiguity towards
the received data: in the absence of match, the current behavior is to
be forced to wait either the end of the connection or a buffer full,
whichever comes first. Only then does the matching diagnostic is
considered conclusive. For instance :
tcp-check connect
tcp-check expect !rstring "^error"
tcp-check expect string "valid"
This check will only succeed if the connection is closed by the server before
the check timeout. Otherwise the first expect rule will wait for more data until
"^error" regex matches or the check expires.
Allow the user to explicitly define an amount of data that will be
considered enough to determine the value of the check.
This allows succeeding on negative rstring rules, as previously
in valid condition no match happened, and the matching was repeated
until the end of the connection. This could timeout the check
while no error was happening.
[Cf: I slighly updated the patch. The parameter was renamed and the value is a
signed integer to support -1 as default value to ignore the parameter.]
When receiving additional data while chaining multiple tcp-check expects,
previous inverse expects might have a different result with the new data. They
need to be evaluated again against the new data.
Add a pointer to the first inverse expect rule of the current expect chain
(possibly of length one) to each expect rule. When receiving new data, the
currently evaluated tcp-check rule is set back to this pointed rule.
Fonctionnaly speaking, it is a bug and it exists since the introduction of the
feature. But there is no way for now to hit it because when an expect rule does
not match, we wait for more data, independently on the inverse flag. The only
way to move to the following rule is to be sure no more data will be received.
This patch depends on the commit "MINOR: mini-clist: Add functions to iterate
backward on a list".
[Cf: I slightly updated the patch. First, it only concerns inverse expect
rule. Normal expect rules are not concerned. Then, I removed the BUG tag
because, for now, it is not possible to move to the following rule when the
current one does not match while more data can be received.]
Replace the generic integer with an enumerated list. This allows light
type check and helps debugging (seeing action = 2 in the struct is not
helpful).
This options is used to force a non-SSL connection to check a SSL server or to
invert a check-ssl option inherited from the default section. The use_ssl field
in the check structure is used to know if a SSL connection must be used
(use_ssl=1) or not (use_ssl=0). The server configuration is used by default.
The problem is that we cannot distinguish the default case (no specific SSL
check option) and the case of an explicit non-SSL check. In both, use_ssl is set
to 0. So the server configuration is always used. For a SSL server, when
no-check-ssl option is set, the check is still performed using a SSL
configuration.
To fix the bug, instead of a boolean value (0=TCP, 1=SSL), we use a ternary value :
* 0 = use server config
* 1 = force SSL
* -1 = force non-SSL
The same is done for the server parameter. It is not really necessary for
now. But it is a good way to know is the server no-ssl option is set.
In addition, the PR_O_TCPCHK_SSL proxy option is no longer used to set use_ssl
to 1 for a check. Instead the flag is directly tested to prepare or destroy the
server SSL context.
This patch should be backported as far as 1.8.
The 'http-check send' directive have been added to add headers and optionnaly a
payload to the request sent during HTTP healthchecks. The request line may be
customized by the "option httpchk" directive but there was not official way to
add extra headers. An old trick consisted to hide these headers at the end of
the version string, on the "option httpchk" line. And it was impossible to add
an extra payload with an "http-check expect" directive because of the
"Connection: close" header appended to the request (See issue #16 for details).
So to make things official and fully support payload additions, the "http-check
send" directive have been added :
option httpchk POST /status HTTP/1.1
http-check send hdr Content-Type "application/json;charset=UTF-8" \
hdr X-test-1 value1 hdr X-test-2 value2 \
body "{id: 1, field: \"value\"}"
When a payload is defined, the Content-Length header is automatically added. So
chunk-encoded requests are not supported yet. For now, there is no special
validity checks on the extra headers.
This patch is inspired by Kiran Gavali's work. It should fix the issue #16 and
as far as possible, it may be backported, at least as far as 1.8.
list_for_each_entry_rev() and list_for_each_entry_from_rev() and corresponding
safe versions have been added to iterate on a list in the reverse order. All
these functions work the same way than the forward versions, except they use the
.p field to move for an element to another.
Server address and port may change at runtime. So the address and port passed as
arguments and as environment variables when an external check is executed must
be updated. The current number of connections on the server was already updated
before executing the command. So the same mechanism is used for the server
address and port. But in addition, command arguments are also updated.
This patch must be backported to all stable versions. It should fix the
issue #577.
The url_decode() function used by the url_dec converter and a few other
call points is ambiguous on its processing of the '+' character which
itself isn't stable in the spec. This one belongs to the reserved
characters for the query string but not for the path nor the scheme,
in which it must be left as-is. It's only in argument strings that
follow the application/x-www-form-urlencoded encoding that it must be
turned into a space, that is, in query strings and POST arguments.
The problem is that the function is used to process full URLs and
paths in various configs, and to process query strings from the stats
page for example.
This patch updates the function to differentiate the situation where
it's parsing a path and a query string. A new argument indicates if a
query string should be assumed, otherwise it's only assumed after seeing
a question mark.
The various locations in the code making use of this function were
updated to take care of this (most call places were using it to decode
POST arguments).
The url_dec converter is usually called on path or url samples, so it
needs to remain compatible with this and will default to parsing a path
and turning the '+' to a space only after a question mark. However in
situations where it would explicitly be extracted from a POST or a
query string, it now becomes possible to enforce the decoding by passing
a non-null value in argument.
It seems to be what was reported in issue #585. This fix may be
backported to older stable releases.
As reported in issue #596, the edx register isn't marked as clobbered
in div64_32(), which could technically allow gcc to try to reuse it
if it needed a copy of the 32 highest bits of the o1 register after
the operation.
Two attempts were tried, one using a dummy 32-bit local variable to
store the intermediary edx and another one switching to "=A" and making
result a long long. It turns out the former makes the resulting object
code significantly dirtier while the latter makes it better and was
kept. This is due to gcc's difficulties at working with register pairs
mixing 32- and 64- bit values on i386. It was verified that no code
change happened at all on x86_64, armv7, aarch64 nor mips32.
In practice it's only used by the frequency counters so this bug
cannot even be triggered but better fix it.
This may be backported to stable branches though it will not fix any
issue.
If haproxy fails to start and emits an alert, then it can be useful
to have it also emit the version and the path used to load it. Some
users may be mistakenly launching the wrong binary due to a misconfigured
PATH variable and this will save them some troubleshooting time when it
reports that some keywords are not understood.
What we do here is that we *try* to extract the binary name from the
AUX vector on glibc, and we report this as a NOTICE tag before the
very first alert is emitted.
Since some systems switched to service managers which hide all warnings
by default, some users are not aware of some possibly important warnings
and get caught too late with errors that could have been detected earlier.
This patch adds a new global keyword, "zero-warning" and an equivalent
command-line option "-dW" to refuse to start in case any warning is
detected. It is recommended to use these with configurations that are
managed by humans in order to catch mistakes very early.
This helps quickly checking if the config produces any warning. For
this we reuse the "warned" bit field to add a new WARN_ANY bit that is
set by ha_warning(). The rest of the bit field was also cleaned from
unused bits.
Before supporting "server" line in "peers" section, such sections without
any local peer were removed from the configuration to get it validated.
This patch fixes the issue where a "server" line without address and port which
is a remote peer without address and port makes the configuration parsing fail.
When encoutering such cases we now ignore such lines remove them from the
configuration.
Thank you to Jérôme Magnin for having reported this bug.
Must be backported to 2.1 and 2.0.
In 'commit ssl cert', instead of trying to regenerate a list of filters
from the SNIs, use the list provided by the crtlist_entry used to
generate the ckch_inst.
This list of filters doesn't need to be free'd anymore since they are
always reused from the crtlist_entry.
Use the refcount of the SSL_CTX' to free them instead of freeing them on
certains conditions. That way we can free the SSL_CTX everywhere its
pointer is used.
The dump and show ssl crt-list commands does the same thing, they dump
the content of a crt-list, but the 'show' displays an ID in the first
column. Delete the 'dump' command so it is replaced by the 'show' one.
The old 'show' command is replaced by an '-n' option to dump the ID.
And the ID which was a pointer is replaced by a line number and placed
after colons in the filename.
Example:
$ echo "show ssl crt-list -n kikyo.crt-list" | socat /tmp/sock1 -
# kikyo.crt-list
kikyo.pem.rsa:1 secure.domain.tld
kikyo.pem.ecdsa:2 secure.domain.tld
This patch fixes a bad stop condition when decoding a protocol buffer variable integer
whose maximum lenghts are 10, shifting a uint64_t value by more than 63.
Thank you to Ilya for having reported this issue.
Must be backported to 2.1 and 2.0.
When updating a ckch_store we may want to update its pointer in the
crtlist_entry which use it. To do this, we need the list of the entries
using the store.
The instances were wrongly inserted in the crtlist entries, all
instances of a crt-list were inserted in the last crt-list entry.
Which was kind of handy to free all instances upon error.
Now that it's done correctly, the error path was changed, it must
iterate on the entries and find the ckch_insts which were generated for
this bind_conf. To avoid wasting time, it stops the iteration once it
found the first unsuccessful generation.
In order to be able to add new certificate in a crt-list, we need the
list of bind_conf that uses this crt-list so we can create a ckch_inst
for each of them.
Add a counter to know the current number of used connections, as well as the
max, this will be used later to refine the algorithm used to kill idle
connections, based on current usage.
With server-template was introduced the possibility to scale the
number of servers in a backend without needing a configuration change
and associated reload. On the other hand it became impractical to
write use-server rules for these servers as they would only accept
existing server labels as argument. This patch allows the use of
log-format notation to describe targets of a use-server rules, such
as in the example below:
listen test
bind *:1234
use-server %[hdr(srv)] if { hdr(srv) -m found }
use-server s1 if { path / }
server s1 127.0.0.1:18080
server s2 127.0.0.1:18081
If a use-server rule is applied because it was conditionned by an
ACL returning true, but the target of the use-server rule cannot be
resolved, no other use-server rule is evaluated and we fall back to
load balancing.
This feature was requested on the ML, and bumped with issue #563.
In srv_add_to_idle_list(), make sure we set the idle_time before we add
the connection to an idle list, not after, otherwise another thread may
grab it, set the idle_time to 0, only to have the original thread set it
back to now_ms.
This may have an impact, as in conn_free() we check idle_time to decide
if we should decrement the idle connection counters for the server.
In connect_server(), if we no longer have any idle connections for the
current thread, attempt to use the new "takeover" mux method to steal a
connection from another thread.
This should have no impact right now, given no mux implements it.
Make the "list" element a struct mt_list, and explicitely use
list_from_mt_list to get a struct list * where it is used as such, so that
mt_list_for_each_entry will be usable with it.
Add a new mux method, "takeover", that will attempt to make the current thread
responsible for the connection.
It should return 0 on success, and non-zero on failure.
Implement a new function, fd_takeover(), that lets you become the thread
responsible for the fd. On architectures that do not have a double-width CAS,
use a global rwlock.
fd_set_running() was also changed to be able to compete with fd_takeover(),
either using a dooble-width CAS on both running_mask and thread_mask, or
by claiming a reader on the global rwlock. This extra operation should not
have any measurable impact on modern architectures where threading is
relevant.
Revamp the server connection lists. We know have 3 lists :
- idle_conns, which contains idling connections
- safe_conns, which contains idling connections that are safe to use even
for the first request
- available_conns, which contains connections that are not idling, but can
still accept new streams (those are HTTP/2 or fastcgi, and are always
considered safe).
Make it so sessions are not responsible for connection anymore, except for
connections that are private, and thus can't be shared, otherwise, as soon
as a request is done, the session will just add the connection to the
orphan connections pool.
This will break http-reuse safe, but it is expected to be fixed later.
The flush_lock was introduced, mostly to be sure that pool_gc() will never
dereference a pointer that has been free'd. __pool_get_first() was acquiring
the lock to, the fear was that otherwise that pointer could get free'd later,
and then pool_gc() would attempt to dereference it. However, that can not
happen, because the only functions that can free a pointer, when using
lockless pools, are pool_gc() and pool_flush(), and as long as those two
are mutually exclusive, nobody will be able to free the pointer while
pool_gc() attempts to access it.
So change the flush_lock to a spinlock, and don't bother acquire/release
it in __pool_get_first(), that way callers of __pool_get_first() won't have
to wait while the pool is flushed. The worst that can happen is we call
__pool_refill_alloc() while the pool is getting flushed, and memory can
get allocated just to be free'd.
This may help with github issue #552
This may be backported to 2.1, 2.0 and 1.9.
Move the definition of WDTSIG and DEBUGSIG from wdt.c and debug.c into
types/signal.h, so that we can access them in another file.
We need those definition to avoid blocking those signals when running
__signal_process_queue().
This should be backported to 2.1, 2.0 and 1.9.
In the struct fdtab, introduce a new mask, running_mask. Each thread should
add its bit before using the fd.
Use the running_mask instead of a lock, in fd_insert/fd_delete, we'll just
spin as long as the mask is non-zero, to be sure we access the data
exclusively.
fd_set_running_excl() spins until the mask is 0, fd_set_running() just
adds the thread bit, and fd_clr_running() removes it.
The crtlist structure defines a crt-list in the HAProxy configuration.
It contains crtlist_entry structures which are the lines in a crt-list
file.
crt-list are now loaded in memory using crtlist and crtlist_entry
structures. The file is read only once. The generation algorithm changed
a little bit, new ckch instances are generated from the crtlist
structures, instead of being generated during the file loading.
The loading function was split in two, one that loads and caches the
crt-list and certificates, and one that looks for a crt-list and creates
the ckch instances.
Filters are also stored in crtlist_entry->filters as a char ** so we can
generate the sni_ctx again if needed. I won't be needed anymore to parse
the sni_ctx to do that.
A crtlist_entry stores the list of all ckch_inst that were generated
from this entry.
With these debug options we still get these warnings:
include/common/memory.h:501:23: warning: null pointer dereference [-Wnull-dereference]
*(volatile int *)0 = 0;
~~~~~~~~~~~~~~~~~~~^~~
include/common/memory.h:460:22: warning: null pointer dereference [-Wnull-dereference]
*(volatile int *)0 = 0;
~~~~~~~~~~~~~~~~~~~^~~
These are purposely there to crash the process at specific locations.
But the annoying warnings do not help with debugging and they are not
even reliable as the compiler may decide to optimize them away. Let's
pass the pointer through DISGUISE() to avoid this.
It's more generic and versatile than the previous shut_your_big_mouth_gcc()
that was used to silence annoying warnings as it's not limited to ignoring
syscalls returns only. This allows us to get rid of the aforementioned
function and the shut_your_big_mouth_gcc_int variable, that started to
look ugly in multi-threaded environments.
Tim reported that BUG_ON() issues warnings on his distro, as the libc marks
some syscalls with __attribute__((warn_unused_result)). Let's pass the
write() result through DISGUISE() to hide it.
This does exactly the same as ALREADY_CHECKED() but does it inline,
returning an identical copy of the scalar variable without letting
the compiler know how it might have been transformed. This can
forcefully disable certain null-pointer checks or result checks when
known undesirable. Typically forcing a crash with *(DISGUISE(NULL))=0
will not cause a null-deref warning.
These are mostly comments in the code. A few error messages were fixed
and are of low enough importance not to deserve a backport. Some regtests
were also fixed.
This patch adds the `unique-id` option to `proxy-v2-options`. If this
option is set a unique ID will be generated based on the `unique-id-format`
while sending the proxy protocol v2 header and stored as the unique id for
the first stream of the connection.
This feature is meant to be used in `tcp` mode. It works on HTTP mode, but
might result in inconsistent unique IDs for the first request on a keep-alive
connection, because the unique ID for the first stream is generated earlier
than the others.
Now that we can send unique IDs in `tcp` mode the `%ID` log variable is made
available in TCP mode.
Remove the list of private connections from server, it has been largely
unused, we only inserted connections in it, but we would never actually
use it.
Implement mt_list_to_list() and list_to_mt_list(), to be able to convert
from a struct list to a struct mt_list, and vice versa.
This is normally of no use, except for struct connection's list field, that
can go in either a struct list or a struct mt_list.
The stream-int code doesn't need to load server.h as it doesn't use
servers at all. However removing this one reveals that proxy.h was
lacking types/checks.h that used to be silently inherited from
types/server.h loaded before in stream_interface.h.
Ryan O'Hara reported that haproxy breaks on fedora-32 using gcc-10
(pre-release). It turns out that constructs such as:
while (item != head) {
item = LIST_ELEM(item.n);
}
loop forever, never matching <item> to <head> despite a printf there
showing them equal. In practice the problem is that the LIST_ELEM()
macro is wrong, it assigns the subtract of two pointers (an integer)
to another pointer through a cast to its pointer type. And GCC 10 now
considers that this cannot match a pointer and silently optimizes the
comparison away. A tested workaround for this is to build with
-fno-tree-pta. Note that older gcc versions even with -ftree-pta do
not exhibit this rather surprizing behavior.
This patch changes the test to instead cast the null-based address to
an int to get the offset and subtract it from the pointer, and this
time it works. There were just a few places to adjust. Ideally
offsetof() should be used but the LIST_ELEM() API doesn't make this
trivial as it's commonly called with a typeof(ptr) and not typeof(ptr*)
thus it would require to completely change the whole API, which is not
something workable in the short term, especially for a backport.
With this change, the emitted code is subtly different even on older
versions. A code size reduction of ~600 bytes and a total executable
size reduction of ~1kB are expected to be observed and should not be
taken as an anomaly. Typically this loop in dequeue_proxy_listeners() :
while ((listener = MT_LIST_POP(...)))
used to produce this code where the comparison is performed on RAX
while the new offset is assigned to RDI even though both are always
identical:
53ded8: 48 8d 78 c0 lea -0x40(%rax),%rdi
53dedc: 48 83 f8 40 cmp $0x40,%rax
53dee0: 74 39 je 53df1b <dequeue_proxy_listeners+0xab>
and now produces this one which is slightly more efficient as the
same register is used for both purposes:
53dd08: 48 83 ef 40 sub $0x40,%rdi
53dd0c: 74 2d je 53dd3b <dequeue_proxy_listeners+0x9b>
Similarly, retrieving the channel from a stream_interface using si_ic()
and si_oc() used to cause this (stream-int in rdi):
1cb7: c7 47 1c 00 02 00 00 movl $0x200,0x1c(%rdi)
1cbe: f6 47 04 10 testb $0x10,0x4(%rdi)
1cc2: 74 1c je 1ce0 <si_report_error+0x30>
1cc4: 48 81 ef 00 03 00 00 sub $0x300,%rdi
1ccb: 81 4f 10 00 08 00 00 orl $0x800,0x10(%rdi)
and now causes this:
1cb7: c7 47 1c 00 02 00 00 movl $0x200,0x1c(%rdi)
1cbe: f6 47 04 10 testb $0x10,0x4(%rdi)
1cc2: 74 1c je 1ce0 <si_report_error+0x30>
1cc4: 81 8f 10 fd ff ff 00 orl $0x800,-0x2f0(%rdi)
There is extremely little chance that this fix wakes up a dormant bug as
the emitted code effectively does what the source code intends.
This must be backported to all supported branches (dropping MT_LIST_ELEM
and the spoa_example parts as needed), since the bug is subtle and may
not always be visible even when compiling with gcc-10.
In MT_LIST_DEL_SAFE(), when the code was changed to use a temporary variable
instead of using the provided pointer directly, we shouldn't have changed
the code that set the pointer to NULL, as we really want the pointer
provided to be nullified, otherwise other parts of the code won't know
we just deleted an element, and bad things will happen.
This should be backported to 2.1.
The splice() syscall has been supported in glibc since version 2.5 issued
in 2006 and is present on supported systems so there's no need for having
our own arch-specific syscall definitions anymore.
This was made to support epoll on patched 2.4 kernels, and on early 2.6
using alternative libcs thanks to the arch-specific syscall definitions.
All the features we support have been around since 2.6.2 and present in
glibc since 2.3.2, neither of which are found in field anymore. Let's
simply drop this and use epoll normally.
The accept4() syscall has been present for a while now, there is no more
reason for maintaining our own arch-specific syscall implementation for
systems lacking it in libc but having it in the kernel.
This was introduced 10 years ago to squeeze a few CPU cycles per syscall
on 32-bit x86 machines and was already quite old by then, requiring to
explicitly enable support for this in the kernel. We don't even know if
it still builds, let alone if it works at all on recent kernels! Let's
completely drop this now.
Since commit 244b070 ("MINOR: ssl/cli: support crt-list filters"),
HAProxy generates a list of filters based on the sni_ctx in memory.
However it's not always relevant, sometimes no filters were configured
and the CN/SAN in the new certificate are not the same.
This patch fixes the issue by using a flag filters in the ckch_inst, so
we are able to know if there were filters or not. In the late case it
uses the CN/SAN of the new certificate to generate the sni_ctx.
note: filters are still only used in the crt-list atm.
We currently have two UUID generation functions, one for the sample
fetch and the other one in the SPOE filter. Both were a bit complicated
since they were made to support random() implementations returning an
arbitrary number of bits, and were throwing away 33 bits every 64. Now
we don't need this anymore, so let's have a generic function consuming
64 bits at once and use it as appropriate.
This is the replacement of failed attempt to add thread safety and
per-process sequences of random numbers initally tried with commit
1c306aa84d ("BUG/MEDIUM: random: implement per-thread and per-process
random sequences").
This new version takes a completely different approach and doesn't try
to work around the horrible OS-specific and non-portable random API
anymore. Instead it implements "xoroshiro128**", a reputedly high
quality random number generator, which is one of the many variants of
xorshift, which passes all quality tests and which is described here:
http://prng.di.unimi.it/
While not cryptographically secure, it is fast and features a 2^128-1
period. It supports fast jumps allowing to cut the period into smaller
non-overlapping sequences, which we use here to support up to 2^32
processes each having their own, non-overlapping sequence of 2^96
numbers (~7*10^28). This is enough to provide 1 billion randoms per
second and per process for 2200 billion years.
The implementation was made thread-safe either by using a double 64-bit
CAS on platforms supporting it (x86_64, aarch64) or by using a local
lock for the time needed to perform the shift operations. This ensures
that all threads pick numbers from the same pool so that it is not
needed to assign per-thread ranges. For processes we use the fast jump
method to advance the sequence by 2^96 for each process.
Before this patch, the following config:
global
nbproc 8
frontend f
bind :4445
mode http
log stdout format raw daemon
log-format "%[uuid] %pid"
redirect location /
Would produce this output:
a4d0ad64-2645-4b74-b894-48acce0669af 12987
a4d0ad64-2645-4b74-b894-48acce0669af 12992
a4d0ad64-2645-4b74-b894-48acce0669af 12986
a4d0ad64-2645-4b74-b894-48acce0669af 12988
a4d0ad64-2645-4b74-b894-48acce0669af 12991
a4d0ad64-2645-4b74-b894-48acce0669af 12989
a4d0ad64-2645-4b74-b894-48acce0669af 12990
82d5f6cd-f6c1-4f85-a89c-36ae85d26fb9 12987
82d5f6cd-f6c1-4f85-a89c-36ae85d26fb9 12992
82d5f6cd-f6c1-4f85-a89c-36ae85d26fb9 12986
(...)
And now produces:
f94b29b3-da74-4e03-a0c5-a532c635bad9 13011
47470c02-4862-4c33-80e7-a952899570e5 13014
86332123-539a-47bf-853f-8c8ea8b2a2b5 13013
8f9efa99-3143-47b2-83cf-d618c8dea711 13012
3cc0f5c7-d790-496b-8d39-bec77647af5b 13015
3ec64915-8f95-4374-9e66-e777dc8791e0 13009
0f9bf894-dcde-408c-b094-6e0bb3255452 13011
49c7bfde-3ffb-40e9-9a8d-8084d650ed8f 13014
e23f6f2e-35c5-4433-a294-b790ab902653 13012
There are multiple benefits to using this method. First, it doesn't
depend anymore on a non-portable API. Second it's thread safe. Third it
is fast and more proven than any hack we could attempt to try to work
around the deficiencies of the various implementations around.
This commit depends on previous patches "MINOR: tools: add 64-bit rotate
operators" and "BUG/MEDIUM: random: initialize the random pool a bit
better", all of which will need to be backported at least as far as
version 2.0. It doesn't require to backport the build fixes for circular
include files dependecy anymore.
This reverts commit 1c306aa84d.
It breaks the build on all non-glibc platforms. I got confused by the
man page (which possibly is the most confusing man page I've ever read
about a standard libc function) and mistakenly understood that random_r
was portable, especially since it appears in latest freebsd source as
well but not in released versions, and with a slightly different API :-/
We need to find a different solution with a fallback. Among the
possibilities, we may reintroduce this one with a fallback relying on
locking around the standard functions, keeping fingers crossed for no
other library function to call them in parallel, or we may also provide
our own PRNG, which is not necessarily more difficult than working
around the totally broken up design of the portable API.
As mentioned in previous patch, the random number generator was never
made thread-safe, which used not to be a problem for health checks
spreading, until the uuid sample fetch function appeared. Currently
it is possible for two threads or processes to produce exactly the
same UUID. In fact it's extremely likely that this will happen for
processes, as can be seen with this config:
global
nbproc 8
frontend f
bind :4445
mode http
log stdout daemon format raw
log-format "%[uuid] %pid"
redirect location /
It typically produces this log:
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30645
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30641
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30644
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30639
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30646
07764439-c24d-4e6f-a5a6-0138be59e7a8 30645
07764439-c24d-4e6f-a5a6-0138be59e7a8 30639
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30643
07764439-c24d-4e6f-a5a6-0138be59e7a8 30646
b6773fdd-678f-4d04-96f2-4fb11ad15d6b 30646
551ce567-0bfb-4bbd-9b58-cdc7e9365325 30642
07764439-c24d-4e6f-a5a6-0138be59e7a8 30642
What this patch does is to use a distinct per-thread and per-process
seed to make sure the same sequences will not appear, and will then
extend these seeds by "burning" a number of randoms that depends on
the global random seed, the thread ID and the process ID. This adds
roughly 20 extra bits of randomness, resulting in 52 bits total per
thread and per process.
It only takes a few milliseconds to burn these randoms and given
that threads start with a different seed, we know they will not
catch each other. So these random extra bits are essentially added
to ensure randomness between boots and cluster instances.
This replaces all uses of random() with ha_random() which uses the
thread-local state.
This must be backported as far as 2.0 or any version having the
UUID sample-fetch function since it's the main victim here.
It's important to note that this patch, in addition to depending on
the previous one "BUG/MEDIUM: init: initialize the random pool a bit
better", also depends on the preceeding build fixes to address a
circular dependency issue in the include files that prevented it
from building. Part or all of these patches may need to be backported
or adapted as well.
Since the UUID sample fetch was created, some people noticed that in
certain virtualized environments they manage to get exact same UUIDs
on different instances started exactly at the same moment. It turns
out that the randoms were only initialized to spread the health checks
originally, not to provide "clean" randoms.
This patch changes this and collects more randomness from various
sources, including existing randoms, /dev/urandom when available,
RAND_bytes() when OpenSSL is available, as well as the timing for such
operations, then applies a SHA1 on all this to keep a 160 bits random
seed available, 32 of which are passed to srandom().
It's worth mentioning that there's no clean way to pass more than 32
bits to srandom() as even initstate() provides an opaque state that
must absolutely not be tampered with since known implementations
contain state information.
At least this allows to have up to 4 billion different sequences
from the boot, which is not that bad.
Note that the thread safety was still not addressed, which is another
issue for another patch.
This must be backported to all versions containing the UUID sample
fetch function, i.e. as far as 2.0.
buffer.h relies on proto/activity because it contains some code and not
just type definitions. It must not be included from types files. It
should probably also be split in two if it starts to include a proto.
This causes some circular dependencies at other places.
This flag was used in some internal functions to be sure the current stream is
able to handle HTTP content. It was introduced when the legacy HTTP code was
still there. Now, It is possible to rely on stream's flags to be sure we have an
HTX stream.
So the flag HLUA_TXN_HTTP_RDY can be removed. Everywhere it was tested, it is
replaced by a call to the IS_HTX_STRM() macro.
This patch is mandatory to allow the support of the filters written in lua.
The htx_find_offset() function may be used to look for a block at a specific
offset in an HTX message, starting from the message head. A compound result is
returned, an htx_ret structure, with the found block and the position of the
offset in the block. If the offset is ouside of the HTX message, the returned
block is NULL.
The b_insert_blk() function may now be used to insert a string, given a pointer
and the string length, at an absolute offset in a buffer, moving data between
this offset and the buffer's tail just after the end of the inserted string. The
buffer's length is automatically updated. This function supports wrapping. All
the string is copied or nothing. So it returns 0 if there are not enough space
to perform the copy. Otherwise, the number of bytes copied is returned.
pattern_finalize_config() uses an inefficient algorithm which is a
problem with very large configuration files. This affects startup, and
therefore reload time. When haproxy is deployed as a router in a
Kubernetes cluster the generated configuration file may be large and
reloads are frequently occuring, which makes this a significant issue.
The old algorithm is O(n^2)
* allocate missing uids - O(n^2)
* sort linked list - O(n^2)
The new algorithm is O(n log n):
* find the user allocated uids - O(n)
* store them for efficient lookup - O(n log n)
* allocate missing uids - n times O(log n)
* sort all uids - O(n log n)
* convert back to linked list - O(n)
Performance examples, startup time in seconds:
pat_refs old new
1000 0.02 0.01
10000 2.1 0.04
20000 12.3 0.07
30000 27.9 0.10
40000 52.5 0.14
50000 77.5 0.17
Please backport to 1.8, 2.0 and 2.1.
`istalloc` allocates memory and returns an `ist` with the size `0` that points
to this allocation.
`istfree` frees the pointed memory and clears the pointer.
It was only possible to go down from the ckch_store to the sni_ctx but
not to go up from the sni_ctx to the ckch_store.
To allow that, 2 pointers were added:
- a ckch_inst pointer in the struct sni_ctx
- a ckckh_store pointer in the struct ckch_inst
Generate a list of the previous filters when updating a certificate
which use filters in crt-list. Then pass this list to the function
generating the sni_ctx during the commit.
This feature allows the update of the crt-list certificates which uses
the filters with "set ssl cert".
This function could be probably replaced by creating a new
ckch_inst_new_load_store() function which take the previous sni_ctx list as
an argument instead of the char **sni_filter, avoiding the
allocation/copy during runtime for each filter. But since are still
handling the multi-cert bundles, it's better this way to avoid code
duplication.
A test on FreeBSD with clang 4 to 8 produces this on a call to a
spinning loop on the CLI:
call trace(5):
| 0x53e2bc [eb 16 48 63 c3 48 c1 e0]: wdt_handler+0x10c
| 0x800e02cfe [e8 5d 83 00 00 8b 18 8b]: libthr:pthread_sigmask+0x53e
with our own function it correctly produces this:
call trace(20):
| 0x53e2dc [eb 16 48 63 c3 48 c1 e0]: wdt_handler+0x10c
| 0x800e02cfe [e8 5d 83 00 00 8b 18 8b]: libthr:pthread_sigmask+0x53e
| 0x800e022bf [48 83 c4 38 5b 41 5c 41]: libthr:pthread_getspecific+0xdef
| 0x7ffffffff003 [48 8d 7c 24 10 6a 00 48]: main+0x7fffffb416f3
| 0x801373809 [85 c0 0f 84 6f ff ff ff]: libc:__sys_gettimeofday+0x199
| 0x801373709 [89 c3 85 c0 75 a6 48 8b]: libc:__sys_gettimeofday+0x99
| 0x801371c62 [83 f8 4e 75 0f 48 89 df]: libc:gettimeofday+0x12
| 0x51fa0a [48 89 df 4c 89 f6 e8 6b]: ha_thread_dump_all_to_trash+0x49a
| 0x4b723b [85 c0 75 09 49 8b 04 24]: mworker_cli_sockpair_new+0xd9b
| 0x4b6c68 [85 c0 75 08 4c 89 ef e8]: mworker_cli_sockpair_new+0x7c8
| 0x532f81 [4c 89 e7 48 83 ef 80 41]: task_run_applet+0xe1
So let's add clang+x86_64 to the list of platforms that will use our
simplified version. As a bonus it will not require to link with
-lexecinfo on FreeBSD and will work out of the box when passing
USE_BACKTRACE=1.
It happens that on aarch64 backtrace() only returns one entry (tested
with gcc 4.7.4, 5.5.0 and 7.4.1). Probably that it refrains from unwinding
the stack due to the risk of hitting a bad pointer. Here we can use
may_access() to know when it's safe, so we can actually unwind the stack
without taking risks. It happens that the faulting function (the one
just after the signal handler) is not listed here, very likely because
the signal handler uses a special stack and did not create a new frame.
So this patch creates a new my_backtrace() function in standard.h that
either calls backtrace() or does its own unrolling. The choice depends
on HA_HAVE_WORKING_BACKTRACE which is set in compat.h based on the build
target.
It's only available for bind line. "ca-verify-file" allows to separate
CA certificates from "ca-file". CA names sent in server hello message is
only compute from "ca-file". Typically, "ca-file" must be defined with
intermediate certificates and "ca-verify-file" with certificates to
ending the chain, like root CA.
Fix issue #404.
We use various hacks at a few places to try to identify known function
pointers in debugging outputs (show threads & show fd). Let's centralize
this into a new function dedicated to this. It already knows about the
functions matched by "show threads" and "show fd", and when built with
USE_DL, it can rely on dladdr1() to resolve other functions. There are
some limitations, as static functions are not resolved, linking with
-rdynamic is mandatory, and even then some functions will not necessarily
appear. It's possible to do a better job by rebuilding the whole symbol
table from the ELF headers in memory but it's less portable and the gains
are still limited, so this solution remains a reasonable tradeoff.
This function dumps <n> bytes from <addr> in hex form into buffer <buf>
enclosed in brackets after the address itself, formatted on 14 chars
including the "0x" prefix. This is meant to be used as a prefix for code
areas. For example: "0x7f10b6557690 [48 c7 c0 0f 00 00 00 0f]: "
It relies on may_access() to know if the bytes are dumpable, otherwise "--"
is emitted. An optional prefix is supported.
We always set them both, which makes sense since errors at the FD level
indicate a terminal condition for the socket that cannot be recovered.
Usually this is detected via a write error, but sometimes such an error
may asynchronously be reported on the read side. Let's simplify this
using only the write bit and calling it RW since it's used like this
everywhere, and leave the R bit spare for future use.
There's no point in trying to be too generic for these flags as the
read and write sides will soon differ a bit. Better explicitly define
the flags for each direction without trying to be direction-agnostic.
this clarifies the code and removes some defines.
This was used only by fd_recv_state() and fd_send_state(), both of
which are unused. This will not work anymore once recv and send flags
start to differ, so let's remove this.
commit c87e46881 ("MINOR: http-rules: Add a flag on redirect rules to know the
rule direction") introduced a new flag for redirect rules, but its value has
bits in common with REDIRECT_FLAG_DROP_QS, which makes us enter this code path
in http_apply_redirect_rule(), which will then drop the query string.
To fix this, just give REDIRECT_FLAG_FROM_REQ its own unique value.
This must be backported where c87e468816 is backported.
This should fix issue 521.
This lock was only needed to protect the buffer_wq list, but now we have
the mt_list for this. This patch simply turns the buffer_wq list to an
mt_list and gets rid of the lock.
It's worth noting that the whole buffer_wait thing still looks totally
wrong especially in a threaded context: the wakeup_cb() callback is
called synchronously from any thread and may end up calling some
connection code that was not expected to run on a given thread. The
whole thing should probably be reworked to use tasklets instead and be
a bit more centralized.
Move the cert_issuer_tree outside the global_ssl structure since it's
not a configuration variable. And move the declaration of the
issuer_chain structure in types/ssl_sock.h
This commit adds ALWAYS_ALIGN(), MAYBE_ALIGN() and ATOMIC_ALIGN() to
be placed as delimitors inside structures to force alignment to a
given size. These depend on the architecture's capabilities so that
it is possible to always align, align only on archs not supporting
unaligned accesses at all, or only on those not supporting them for
atomic accesses (e.g. before a lock).
The isalnum(), isalpha(), isdigit() etc functions from ctype.h are
supposed to take an int in argument which must either reflect an
unsigned char or EOF. In practice on some platforms they're implemented
as macros referencing an array, and when passed a char, they either cause
a warning "array subscript has type 'char'" when lucky, or cause random
segfaults when unlucky. It's quite unconvenient by the way since none of
them may return true for negative values. The recent introduction of
cygwin to the list of regularly tested build platforms revealed a lot
of breakage there due to the same issues again.
So this patch addresses the problem all over the code at once. It adds
unsigned char casts to every valid use case, and also drops the unneeded
double cast to int that was sometimes added on top of it.
It may be backported by dropping irrelevant changes if that helps better
support uncommon platforms. It's unlikely to fix bugs on platforms which
would already not emit any warning though.
This used to be a minor optimization on ix86 where registers are scarce
and the calling convention not very efficient, but this platform is not
relevant enough anymore to warrant all this dirt in the code for the sake
of saving 1 or 2% of performance. Modern platforms don't use this at all
since their calling convention already defaults to using several registers
so better get rid of this once for all.
This patch turns the double negation of 'not unlikely' into 'likely'
and then turns the negation of 'not smaller' into 'greater or equal'
in an attempt to improve readability of the condition.
[wt: this was not a bug but purposely written like this to improve code
generation on older compilers but not needed anymore as described here:
https://www.mail-archive.com/haproxy@formilux.org/msg36392.html ]
Move the `!` inside the likely and negate it to unlikely.
The previous version should not have caused issues, because it is converted
to a boolean / integral value before being passed to __builtin_expect(), but
it's certainly unusual.
[wt: this was not a bug but purposely written like this to improve code
generation on older compilers but not needed anymore as described here:
https://www.mail-archive.com/haproxy@formilux.org/msg36392.html ]
We used to special-case the likely()/unlikely() macros for a series of
early gcc 4.x compilers which used to produce very bad code when using
__builtin_expect(x,1), which basically used to build an integer (0 or 1)
from a condition then compare it to integer 1. This was already fixed in
5.x, but even now, looking at the code produced by various flavors of 4.x
this bad behavior couldn't be witnessed anymore. So let's consider it as
fixed by now, which will allow to get rid of some ugly tricks at some
specific places. A test on 4.7.4 shows that the code shrinks by about 3kB
now, thanks to some tests being inlined closer to the call place and the
unlikely case being moved to real functions. See the link below for more
background on this.
Link: https://www.mail-archive.com/haproxy@formilux.org/msg36392.html
These ones are irrelevant to the config but rather to the platform, and
as such are better placed in compiler.h.
Here we take the opportunity for declaring a few extra capabilities:
- HA_UNALIGNED : CPU supports unaligned accesses
- HA_UNALIGNED_LE : CPU supports unaligned accesses in little endian
- HA_UNALIGNED_FAST : CPU supports fast unaligned accesses
- HA_UNALIGNED_ATOMIC : CPU supports unaligned accesses in atomics
This will help remove a number of #ifdefs with arch-specific statements.
Add a function that finds a character in an ist and returns an
updated ist with the length of the portion of the original string
that doesn't contain the char.
Might be backported to 2.1
This flag is currently supported by raw_sock to perform a single recv()
attempt and avoid subscribing. Typically on the request and response
paths with keep-alive, with short messages we know that it's very likely
that the first message is enough.
This marks the end of the transition from the connection polling states
introduced in 1.5-dev12 and the subscriptions in that arrived in 1.9.
The socket layer can now safely use its FD while all upper layers rely
exclusively on subscriptions. These old functions were removed. Some may
deserve some renaming to improved clarty though. The single call to
conn_xprt_stop_both() was dropped in favor of conn_cond_update_polling()
which already does the same.
The last few calls to conn_xprt_{want,stop}_{recv,send} in the central
connection code were replaced with their strictly exact equivalent fd_*,
adding the call to conn_ctrl_ready() when it was missing.
Historically we used to require that the connections held the desired
polling states for the data layer and the socket layer. Then with muxes
these were more or less merged into the transport layer, and now it
happens that with all transport layers having their own state, the
"transport layer state" as we have it in the connection (XPRT_RD_ENA,
XPRT_WR_ENA) is only an exact copy of the undelying file descriptor
state, but with a delay. All of this is causing some difficulties at
many places in the code because there are still some locations which
use the conn_want_* API to remain clean and only rely on connection,
and count on a later collection call to conn_cond_update_polling(),
while others need an immediate action and directly use the FD updates.
Since our updates are now much cheaper, most of them being only an
atomic test-and-set operation, and since our I/O callbacks are deferred,
there's no benefit anymore in trying to "cache" the transient state
change in the connection flags hoping to cancel them before they
become an FD event. Better make such calls transparent indirections
to the FD layer instead and get rid of the deferred operations which
needlessly complicate the logic inside.
This removes flags CO_FL_XPRT_{RD,WR}_ENA and CO_FL_WILL_UPDATE.
A number of functions related to polling updates were either greatly
simplified or removed.
Two places were using CO_FL_XPRT_WR_ENA as a hint to know if more data
were expected to be sent after a PROXY protocol or SOCKSv4 header. These
ones were simply replaced with a check on the subscription which is
where we ought to get the autoritative information from.
Now the __conn_xprt_want_* and their conn_xprt_want_* counterparts
are the same. conn_stop_polling() and conn_xprt_stop_both() are the
same as well. conn_cond_update_polling() only causes errors to stop
polling. It also becomes way more obvious that muxes should not at
all employ conn_xprt_{want|stop}_{recv,send}(), and that the call
to __conn_xprt_stop_recv() in case a mux failed to allocate a buffer
is inappropriate, it ought to unsubscribe from reads instead. All of
this definitely requires a serious cleanup.
http_get_hdrs_size() function may now be used to get the bytes held by headers
in an HTX message. It only works if the headers were not already
forwarded. Metadata are not counted here.
The function is_idchar() was added by commit 36f586b ("MINOR: tools:
add is_idchar() to tell if a char may belong to an identifier") to
ease matching of sample fetch/converter names. But it lacked support
for the '+' character used in "base32+src" and "url32+src". A quick
way to figure the list of supported sample fetch+converter names is
to issue the following command:
git grep '"[^"]*",.*SMP_T_.*SMP_USE_'|cut -f2 -d'"'|sort -u
No more entry is reported once searching for characters not covered
by is_idchar().
No backport is needed.
When an end pointer is passed, instead of complaining that a comma is
missing after a keyword, sample_parse_expr() will silently return the
pointer to the current location into this return pointer so that the
caller can continue its parsing. This will be used by more complex
expressions which embed sample expressions, and may even permit to
embed sample expressions into arguments of other expressions.
The main problem we're having with argument parsing is that at the
moment the caller looks for the first character looking like an end
of arguments (')') and calls make_arg_list() on the sub-string inside
the parenthesis.
Let's first change the way it works so that make_arg_list() also
consumes the parenthesis and returns the pointer to the first char not
consumed. This will later permit to refine each argument parsing.
For now there is no functional change.
This does like chunk_strcpy() except that the maximum string length may
be limited by the caller. A trailing zero is always appended. This is
particularly handy to extract portions of strings to put into the trash
for use with libc functions requiring a nul-terminated string.
While looking for other occurrences of do { continue; } while (0) I
found these few leftovers in mini-clist where an outer loop was made
around "do { } while (0)" then another loop was placed inside just to
handle the continue. Let's clean this up by just removing the outer
one. Most of the patch is only the inner part of the loop that is
reindented. It was verified that the resulting code is the same.
When an action successfully finishes, the action return code (ACT_RET_*) is now
retrieve on the stack, ff the first element is an integer. In addition, in
hlua_txn_done(), the value ACT_RET_DONE is pushed on the stack before
exiting. Thus, when a script uses this function, the corresponding action still
finishes with the good code. Thanks to this change, the flag HLUA_STOP is now
useless. So it has been removed.
It is a mandatory step to allow a lua action to return any action return code.
It is not possible anymore to alter the HTTP parser state from lua sample
fetches or lua actions. So there is no reason to still check for the parser
state consistency.
It is now possible to append extra headers to the generated responses by HTTP
return actions, while it is not based on an errorfile. For return actions based
on errorfiles, these extra headers are ignored. To define an extra header, a
"hdr" argument must be used with a name and a value. The value is a log-format
string. For instance:
http-request status 200 hdr "x-src" "%[src]" hdr "x-dst" "%[dst]"
Thanks to this new action, it is now possible to return any responses from
HAProxy, with any status code, based on an errorfile, a file or a string. Unlike
the other internal messages generated by HAProxy, these ones are not interpreted
as errors. And it is not necessary to use a file containing a full HTTP
response, although it is still possible. In addition, using a log-format string
or a log-format file, it is possible to have responses with a dynamic
content. This action can be used on the request path or the response path. The
only constraint is to have a responses smaller than a buffer. And to avoid any
warning the buffer space reserved to the headers rewritting should also be free.
When a response is returned with a file or a string as payload, it only contains
the content-length header and the content-type header, if applicable. Here are
examples:
http-request return content-type image/x-icon file /var/www/favicon.ico \
if { path /favicon.ico }
http-request return status 403 content-type text/plain \
lf-string "Access denied. IP %[src] is blacklisted." \
if { src -f /etc/haproxy/blacklist.lst }
This patch introduces the 'http-after-response' rules. These rules are evaluated
at the end of the response analysis, just before the data forwarding, on ALL
HTTP responses, the server ones but also all responses generated by
HAProxy. Thanks to this ruleset, it is now possible for instance to add some
headers to the responses generated by the stats applet. Following actions are
supported :
* allow
* add-header
* del-header
* replace-header
* replace-value
* set-header
* set-status
* set-var
* strict-mode
* unset-var
Operations performed when internal responses (redirect/deny/auth/errors) are
returned are always the same. The http_forward_proxy_resp() function is added to
group all of them under a unique function.
The http_server_error() function now relies on http_reply_and_close(). Both do
almost the same actions. In addtion, http_server_error() sets the error flag and
the final state flag on the stream.
HTTP redirect rules can be evaluated on the request or the response path. So
when a redirect rule is evaluated, it is important to have this information
because some specific processing may be performed depending on the direction. So
the REDIRECT_FLAG_FROM_REQ flag has been added. It is set when applicable on the
redirect rule during the parsing.
This patch is mandatory to fix a bug on redirect rule. It must be backported to
all stable versions.
<.arg.dns.dns_opts> field in the act_rule structure is now dynamically allocated
when a do-resolve rule is parsed. This drastically reduces the structure size.
The channel_htx_copy_msg() function can now be used to copy an HTX message in a
channel's buffer. This function takes care to not overwrite existing data.
This patch depends on the commit "MINOR: htx: Add a function to append an HTX
message to another one". Both are mandatory to fix a bug in
http_reply_and_close() function. Be careful to backport both first.
the htx_append_msg() function can now be used to append an HTX message to
another one. All the message is copied or nothing. If an error occurs during the
copy, all changes are rolled back.
This patch is mandatory to fix a bug in http_reply_and_close() function. Be
careful to backport it first.
In __pool_get_first(), don't forget to unlock the pool lock if the pool is
empty, otherwise no writer will be able to take the lock, and as it is done
when reloading, it leads to an infinite loop on reload.
This should be backported with commit 04f5fe87d3
When using lockless pools, add a new rwlock, flush_pool. read-lock it when
getting memory from the pool, so that concurrenct access are still
authorized, but write-lock it when we're about to free memory, in
pool_flush() and pool_gc().
The problem is, when removing an item from the pool, we unreference it
to get the next one, however, that pointer may have been free'd in the
meanwhile, and that could provoke a crash if the pointer has been unmapped.
It should be OK to use a rwlock, as normal operations will still be able
to access the pool concurrently, and calls to pool_flush() and pool_gc()
should be pretty rare.
This should be backported to 2.1, 2.0 and 1.9.
Commit a17664d829 ("MEDIUM: tasks: automatically requeue into the bulk
queue an already running tasklet") tried to inflict a penalty to
self-requeuing tasks/tasklets which correspond to those involved in
large, high-latency data transfers, for the benefit of all other
processing which requires a low latency. However, it turns out that
while it ought to do this on a case-by-case basis, basing itself on
the RUNNING flag isn't accurate because this flag doesn't leave for
tasklets, so we'd rather need a distinct flag to tag such tasklets.
This commit introduces TASK_SELF_WAKING to mark tasklets acting like
this. For now it's still set when TASK_RUNNING is present but this
will have to change. The flag is kept across wakeups.
When a tasklet re-runs itself such as in this chain:
si_cs_io_cb -> si_cs_process -> si_notify -> si_chk_rcv
then we know it can easily clobber the run queue and harm latency. Now
what the scheduler does when it detects this is that such a tasklet is
automatically placed into the bulk list so that it's processed with the
remaining CPU bandwidth only. Thanks to this the CLI becomes instantly
responsive again even under heavy stress at 50 Gbps over 40kcon and
100% CPU on 16 threads.
We used to mix high latency tasks and low latency tasklets in the same
list, and to even refill bulk tasklets there, causing some unfairness
in certain situations (e.g. poll-less transfers between many connections
saturating the machine with similarly-sized in and out network interfaces).
This patch changes the mechanism to split the load into 3 lists depending
on the task/tasklet's desired classes :
- URGENT: this is mainly for tasklets used as deferred callbacks
- NORMAL: this is for regular tasks
- BULK: this is for bulk tasks/tasklets
Arbitrary ratios of max_processed are picked from each of these lists in
turn, with the ability to complete in one list from what was not picked
in the previous one. After some quick tests, the following setup gave
apparently good results both for raw TCP with splicing and for H2-to-H1
request rate:
- 0 to 75% for urgent
- 12 to 50% for normal
- 12 to what remains for bulk
Bulk is not used yet.
As mentioned in commit c192b0ab95 ("MEDIUM: connection: remove
CO_FL_CONNECTED and only rely on CO_FL_WAIT_*"), there is a lack of
consistency on which flags are checked among L4/L6/HANDSHAKE depending
on the code areas. A number of sample fetch functions only check for
L4L6 to report MAY_CHANGE, some places only check for HANDSHAKE and
many check both L4L6 and HANDSHAKE.
This patch starts to make all of this more consistent by introducing a
new mask CO_FL_WAIT_XPRT which is the union of L4/L6/HANDSHAKE and
reports whether the transport layer is ready or not.
All inconsistent call places were updated to rely on this one each time
the goal was to check for the readiness of the transport layer.
Most places continue to check CO_FL_HANDSHAKE while in fact they should
check CO_FL_HANDSHAKE_NOSSL, which contains all handshakes but the one
dedicated to SSL renegotiation. In fact the SSL layer should be the
only one checking CO_FL_SSL_WAIT_HS, so as to avoid processing data
when a renegotiation is in progress, but other ones randomly include it
without knowing. And ideally it should even be an internal flag that's
not exposed in the connection.
This patch takes CO_FL_SSL_WAIT_HS out of CO_FL_HANDSHAKE, uses this flag
consistently all over the code, and gets rid of CO_FL_HANDSHAKE_NOSSL.
In order to limit the confusion that has accumulated over time, the
CO_FL_SSL_WAIT_HS flag which indicates an ongoing SSL handshake,
possibly used by a renegotiation was moved after the other ones.
Commit 477902bd2e ("MEDIUM: connections: Get ride of the xprt_done
callback.") broke the master CLI for a very obscure reason. It happens
that short requests immediately terminated by a shutdown are properly
received, CS_FL_EOS is correctly set, but in si_cs_recv(), we refrain
from setting CF_SHUTR on the channel because CO_FL_CONNECTED was not
yet set on the connection since we've not passed again through
conn_fd_handler() and it was not done in conn_complete_session(). While
commit a8a415d31a ("BUG/MEDIUM: connections: Set CO_FL_CONNECTED in
conn_complete_session()") fixed the issue, such accident may happen
again as the root cause is deeper and actually comes down to the fact
that CO_FL_CONNECTED is lazily set at various check points in the code
but not every time we drop one wait bit. It is not the first time we
face this situation.
Originally this flag was used to detect the transition between WAIT_*
and CONNECTED in order to call ->wake() from the FD handler. But since
at least 1.8-dev1 with commit 7bf3fa3c23 ("BUG/MAJOR: connection: update
CO_FL_CONNECTED before calling the data layer"), CO_FL_CONNECTED is
always synchronized against the two others before being checked. Moreover,
with the I/Os moved to tasklets, the decision to call the ->wake() function
is performed after the I/Os in si_cs_process() and equivalent, which don't
care about this transition either.
So in essence, checking for CO_FL_CONNECTED has become a lazy wait to
check for (CO_FL_WAIT_L4_CONN | CO_FL_WAIT_L6_CONN), but that always
relies on someone else having synchronized it.
This patch addresses it once for all by killing this flag and only checking
the two others (for which a composite mask CO_FL_WAIT_L4L6 was added). This
revealed a number of inconsistencies that were purposely not addressed here
for the sake of bisectability:
- while most places do check both L4+L6 and HANDSHAKE at the same time,
some places like assign_server() or back_handle_st_con() and a few
sample fetches looking for proxy protocol do check for L4+L6 but
don't care about HANDSHAKE ; these ones will probably fail on TCP
request session rules if the handshake is not complete.
- some handshake handlers do validate that a connection is established
at L4 but didn't clear CO_FL_WAIT_L4_CONN
- the ->ctl method of mux_fcgi, mux_pt and mux_h1 only checks for L4+L6
before declaring the mux ready while the snd_buf function also checks
for the handshake's completion. Likely the former should validate the
handshake as well and we should get rid of these extra tests in snd_buf.
- raw_sock_from_buf() would directly set CO_FL_CONNECTED and would only
later clear CO_FL_WAIT_L4_CONN.
- xprt_handshake would set CO_FL_CONNECTED itself without actually
clearing CO_FL_WAIT_L4_CONN, which could apparently happen only if
waiting for a pure Rx handshake.
- most places in ssl_sock that were checking CO_FL_CONNECTED don't need
to include the L4 check as an L6 check is enough to decide whether to
wait for more info or not.
It also becomes obvious when reading the test in si_cs_recv() that caused
the failure mentioned above that once converted it doesn't make any sense
anymore: having CS_FL_EOS set while still waiting for L4 and L6 to complete
cannot happen since for CS_FL_EOS to be set, the other ones must have been
validated.
Some of these parts will still deserve further cleanup, and some of the
observations above may induce some backports of potential bug fixes once
totally analyzed in their context. The risk of breaking existing stuff
is too high to blindly backport everything.
The xprt_done_cb callback was used to defer some connection initialization
until we're connected and the handshake are done. As it mostly consists of
creating the mux, instead of using the callback, introduce a conn_create_mux()
function, that will just call conn_complete_session() for frontend, and
create the mux for backend.
In h2_wake(), make sure we call the wake method of the stream_interface,
as we no longer wakeup the stream task.
In connect_server(), when creating a new connection for which we don't yet
know the mux (because it'll be decided by the ALPN), instead of associating
the connection to the stream_interface, always create a conn_stream. This way,
we have less special-casing needed. Store the conn_stream in conn->ctx,
so that we can reach the upper layers if needed.
"set ssl cert <filename> <payload>" CLI command should have the same
result as reload HAproxy with the updated pem file (<filename>).
Is not the case, DHparams/cert-chain is kept from the previous
context if no DHparams/cert-chain is set in the context (<payload>).
This patch should be backport to 2.1
For complex stick tables with many entries/columns, it can be beneficial
to filter using multiple criteria. The maximum number of filter entries
can be controlled by defining STKTABLE_FILTER_LEN during build time.
This patch can be backported to older releases.
while working on issue #429, I encountered build failures with various
non-released openssl versions, let us improve ssl defines, switch to
features, not versions, for EVP_CTRL_AEAD_SET_IVLEN and
EVP_CTRL_AEAD_SET_TAG.
No backport is needed as there is no valid reason to build a stable haproxy
version against a development version of openssl.
It was inherited from acl_time, introduced in 1.3.10 by commit a84d374367
("[MAJOR] new framework for generic ACL support") and was never ever used.
Let's simply drop it now.
Signed bitfields of size `1` hold the values `0` and `-1`, but are
usually assigned `1`, possibly leading to subtle bugs when the value
is explicitely compared against `1`.
Most DNS servers provide A/AAAA records in the Additional section of a
response, which correspond to the SRV records from the Answer section:
;; QUESTION SECTION:
;_http._tcp.be1.domain.tld. IN SRV
;; ANSWER SECTION:
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A1.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A8.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A5.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A6.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A4.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A3.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A2.domain.tld.
_http._tcp.be1.domain.tld. 3600 IN SRV 5 500 80 A7.domain.tld.
;; ADDITIONAL SECTION:
A1.domain.tld. 3600 IN A 192.168.0.1
A8.domain.tld. 3600 IN A 192.168.0.8
A5.domain.tld. 3600 IN A 192.168.0.5
A6.domain.tld. 3600 IN A 192.168.0.6
A4.domain.tld. 3600 IN A 192.168.0.4
A3.domain.tld. 3600 IN A 192.168.0.3
A2.domain.tld. 3600 IN A 192.168.0.2
A7.domain.tld. 3600 IN A 192.168.0.7
SRV record support was introduced in HAProxy 1.8 and the first design
did not take into account the records from the Additional section.
Instead, a new resolution is associated to each server with its relevant
FQDN.
This behavior generates a lot of DNS requests (1 SRV + 1 per server
associated).
This patch aims at fixing this by:
- when a DNS response is validated, we associate A/AAAA records to
relevant SRV ones
- set a flag on associated servers to prevent them from running a DNS
resolution for said FADN
- update server IP address with information found in the Additional
section
If no relevant record can be found in the Additional section, then
HAProxy will failback to running a dedicated resolution for this server,
as it used to do.
This behavior is the one described in RFC 2782.
It is now possible to insert any attribute when a cookie is inserted by
HAProxy. Any value may be set, no check is performed except the syntax validity
(CTRL chars and ';' are forbidden). For instance, it may be used to add the
SameSite attribute:
cookie SRV insert attr "SameSite=Strict"
The attr option may be repeated to add several attributes.
This patch should fix the issue #361.
It is now possible to set the error message to use when a deny rule is
executed. It may be a specific error file, adding "errorfile <file>" :
http-request deny deny_status 400 errorfile /etc/haproxy/errorfiles/400badreq.http
It may also be an error file from an http-errors section, adding "errorfiles
<name>" :
http-request deny errorfiles my-errors # use 403 error from "my-errors" section
When defined, this error message is set in the HTTP transaction. The tarpit rule
is also concerned by this change.
It is now possible to set the error message to return to client in the HTTP
transaction. If it is defined, this error message is used instead of proxy's
errors or default errors.
It is now possible to import in a proxy, fully or partially, error files
declared in an http-errors section. It may be done using the "errorfiles"
directive, followed by a name and optionally a list of status code. If there is
no status code specified, all error files of the http-errors section are
imported. Otherwise, only error files associated to the listed status code are
imported. For instance :
http-errors my-errors
errorfile 400 ...
errorfile 403 ...
errorfile 404 ...
frontend frt
errorfiles my-errors 403 404 # ==> error 400 not imported
A new section may now be declared in the configuration to create global groups
of HTTP errors. These groups are not linked to a proxy and are referenced by
name. The section must be declared using the keyword "http-errors" followed by
the group name. This name must be unique. A list of "errorfile" directives may
be declared in such section. For instance:
http-errors website-1
errorfile 400 /path/to/site1/400.http
errorfile 404 /path/to/site1/404.http
http-errors website-2
errorfile 400 /path/to/site2/400.http
errorfile 404 /path/to/site2/404.http
For now, it is just possible to create "http-errors" sections. There is no
documentation because these groups are not used yet.
All custom HTTP errors are now stored in a global tree. Proxies use a references
on these messages. The key used for errorfile directives is the file name as
specified in the configuration. For errorloc directives, a key is created using
the redirect code and the url. This means that the same custom error message is
now stored only once. It may be used in several proxies or for several status
code, it is only parsed and stored once.
http_parse_errorloc() may now be used to create an HTTP 302 or 303 redirect
message with a specific url passed as parameter. A parameter is used to known if
it is a 302 or a 303 redirect. A status code is passed as parameter. It must be
one of the supported HTTP error codes to be valid. Otherwise an error is
returned. It aims to be used to parse "errorloc" directives. It relies on
http_load_errormsg() to do most of the job, ie converting it in HTX.
http_parse_errorfile() may now be used to parse a raw HTTP message from a
file. A status code is passed as parameter. It must be one of the supported HTTP
error codes to be valid. Otherwise an error is returned. It aims to be used to
parse "errorfile" directives. It relies on http_load_errorfile() to do most of
the job, ie reading the file content and converting it in HTX.
Now, this action is use its own dedicated function and is no longer handled "in
place" during the TCP rules evaluation. Thus the action name ACT_TCP_CAPTURE is
removed. The action type is set to ACT_CUSTOM and a check function is used to
know if the rule depends on request contents while there is no inspect-delay.
Now, these actions use their own dedicated function and are no longer handled
"in place" during the TCP/HTTP rules evaluation. Thus the action names
ACT_ACTION_TRK_SC0 and ACT_ACTION_TRK_SCMAX are removed. The action type is now
the tracking index. Thus the function trk_idx() is no longer needed.
Now, the early-hint action uses its own dedicated action and is no longer
handled "in place" during the HTTP rules evaluation. Thus the action name
ACT_HTTP_EARLY_HINT is removed. In additionn, http_add_early_hint_header() and
http_reply_103_early_hints() are also removed. This part is now handled in the
new action_ptr callback function.
Now, these actions use their own dedicated function and are no longer handled
"in place" during the HTTP rules evaluation. Thus the action names
ACT_HTTP_*_ACL and ACT_HTTP_*_MAP are removed. The action type is now mapped as
following: 0 = add-acl, 1 = set-map, 2 = del-acl and 3 = del-map.
Now, these actions use their own dedicated function and are no longer handled
"in place" during the HTTP rules evaluation. Thus the action names
ACT_HTTP_SET_HDR and ACT_HTTP_ADD_VAL are removed. The action type is now set to
0 to set a header (so remove existing ones if any and add a new one) or to 1 to
add a header (add without remove).
Now, these actions use their own dedicated function and are no longer handled
"in place" during the HTTP rules evaluation. Thus the action names
ACT_HTTP_REPLACE_HDR and ACT_HTTP_REPLACE_VAL are removed. The action type is
now set to 0 to evaluate the whole header or to 1 to evaluate every
comma-delimited values.
The function http_transform_header_str() is renamed to http_replace_hdrs() to be
more explicit and the function http_transform_header() is removed. In fact, this
last one is now more or less the new action function.
The lua code has been updated accordingly to use http_replace_hdrs().
<action> field in the act_rule structure is now an integer. The act_name values
are used for all actions without action function (but it is not a pre-requisit
though) or the action will have no effect. But for all other actions, any
integer value may used, only the action function will take care of it. The
default for such actions is ACT_CUSTOM.
Some flags can now be set on an action when it is registered. The flags are
defined in the act_flag enum. For now, only ACT_FLAG_FINAL may be set on an
action to specify if it stops the rules evaluation. It is set on
ACT_ACTION_ALLOW, ACT_ACTION_DENY, ACT_HTTP_REQ_TARPIT, ACT_HTTP_REQ_AUTH,
ACT_HTTP_REDIR and ACT_TCP_CLOSE actions. But, when required, it may also be set
on custom actions.
Consequently, this flag is checked instead of the action type during the
configuration parsing to trigger a warning when a rule inhibits all the
following ones.
The flags in the act_flag enum have been renamed act_opt. It means ACT_OPT
prefix is used instead of ACT_FLAG. The purpose of this patch is to reserve the
action flags for the actions configuration.
When TCP and HTTP rules are evaluated, if an action function (action_ptr field
in the act_rule structure) is defined for a given action, it is now always
called in priority over the test on the action type. Concretly, for now, only
custom actions define it. Thus there is no change. It just let us the choice to
extend the action type beyond the existing ones in the enum.
Info used by HTTP rules manipulating the message itself are splitted in several
structures in the arg union. But it is possible to group all of them in a unique
struct. Now, <arg.http> is used by most of these rules, which contains:
* <arg.http.i> : an integer used as status code, nice/tos/mark/loglevel or
action id.
* <arg.http.str> : an IST used as header name, reason string or auth realm.
* <arg.http.fmt> : a log-format compatible expression
* <arg.http.re> : a regular expression used by replace rules
Arguments used by actions are never released during HAProxy deinit. Now, it is
possible to specify a function to do so. ".release_ptr" field in the act_rule
structure may be set during the configuration parsing to a specific deinit
function depending on the action type.
In HTTP rules, error handling during a rewrite is now handle the same way for
all rules. First, allocation errors are reported as internal errors. Then, if
soft rewrites are allowed, rewrite errors are ignored and only the
failed_rewrites counter is incremented. Otherwise, when strict rewrites are
mandatory, interanl errors are returned.
For now, only soft rewrites are supported. Note also that the warning sent to
notify a rewrite failure was removed. It will be useless once the strict
rewrites will be possible.
the HTTP_MSGF_SOFT_RW flag must now be set on the HTTP transaction to ignore
rewrite errors on a message, from HTTP rules. The mode is called the soft
rewrites. If thes flag is not set, strict rewrites are performed. In this mode,
if a rewrite error occurred, an internal error is reported.
For now, HTTP_MSGF_SOFT_RW is always set and there is no way to switch a
transaction in strict mode.
The failed_secu counter is only used for the servers stats. It is used to report
the number of denied responses. On proxies, the same info is stored in the
denied_resp counter. So, it is more consistent to use the same field for
servers.
The stats field ST_F_EINT has been added to report internal errors encountered
per proxy, per listener and per server. It appears in the CLI export and on the
HTML stats page.
When HTTP/TCP rules are evaluated, especially HTTP ones, some results are
possible for normal actions and not for custom ones. So missing return codes
(ACT_RET_) have been added to let custom actions act as normal ones. Concretely
following codes have been added:
* ACT_RET_DENY : deny the request/response. It must be handled by the caller
* ACT_RET_ABRT : abort the request/response, handled by action itsleft.
* ACT_RET_INV : invalid request/response
Now, when HTTP rules are evaluated, HTTP_RULE_RES_ERROR must be returned when an
internal error is catched. It is a way to make the difference between a bad
request or a bad response and an error during its processing.
This counter, named 'internal_errors', has been added in frontend and backend
counters. It should be used when a internal error is encountered, instead for
failed_req or failed_resp.
Functions to deinitialize the HTTP rules are buggy. These functions does not
check the action name to release the right part in the arg union. Only few info
are released. For auth rules, the realm is released and there is no problem
here. But the regex <arg.hdr_add.re> is always unconditionally released. So it
is easy to make these functions crash. For instance, with the following rule
HAProxy crashes during the deinit :
http-request set-map(/path/to/map) %[src] %[req.hdr(X-Value)]
For now, These functions are simply removed and we rely on the deinit function
used for TCP rules (renamed as deinit_act_rules()). This patch fixes the
bug. But arguments used by actions are not released at all, this part will be
addressed later.
This patch must be backported to all stable versions.
The subscriber used to be passed as a "void *param" that was systematically
cast to a struct wait_event*. By now it appears clear that the subscribe()
call at every layer is well defined and always takes a pointer to an event
subscriber of type wait_event, so let's enforce this in the functions'
prototypes, remove the intermediary variables used to cast it and clean up
the comments to clarify what all these functions do in their context.
In practice all callers use the same wait_event notification for any I/O
so instead of keeping specific code to handle them separately, let's merge
them and it will allow us to create new events later.
For more than a decade we've kept all the sess_update_st_*() functions
in stream.c while they're only there to work in relation with what is
currently being done in backend.c (srv_redispatch_connect, connect_server,
etc). Let's move all this pollution over there and take this opportunity
to try to find slightly less confusing names for these old functions
whose role is only to handle transitions from one specific stream-int
state:
sess_update_st_rdy_tcp() -> back_handle_st_rdy()
sess_update_st_con_tcp() -> back_handle_st_con()
sess_update_st_cer() -> back_handle_st_cer()
sess_update_stream_int() -> back_try_conn_req()
sess_prepare_conn_req() -> back_handle_st_req()
sess_establish() -> back_establish()
The last one remained in stream.c because it's more or less a completion
function which does all the initialization expected on a connection
success or failure, can set analysers and emit logs.
The other ones could possibly slightly benefit from being modified to
take a stream-int instead since it's really what they're working with,
but it's unimportant here.
These ones used to serve as a set of switches between CO_FL_SOCK_* and
CO_FL_XPRT_*, and now that the SOCK layer is gone, they're always a
copy of the last know CO_FL_XPRT_* ones that is resynchronized before
I/O events by calling conn_refresh_polling_flags(), and that are pushed
back to FDs when detecting changes with conn_xprt_polling_changes().
While these functions are not particularly heavy, what they do is
totally redundant by now because the fd_want_*/fd_stop_*() actions
already perform test-and-set operations to decide to create an entry
or not, so they do the exact same thing that is done by
conn_xprt_polling_changes(). As such it is pointless to call that
one, and given that the only reason to keep CO_FL_CURR_* is to detect
changes there, we can now remove them.
Even if this does only save very few cycles, this removes a significant
complexity that has been responsible for many bugs in the past, including
the last one affecting FreeBSD.
All tests look good, and no performance regressions were observed.
CO_FL_WAIT_ROOM is set by the splicing function in raw_sock, and cleared
by the stream-int when splicing is disabled, as well as in
conn_refresh_polling_flags() so that a new call to ->rcv_pipe() could
be attempted by the I/O callbacks called from conn_fd_handler(). This
clearing in conn_refresh_polling_flags() makes no sense anymore and is
in no way related to the polling at all.
Since we don't call them from there anymore it's better to clear it
before attempting to receive, and to set it again later. So let's move
this operation where it should be, in raw_sock_to_pipe() so that it's
now symmetric. It was also placed in raw_sock_to_buf() so that we're
certain that it gets cleared if an attempt to splice is replaced with
a subsequent attempt to recv(). And these were currently already achieved
by the call to conn_refresh_polling_flags(). Now it could theorically be
removed from the stream-int.
Commit c640ef1a7d ("BUG/MINOR: stream-int: avoid calling rcv_buf() when
splicing is still possible") fixed splicing in TCP and legacy mode but
broke it badly in HTX mode.
What happens in HTX mode is that the channel's to_forward value remains
set to CHN_INFINITE_FORWARD during the whole transfer, and as such it is
not a reliable signal anymore to indicate whether more data are expected
or not. Thus, when data are spliced out of the mux using rcv_pipe(), even
when the end is reached (that only the mux knows about), the call to
rcv_buf() to get the final HTX blocks completing the message were skipped
and there was often no new event to wake this up, resulting in transfer
timeouts at the end of large objects.
All this goes down to the fact that the channel has no more information
about whether it can splice or not despite being the one having to take
the decision to call rcv_pipe() or not. And we cannot afford to call
rcv_buf() inconditionally because, as the commit above showed, this
reduces the forwarding performance by 2 to 3 in TCP and legacy modes
due to data lying in the buffer preventing splicing from being used
later.
The approach taken by this patch consists in offering the muxes the ability
to report a bit more information to the upper layers via the conn_stream.
This information could simply be to indicate that more data are awaited
but the real need being to distinguish splicing and receiving, here
instead we clearly report the mux's willingness to be called for splicing
or not. Hence the flag's name, CS_FL_MAY_SPLICE.
The mux sets this flag when it knows that its buffer is empty and that
data waiting past what is currently known may be spliced, and clears it
when it knows there's no more data or that the caller must fall back to
rcv_buf() instead.
The stream-int code now uses this to determine if splicing may be used
or not instead of looking at the rcv_pipe() callbacks through the whole
chain. And after the rcv_pipe() call, it checks the flag again to decide
whether it may safely skip rcv_buf() or not.
All this bitfield dance remains a bit complex and it starts to appear
obvious that splicing vs reading should be a decision of the mux based
on permission granted by the data layer. This would however increase
the API's complexity but definitely need to be thought about, and should
even significantly simplify the data processing layer.
The way it was integrated in mux-h1 will also result in no more calls
to rcv_pipe() on chunked encoded data, since these ones are currently
disabled at the mux level. However once the issue with chunks+splice
is fixed, it will be important to explicitly check for curr_len|CHNK
to set MAY_SPLICE, so that we don't call rcv_buf() after each chunk.
This fix must be backported to 2.1 and 2.0.
Wietse Venema reported in the thread below that we have a signedness
issue with our hashes implementations: due to the use of const char*
for the input key that's often text, the crc32, sdbm, djb2, and wt6
algorithms return a platform-dependent value for binary input keys
containing bytes with bit 7 set. This means that an ARM or PPC
platform will hash binary inputs differently from an x86 typically.
Worse, some algorithms are well defined in the industry (like CRC32)
and do not provide the expected result on x86, possibly causing
interoperability issues (e.g. a user-agent would fail to compare the
CRC32 of a message body against the one computed by haproxy).
Fortunately, and contrary to the first impression, the CRC32c variant
used in the PROXY protocol processing is not affected. Thus the impact
remains very limited (the vast majority of input keys are text-based,
such as user-agent headers for exmaple).
This patch addresses the issue by fixing all hash functions' prototypes
(even those not affected, for API consistency). A reg test will follow
in another patch.
The vast majority of users do not use these hashes. And among those
using them, very few will pass them on binary inputs. However, for the
rare ones doing it, this fix MAY have an impact during the upgrade. For
example if the package is upgraded on one LB then on another one, and
the CRC32 of a binary input is used as a stick table key (why?) then
these CRCs will not match between both nodes. Similarly, if
"hash-type ... crc32" is used, LB inconsistency may appear during the
transition. For this reason it is preferable to apply the patch on all
nodes using such hashes at the same time. Systems upgraded via their
distros will likely observe the least impact since they're expected to
be upgraded within a short time frame.
And it is important for distros NOT to skip this fix, in order to avoid
distributing an incompatible implementation of a hash. This is the
reason why this patch is tagged as MAJOR, eventhough it's extremely
unlikely that anyone will ever notice a change at all.
This patch must be backported to all supported branches since the
hashes were introduced in 1.5-dev20 (commit 98634f0c). Some parts
may be dropped since implemented later.
Link to Wietse's report:
https://marc.info/?l=postfix-users&m=157879464518535&w=2
In order to properly close connections established from Lua in case
a Lua context dies, the context currently automatically gets a flag
HLUA_MUST_GC set whenever an outgoing connection is used. This causes
the GC to be enforced on the context's death as well as on yield. First,
it does not appear necessary to do it when yielding, since if the
connections die they are already cleaned up. Second, the problem with
the flag is that even if a connection gets properly closed, the flag is
not removed and the GC continues to be called on the Lua context.
The impact on performance looks quite significant, as noticed and
diagnosed by Sadasiva Gujjarlapudi in the following thread:
https://www.mail-archive.com/haproxy@formilux.org/msg35810.html
This patch changes the flag for a counter so that each created
connection increments it and each cleanly closed connection decrements
it. That way we know we have to call the GC on the context's death only
if the count is non-null. As reported in the thread above, the Lua
performance gain is now over 20% by doing this.
Thanks to Sada and Thierry for the design discussion and tests that
led to this solution.
In tasklet_free(), to attempt to remove ourself, use MT_LIST_DEL, we can't
just use LIST_DEL(), as we theorically could be in the shared tasklet list.
This should be backported to 2.1.
In order to reduce the number of poller updates, we can benefit from
the fact that modern pollers use sampling to report readiness and that
under load they rarely report the same FD multiple times in a row. As
such it's not always necessary to disable such FDs especially when we're
almost certain they'll be re-enabled again and will require another set
of syscalls.
Now instead of creating an update for a (possibly temporary) removal,
we only perform this removal if the FD is reported again as ready while
inactive. In addition this is performed via another update so that
alternating workloads like transfers have a chance to re-enable the
FD without any syscall during the loop (typically after the data that
filled a buffer have been sent). However we only do that for single-
threaded FDs as the other ones require a more complex setup and are not
on the critical path.
This does cause a few spurious wakeups but almost totally eliminates the
calls to epoll_ctl() on connections seeing intermitent traffic like HTTP/1
to a server or client.
A typical example with 100k requests for 4 kB objects over 200 connections
shows that the number of epoll_ctl() calls doesn't depend on the number
of requests anymore but most exclusively on the number of established
connections:
Before:
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
57.09 0.499964 0 654361 321190 recvfrom
38.33 0.335741 0 369097 1 epoll_wait
4.56 0.039898 0 44643 epoll_ctl
0.02 0.000211 1 200 200 connect
------ ----------- ----------- --------- --------- ----------------
100.00 0.875814 1068301 321391 total
After:
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
59.25 0.504676 0 657600 323630 recvfrom
40.68 0.346560 0 374289 1 epoll_wait
0.04 0.000370 0 620 epoll_ctl
0.03 0.000228 1 200 200 connect
------ ----------- ----------- --------- --------- ----------------
100.00 0.851834 1032709 323831 total
As expected there is also a slight increase of epoll_wait() calls since
delaying de-activation of events can occasionally cause one spurious
wakeup.
For now this almost never happens but with subsequent patches it will
become more important not to uselessly call the I/O handlers if the FD
is not active.
Both flags became equal in commit 82967bf9 ("MINOR: connection: adjust
CO_FL_NOTIFY_DATA after removal of flags"), which already predicted the
overlap between xprt_done_cb() and wake() after the removal of the DATA
specific flags in 1.8. Let's simply remove CO_FL_NOTIFY_DATA since the
"_DONE" version already covers everything and explains the intent well
enough.
The function is not TCP-specific at all, it covers all FD-based sockets
so let's move this where other similar functions are, in connection.c,
and rename it conn_fd_check().
In practice it's all pollers except select(). It turns out that we're
keeping some legacy code only for select and enforcing it on all
pollers, let's offer the pollers the ability to declare that they
do not need that.
SSL_CTX_set_ecdh_auto() is not defined when OpenSSL 1.1.1 is compiled
with the no-deprecated option. Remove existing, incomplete guards and
add a compatibility macro in openssl-compat.h, just as OpenSSL does:
bf4006a6f9/include/openssl/ssl.h (L1486)
This should be backported as far as 2.0 and probably even 1.9.
LIBRESSL_VERSION_NUMBER evaluates to 0 under OpenSSL, making the condition
always true. Check for the define before checking it.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
[wt: to be backported as far as 1.9]
Since 1.9 with commit b20aa9eef3 ("MAJOR: tasks: create per-thread wait
queues") a task bound to a single thread will not use locks when being
queued or dequeued because the wait queue is assumed to be the owner
thread's.
But there exists a rare situation where this is not true: the health
check tasks may be running on one thread waiting for a response, and
may in parallel be requeued by another thread calling health_adjust()
after a detecting a response error in traffic when "observe l7" is set,
and "fastinter" is lower than "inter", requiring to shorten the running
check's timeout. In this case, the task being requeued was present in
another thread's wait queue, thus opening a race during task_unlink_wq(),
and gets requeued into the calling thread's wait queue instead of the
running one's, opening a second race here.
This patch aims at protecting against the risk of calling task_unlink_wq()
from one thread while the task is queued on another thread, hence unlocked,
by introducing a new TASK_SHARED_WQ flag.
This new flag indicates that a task's position in the wait queue may be
adjusted by other threads than then one currently executing it. This means
that such WQ manipulations must be performed under a lock. There are two
types of such tasks:
- the global ones, using the global wait queue (technically speaking,
those whose thread_mask has at least 2 bits set).
- some local ones, which for now will be placed into the global wait
queue as well in order to benefit from its lock.
The flag is automatically set on initialization if the task's thread mask
indicates more than one thread. The caller must also set it if it intends
to let other threads update the task's expiration delay (e.g. delegated
I/Os), or if it intends to change the task's affinity over time as this
could lead to the same situation.
Right now only the situation described above seems to be affected by this
issue, and it is very difficult to trigger, and even then, will often have
no visible effect beyond stopping the checks for example once the race is
met. On my laptop it is feasible with the following config, chained to
httpterm:
global
maxconn 400 # provoke FD errors, calling health_adjust()
defaults
mode http
timeout client 10s
timeout server 10s
timeout connect 10s
listen px
bind :8001
option httpchk /?t=50
server sback 127.0.0.1:8000 backup
server-template s 0-999 127.0.0.1:8000 check port 8001 inter 100 fastinter 10 observe layer7
This patch will automatically address the case for the checks because
check tasks are created with multiple threads bound and will get the
TASK_SHARED_WQ flag set.
If in the future more tasks need to rely on this (multi-threaded muxes
for example) and the use of the global wait queue becomes a bottleneck
again, then it should not be too difficult to place locks on the local
wait queues and queue the task on its bound thread.
This patch needs to be backported to 2.1, 2.0 and 1.9. It depends on
previous patch "MINOR: task: only check TASK_WOKEN_ANY to decide to
requeue a task".
Many thanks to William Dauchy for providing detailed traces allowing to
spot the problem.
During H1 parsing, the HTX EOM block is added before switching the message state
to H1_MSG_DONE. It is an exception in the way to convert an H1 message to
HTX. Except for this block, the message is first switched to the right state
before starting to add the corresponding HTX blocks. For instance, the message
is switched in H1_MSG_DATA state and then the HTX DATA blocks are added.
With this patch, the message is switched to the H1_MSG_DONE state when all data
blocks or trailers were processed. It is the caller responsibility to call
h1_parse_msg_eom() when the H1_MSG_DONE state is reached. This way, it is far
easier to catch failures when the HTX buffer is full.
The H1 and FCGI muxes have been updated accordingly.
This patch may eventually be backported to 2.1 if it helps other backports.
As reported in issue #380, the state check in listener_state_str() is
invalid as it allows state value 9 to report crap. We don't use such
a state value so the issue should never happen unless the memory is
already corrupted, but better clean this now while it's harmless.
This should be backported to all maintained branches.
If a new process is started with -sf and it fails to bind, it may send
a SIGTTOU to the master process in hope that it will temporarily unbind.
Unfortunately this one doesn't catch it and stops to background instead
of forwarding the signal to the workers. The same is true for SIGTTIN.
This commit simply implements an extra signal handler for the master to
deal with such signals that must be passed down to the workers. It must
be backported as far as 1.8, though there the code differs in that it's
entirely in haproxy.c and doesn't require an extra sig handler.
We used to have wake_expired_tasks() wake up tasks and return the next
expiration delay. The problem this causes is that we have to call it just
before poll() in order to consider latest timers, but this also means that
we don't wake up all newly expired tasks upon return from poll(), which
thus systematically requires a second poll() round.
This is visible when running any scheduled task like a health check, as there
are systematically two poll() calls, one with the interval, nothing is done
after it, and another one with a zero delay, and the task is called:
listen test
bind *:8001
server s1 127.0.0.1:1111 check
09:37:38.200959 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=8696843}) = 0
09:37:38.200967 epoll_wait(3, [], 200, 1000) = 0
09:37:39.202459 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=8712467}) = 0
>> nothing run here, as the expired task was not woken up yet.
09:37:39.202497 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=8715766}) = 0
09:37:39.202505 epoll_wait(3, [], 200, 0) = 0
09:37:39.202513 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=8719064}) = 0
>> now the expired task was woken up
09:37:39.202522 socket(AF_INET, SOCK_STREAM, IPPROTO_TCP) = 7
09:37:39.202537 fcntl(7, F_SETFL, O_RDONLY|O_NONBLOCK) = 0
09:37:39.202565 setsockopt(7, SOL_TCP, TCP_NODELAY, [1], 4) = 0
09:37:39.202577 setsockopt(7, SOL_TCP, TCP_QUICKACK, [0], 4) = 0
09:37:39.202585 connect(7, {sa_family=AF_INET, sin_port=htons(1111), sin_addr=inet_addr("127.0.0.1")}, 16) = -1 EINPROGRESS (Operation now in progress)
09:37:39.202659 epoll_ctl(3, EPOLL_CTL_ADD, 7, {EPOLLOUT, {u32=7, u64=7}}) = 0
09:37:39.202673 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=8814713}) = 0
09:37:39.202683 epoll_wait(3, [{EPOLLOUT|EPOLLERR|EPOLLHUP, {u32=7, u64=7}}], 200, 1000) = 1
09:37:39.202693 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=8818617}) = 0
09:37:39.202701 getsockopt(7, SOL_SOCKET, SO_ERROR, [111], [4]) = 0
09:37:39.202715 close(7) = 0
Let's instead split the function in two parts:
- the first part, wake_expired_tasks(), called just before
process_runnable_tasks(), wakes up all expired tasks; it doesn't
compute any timeout.
- the second part, next_timer_expiry(), called just before poll(),
only computes the next timeout for the current thread.
Thanks to this, all expired tasks are properly woken up when leaving
poll, and each poll call's timeout remains up to date:
09:41:16.270449 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=10223556}) = 0
09:41:16.270457 epoll_wait(3, [], 200, 999) = 0
09:41:17.270130 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=10238572}) = 0
09:41:17.270157 socket(AF_INET, SOCK_STREAM, IPPROTO_TCP) = 7
09:41:17.270194 fcntl(7, F_SETFL, O_RDONLY|O_NONBLOCK) = 0
09:41:17.270204 setsockopt(7, SOL_TCP, TCP_NODELAY, [1], 4) = 0
09:41:17.270216 setsockopt(7, SOL_TCP, TCP_QUICKACK, [0], 4) = 0
09:41:17.270224 connect(7, {sa_family=AF_INET, sin_port=htons(1111), sin_addr=inet_addr("127.0.0.1")}, 16) = -1 EINPROGRESS (Operation now in progress)
09:41:17.270299 epoll_ctl(3, EPOLL_CTL_ADD, 7, {EPOLLOUT, {u32=7, u64=7}}) = 0
09:41:17.270314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=10337841}) = 0
09:41:17.270323 epoll_wait(3, [{EPOLLOUT|EPOLLERR|EPOLLHUP, {u32=7, u64=7}}], 200, 1000) = 1
09:41:17.270332 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=10341860}) = 0
09:41:17.270340 getsockopt(7, SOL_SOCKET, SO_ERROR, [111], [4]) = 0
09:41:17.270367 close(7) = 0
This may be backported to 2.1 and 2.0 though it's unlikely to bring any
user-visible improvement except to clarify debugging.
Commit 0742c314c3 ("BUG/MEDIUM: tasks: Make sure we switch wait queues
in task_set_affinity().") had a slight side effect on expired timeouts,
which is that when used before a timeout is updated, it will cause an
existing task to be requeued earlier than its expected timeout when done
before being updated, resulting in the next poll wakup timeout too early
or even instantly if the previous wake up was done on a timeout. This is
visible in strace when health checks are enabled because there are two
poll calls, one of which has a short or zero delay. The correct solution
is to only requeue a task if it was already in the queue.
This can be backported to all branches having the fix above.
The global listener queue code and declarations were still lying in
haproxy.c while not needed there anymore at all. This complicates
the code for no reason. As a result, the global_listener_queue_task
and the global_listener_queue were made static.
We use it half times for the global_listener_queue and half times
for a proxy's queue and this requires the callers to take care of
these. Let's split it in two versions, the current one working only
on the global queue and another one dedicated to proxies for the
per-proxy queues. This cleans up quite a bit of code.
HAProxy doesn't need to call executables at run time (except when using
external checks which are strongly recommended against), and is even expected
to isolate itself into an empty chroot. As such, there basically is no valid
reason to allow a setuid executable to be called without the user being fully
aware of the risks. In a situation where haproxy would need to call external
checks and/or disable chroot, exploiting a vulnerability in a library or in
haproxy itself could lead to the execution of an external program. On Linux
it is possible to lock the process so that any setuid bit present on such an
executable is ignored. This significantly reduces the risk of privilege
escalation in such a situation. This is what haproxy does by default. In case
this causes a problem to an external check (for example one which would need
the "ping" command), then it is possible to disable this protection by
explicitly adding this directive in the global section. If enabled, it is
possible to turn it back off by prefixing it with the "no" keyword.
Before the option:
$ socat - /tmp/sock1 <<< "expert-mode on; debug dev exec sudo /bin/id"
uid=0(root) gid=0(root) groups=0(root
After the option:
$ socat - /tmp/sock1 <<< "expert-mode on; debug dev exec sudo /bin/id"
sudo: effective uid is not 0, is /usr/bin/sudo on a file system with the
'nosuid' option set or an NFS file system without root privileges?
In task_set_affinity(), leave the wait_queue if any before changing the
affinity, and re-enter a wait queue once it is done. If we don't do that,
the task may stay in the wait queue of another thread, and we later may
end up modifying that wait queue while holding no lock, which could lead
to memory corruption.
THis should be backported to 2.1, 2.0 and 1.9.
Some concerns are regularly raised about the risk to inherit some Lua
files which make use of a fork (e.g. via os.execute()) as well as
whether or not some of bugs we fix might or not be exploitable to run
some code. Given that haproxy is event-driven, any foreground activity
completely stops processing and is easy to detect, but background
activity is a different story. A Lua script could very well discretely
fork a sub-process connecting to a remote location and taking commands,
and some injected code could also try to hide its activity by creating
a process or a thread without blocking the rest of the processing. While
such activities should be extremely limited when run in an empty chroot
without any permission, it would be better to get a higher assurance
they cannot happen.
This patch introduces something very simple: it limits the number of
processes and threads to zero in the workers after the last thread was
created. By doing so, it effectively instructs the system to fail on
any fork() or clone() syscall. Thus any undesired activity has to happen
in the foreground and is way easier to detect.
This will obviously break external checks (whose concept is already
totally insecure), and for this reason a new option
"insecure-fork-wanted" was added to disable this protection, and it
is suggested in the fork() error report from the checks. It is
obviously recommended not to use it and to reconsider the reasons
leading to it being enabled in the first place.
If for any reason we fail to disable forks, we still start because it
could be imaginable that some operating systems refuse to set this
limit to zero, but in this case we emit a warning, that may or may not
be reported since we're after the fork point. Ideally over the long
term it should be conditionned by strict-limits and cause a hard fail.
Typically server line like:
'server-template srv 1-1000 *:443 ssl ca-file ca-certificates.crt'
load ca-certificates.crt 1000 times and stay duplicated in memory.
Same case for bind line: ca-file is loaded for each certificate.
Same 'ca-file' can be load one time only and stay deduplicated in
memory.
As a corollary, this will prevent file access for ca-file when
updating a certificate via CLI.
This new function looks for the first control character in a string (a
char whose value is between 0x00 and 0x1F included) and returns it, or
NULL if there is none. It is optimized for quickly evicting non-matching
strings and scans ~0.43 bytes per cycle. It can be used as an accelerator
when it's needed to look up several of these characters (e.g. CR/LF/NUL).
The link to the known bugs page for the current version is built and
reported there. When it is a development version (less than 2 dots),
instead a link to github open issues is reported as there's no way to
be sure about the current situation in this case and it's better that
users report their trouble there.
As discussed on Discourse here:
https://discourse.haproxy.org/t/haproxy-branch-support-lifetime/4466
it's not always easy for end users to know the lifecycle of the version
they are using. This patch introduces a "Status" line in the output of
"haproxy -vv" indicating whether it's a development, stable, long-term
supported version, possibly with an estimated end of life for the branch
when it can be anticipated (e.g. for stable versions). This field should
be adjusted when creating a major release to reflect the new status.
It may make sense to backport this to other branches to clarify the
situation.
Rework ssl_sock_prepare_ctx() so it fills a buffer with the error
messages instead of using ha_alert()/ha_warning(). Also returns an error
code (ERR_*) instead of the number of errors.
It was noted in #48 that there are times when a configuration
may use the server-template directive with SRV records and
simultaneously want to control weights using an agent-check or
through the runtime api. This patch adds a new option
"ignore-weight" to the "resolve-opts" directive.
When specified, any weight indicated within an SRV record will
be ignored. This is for both initial resolution and ongoing
resolution.
This patch adds three counters to help in debugging peers protocol issues
to "peer" struct:
->no_hbt counts the number of reconnection period without receiving heartbeat
->new_conn counts the number of reconnections after ->reconnect timeout expirations.
->proto_err counts the number of protocol errors.
Add RX/TX heartbeat counters to "peer" struct to have an idead about which
peer is alive or not.
Dump these counters values on the CLI via "show peers" command.
Allow the sc-set-gpt0 action to set GPT0 to a value dynamically evaluated from
its <expr> argument (in addition to the existing static <int> alternative).
The copy of the startup logs used to rely on a re-allocated memory area
on the fly, that would attempt to be delivered at once over the CLI. But
if it's too large (too many warnings) it will take time to start up, and
may not even show up on the CLI as it doesn't fit in a buffer.
The ring buffer infrastructure solves all this with no more code, let's
switch to this instead. It simply requires a parsing function to attach
the ring via ring_attach_cli() and all the rest is automatically handled.
Initially this was imagined as a code cleanup, until a test with a config
involving 100k backends and just one occurrence of
"load-server-state-from-file global" in the defaults section took approx
20 minutes to parse due to the O(N^2) cost of concatenating the warnings
resulting in ~1 TB of data to be copied, while it took only 0.57s with
the ring.
Ideally this patch should be backported to 2.0 and 1.9, though it relies
on the ring infrastructure which will then also need to be backported.
Configs able to trigger the bug are uncommon, so another workaround for
older versions without backporting the rings would consist in simply
limiting the size of the error message in print_message() to something
always printable, which will only return the first errors.
Now, for the sessions, the maximum times (queue, connect, response, total) are
reported in addition of the averages over the last 1024 connections. These
values are called qtime_max, ctime_max, rtime_max and ttime_max.
This patch is related to #272.
For backends and servers, some average times for last 1024 connections are
already calculated. For the moment, the averages for the time passed in the
queue, the connect time, the response time (for HTTP session only) and the total
time are calculated. Now, in addition, the maximum time observed for these
values are also stored.
In addition, These new counters are cleared as all other max values with the CLI
command "clear counters".
This patch is related to #272.
This change make the payload filtering uniform between TCP and HTTP
filters. Now, in TCP, like in HTTP, there is only one callback responsible to
forward data. Thus, old callbacks, tcp_data() and tcp_forward_data(), are
replaced by a single callback function, tcp_payload(). This new callback gets
the offset in the payload to (re)start the filtering and the maximum amount of
data it can forward. It is the filter's responsibility to be compatible with HTX
streams. If not, it must not set the flag FLT_CFG_FL_HTX.
Because of this change, nxt and fwd offsets are no longer needed. Thus they are
removed from the filter structure with their update functions,
flt_change_next_size() and flt_change_forward_size(). Moreover, the trace filter
has been updated accordingly.
This patch breaks the compatibility with the old API. Thus it should probably
not be backported. But, AFAIK, there is no TCP filter, thus the breakage is very
limited.
In tasklet_remove_from_tasket_list(), we can be called for a tasklet that is
either in the private task list, or in the shared tasklet list. Take that into
account and always use MT_LIST_DEL() to remove it, otherwise if we're in the
shared list and another thread attempts to add a tasklet in it, bad things
will happen.
__tasklet_remove_from_tasklet_list() is left unchanged, it's only supposed
to be used by process_runnable_task() to remove task/tasklets from the private
tast list.
This should not be backported.
This should fix github issue #357.
Since the legacy HTTP mode was removed, the stream is always released at the end
of each HTTP transaction and a new is created to handle the next request for
keep-alive connections. So the HTTP transaction is no longer reset and the
function http_reset_txn() can be removed.
Runtime traces are now supported for the streams, only if compiled with
debug. process_stream() is covered as well as TCP/HTTP analyzers and filters.
In traces, the first argument is always a stream. So it is easy to get the info
about the channels and the stream-interfaces. The second argument, when defined,
is always a HTTP transaction. And the third one is an HTTP message. The trace
message is adapted to report HTTP info when possible.
The macros DBG_TRACE_*() can be used instead of existing trace macros to emit
trace messages in debug mode only, ie, when HAProxy is compiled with DEBUG_FULL
or DEBUG_DEV. Otherwise, these macros do nothing. So it is possible to add
traces for development purpose without impacting performance of production
instances.
If the SSL_CTX of a previous instance (ckch_inst) was used as a
default_ctx, replace the default_ctx of the bind_conf by the first
SSL_CTX inserted in the SNI tree.
Use the RWLOCK of the sni tree to handle the change of the default_ctx.
It can be sometimes interesting to have a timestamp with a
resolution of less than a second.
It is currently painful to obtain this, because concatenation
of date and date_us lead to a shorter timestamp during first
100ms of a second, which is not parseable and needs ugly ACLs
in configuration to prepend 0s when needed.
To improve this, add an optional <unit> parameter to date sample
to report an integer with desired unit.
Also support this unit in http_date converter to report
a date string with sub-second precision.
Remove the leftovers of the certificate + bundle updating in 'ssl set
cert' and 'commit ssl cert'.
* Remove the it variable in appctx.ctx.ssl.
* Stop doing everything twice.
* Indent
This patch splits the 'set ssl cert' CLI command into 2 commands.
The previous way of updating the certificate on the CLI was limited with
the bundles. It was only able to apply one of the tree part of the
certificate during an update, which mean that we needed 3 updates to
update a full 3 certs bundle.
It was also not possible to apply atomically several part of a
certificate with the ability to rollback on error. (For example applying
a .pem, then a .ocsp, then a .sctl)
The command 'set ssl cert' will now duplicate the certificate (or
bundle) and update it in a temporary transaction..
The second command 'commit ssl cert' will commit all the changes made
during the transaction for the certificate.
This commit breaks the ability to update a certificate which was used as
a unique file and as a bundle in the HAProxy configuration. This way of
using the certificates wasn't making any sense.
Example:
// For a bundle:
$ echo -e "set ssl cert localhost.pem.rsa <<\n$(cat kikyo.pem.rsa)\n" | socat /tmp/sock1 -
Transaction created for certificate localhost.pem!
$ echo -e "set ssl cert localhost.pem.dsa <<\n$(cat kikyo.pem.dsa)\n" | socat /tmp/sock1 -
Transaction updated for certificate localhost.pem!
$ echo -e "set ssl cert localhost.pem.ecdsa <<\n$(cat kikyo.pem.ecdsa)\n" | socat /tmp/sock1 -
Transaction updated for certificate localhost.pem!
$ echo "commit ssl cert localhost.pem" | socat /tmp/sock1 -
Committing localhost.pem.
Success!
this patch introduces a strict-limits parameter which enforces the
setrlimit setting instead of a warning. This option can be forcingly
disable with the "no" keyword.
The general aim of this patch is to avoid bad surprises on a production
environment where you change the maxconn for example, a new fd limit is
calculated, but cannot be set because of sysfs setting. In that case you
might want to have an explicit failure to be aware of it before seeing
your traffic going down. During a global rollout it is also useful to
explictly fail as most progressive rollout would simply check the
general health check of the process.
As discussed, plan to use the strict by default mode starting from v2.3.
Signed-off-by: William Dauchy <w.dauchy@criteo.com>
In si_connect(), only switch the strema_interface status to SI_ST_RDY if
we're reusing a connection and if the connection's mux is ready. Otherwise,
maybe we're reusing a connection that is not fully established yet, and may
fail, and setting SI_ST_RDY would mean we would not be able to retry to
connect.
This should be backported to 1.9 and 2.0.
This commit depends on 55234e33708c5a584fb9efea81d71ac47235d518.
Add a new method, ctl(), to muxes. It uses a "enum mux_ctl_type" to
let it know which information we're asking for, and can output it either
directly by returning the expected value, or by using an optional argument.
"output" argument.
Right now, the only known mux_ctl_type is MUX_STATUS, that will return 0 if
the mux is not ready, or MUX_STATUS_READY if the mux is ready.
We probably want to backport this to 1.9 and 2.0.
This reverts commit 9e46496d45. It was
wrong and is not reliable, depending on the compiler's version and
optimization, as the struct is assigned inside a statement, thus on
its own stack. It's not needed anymore now so let's remove this.
We previously relied on chunk_cat(dst, b_fromist(src)) for this but it
is not reliable as the allocated buffer is inside the expression and
may be on a temporary stack. While it's possible to allocate stack space
for a struct and return a pointer to it, it's not possible to initialize
it form a temporary variable to prevent arguments from being evaluated
multiple times. Since this is only used to append an ist after a chunk,
let's instead have a chunk_istcat() function to perform exactly this
from a native ist.
The only call place (URI computation in the cache) was updated.
Debug commands will usually mark the fate of the process. We'd rather
have them counted and visible in a core or in stats output than trying
to guess how a flag combination could happen. The counter is only
incremented when the command is about to be issued however, so that
failed attempts are ignored.
Some commands like the debug ones are not enabled by default but can be
useful on some production environments. In order to avoid the temptation
of using them incorrectly, let's introduce an "expert" mode for a CLI
connection, which allows some commands to appear and be used. It is
enabled by command "expert-mode on" which is not listed by default.
8c1cddef ("MINOR: ssl: new functions duplicate and free a ckch_store")
use some OpenSSL refcount functions that were introduced in OpenSSL
1.0.2 and OpenSSL 1.1.0.
Fix the problem by introducing them in openssl-compat.h.
Fix#336.
To avoid affecting too much the traffic during a certificate update,
create the SNIs in a IO handler which yield every 10 ckch instances.
This way haproxy continues to respond even if we tries to update a
certificate which have 50 000 instances.
As reported in issue #335, a lot of contention happens on the PATLRU lock
when performing expensive regex lookups. This is absurd since the purpose
of the LRU cache was to have a fast cache for expressions, thus the cache
must not be shared between threads and must remain lockless.
This commit makes the LRU cache thread-local and gets rid of the PATLRU
lock. A test with 7 threads on 4 cores climbed from 67kH/s to 369kH/s,
or a scalability factor of 5.5.
Given the huge performance difference and the regression caused to
users migrating from processes to threads, this should be backported at
least to 2.0.
Thanks to Brian Diekelman for his detailed report about this regression.
Tasklets may be woken up to run on the calling thread or by a specific thread
(the owner). But since we use a non-thread safe mechanism when the calling
thread is also the for the owner, there may sometimes be collisions when two
threads decide to wake the same tasklet up at the same time and one of them
is the owner.
This is more of a matter of usage than code, in that a tasklet usually is
designed to be woken up and executed on the calling thread only (most cases)
or on a specific thread. Thus it is a property of the tasklet itself as this
solely depends how the code is constructed around it.
This patch performs a small change to address this. By default tasklet_new()
creates a "local" tasklet, which will run on the calling thread, like in 2.0.
This is done by setting tl->tid to a negative value. If the caller wants the
tasklet to run exclusively on a specific thread, it just has to set tl->tid,
which is already what shared tasklet callers do anyway.
No backport is needed.
The use of ~(1 << tid) to compute the sleeping_mask in tasklet_wakeup()
will result in breakage above 32 threads, because (1<<31) = 0xFFFFFFFF8000000,
and upper values will lead to theorically undefined results, but practically
will wrap over 0x1 to 0x80000000 again and indicate wrong sleeping masks. It
seems that the main visible effect maybe extra latency on some threads or
short CPU loops on others.
No backport is needed.
In MT_LIST_BEHEAD(), explicitely set the next element of the prev to NULL,
instead of setting it to the prev of the next. If we only had one element,
then we'd set the next and the prev to the element itself, and thus it would
make the element appear to be outside any list.
A lot of our chunk-based functions are able to work on a buffer pointer
but not on an ist. Instead of duplicating all of them to also take an
ist as a source, let's have a macro to make a temporary dummy buffer
from an ist. This will only result in structure field manipulations
that the compiler will quickly figure to eliminate them with inline
functions, and in other cases it will just use 4 words in the stack
before calling a function, instead of performing intermediary
conversions.
The flag HTX_FL_PROXY_RESP is now set on responses generated by HAProxy,
excluding responses returned by applets and services. It is an informative flag
set by the applicative layer.
It currently is not possible to figure the exact haproxy version from a
core file for the sole reason that the version is stored into a const
string and as such ends up in the .text section that is not part of a
core file. By turning them into variables we move them to the data
section and they appear in core files. In order to help finding them,
we just prepend an extra variable in front of them and we're able to
immediately spot the version strings from a core file:
$ strings core | fgrep -A2 'HAProxy version'
HAProxy version follows
2.1-dev2-e0f48a-88
2019/10/15
(These are haproxy_version and haproxy_date respectively). This may be
backported to 2.0 since this part is not support to impact anything but
the developer's time spent debugging.
When raw data are copied or appended in a chunk, the result must not exceed the
chunk size but it can reach it. Unlike functions to copy or append a string,
there is no terminating null byte.
This patch must be backported as far as 1.8. Note in 1.8, the functions
chunk_cpy() and chunk_cat() don't exist.
Don't try to load the files containing the issuer and the OCSP response
each time we generate a SSL_CTX.
The .ocsp and the .issuer are now loaded in the struct
cert_key_and_chain only once and then loaded from this structure when
creating a SSL_CTX.
Don't try to load the file containing the sctl each time we generate a
SSL_CTX.
The .sctl is now loaded in the struct cert_key_and_chain only once and
then loaded from this structure when creating a SSL_CTX.
Note that this now make possible the use of sctl with multi-cert
bundles.
$ echo -e "set ssl cert certificate.pem <<\n$(cat certificate2.pem)\n" | \
socat stdio /var/run/haproxy.stat
Certificate updated!
The operation is locked at the ckch level with a HA_SPINLOCK_T which
prevents the ckch architecture (ckch_store, ckch_inst..) to be modified
at the same time. So you can't do a certificate update at the same time
from multiple CLI connections.
SNI trees are also locked with a HA_RWLOCK_T so reading operations are
locked only during a certificate update.
Bundles are supported but you need to update each file (.rsa|ecdsa|.dsa)
independently. If a file is used in the configuration as a bundle AND
as a unique certificate, both will be updated.
Bundles, directories and crt-list are supported, however filters in
crt-list are currently unsupported.
The code tries to allocate every SNIs and certificate instances first,
so it can rollback the operation if that was unsuccessful.
If you have too much instances of the certificate (at least 20000 in my
tests on my laptop), the function can take too much time and be killed
by the watchdog. This will be fixed later. Also with too much
certificates it's possible that socat exits before the end of the
generation without displaying a message, consider changing the socat
timeout in this case (-t2 for example).
The size of the certificate is currently limited by the maximum size of
a payload, that must fit in a buffer.
In order to allow the creation of sni_ctx in runtime, we need to split
the function to allow rollback.
We need to be able to allocate all sni_ctxs required before inserting
them in case we need to rollback if we didn't succeed the allocation.
The function was splitted in 2 parts.
The first one ckch_inst_add_cert_sni() allocates a struct sni_ctx, fill
it with the right data and insert it in the ckch_inst's list of sni_ctx.
The second will take every sni_ctx in the ckch_inst and insert them in
the bind_conf's sni tree.
struct ckch_inst represents an instance of a certificate (ckch_node)
used in a bind_conf. Every sni_ctx created for 1 ckch_node in a
bind_conf are linked in this structure.
This patch allocate the ckch_inst for each bind_conf and inserts the
sni_ctx in its linked list.
As using an mt_list for the tasklet list is costly, instead use a regular list,
but add an mt_list for tasklet woken up by other threads, to be run on the
current thread. At the beginning of process_runnable_tasks(), we just take
the new list, and merge it into the task_list.
This should give us performances comparable to before we started using a
mt_list, but allow us to use tasklet_wakeup() from other threads.
This macro atomically cuts the head of a list and returns the list
of elements as a detached list, meaning that they're all linked
together without any head. If the list was empty, NULL is returned.
Several times some users have expressed the non-intuitive aspect of some
of our stat/info metrics and suggested to add some help. This patch
replaces the char* arrays with an array of name_desc so that we now have
some reserved room to store a description with each stat or info field.
These descriptions are currently empty and not reported yet.
Now "show info" and "show stat" can parse "desc" as an output format
modifier that will be passed down the chain to add some descriptions
to the fields depending on the format in use. For now it is not
exploited.
This flag is used to decide to show the check box in front of a proxy
on the HTML stat page. It is always equal to STAT_ADMIN except when the
proxy has no backend capability (i.e. a pure frontend) or has no server,
in which case it's only used to avoid leaving an empty column at the
beginning of the table. Not only this is pretty useless, but it also
causes the columns not to align well when mixing multiple proxies with
or without servers.
Let's simply always use STAT_ADMIN and get rid of this flag.
We used to rely on some config flags defined in uri_auth.h set during
parsing, and another set of STAT_* flags defined in stats.h set at run
time, with a somewhat gray area between the two sets. This is confusing
in the stats code as both are called "flags" in various functions and
it's quite hard to know which one describes what.
This patch cleans this up by replacing all ST_* by a newly assigned
value from the STAT_* set so that we can now use unified flags to
describe both the configuration and the current state. There is no
functional change at all.
This flag was added in 1.4-rc1 by commit 329f74d463 ("[BUG] uri_auth: do
not attemp to convert uri_auth -> http-request more than once") to
address the case where two proxies inherit the stats settings from
the defaults instance, and the first one compiles the expression while
the second one uses it. In this case since they use the exact same
uri_auth pointer, only the first one should compile and the second one
must not fail the check. This was addressed by adding an ST_CONVDONE
flag indicating that the expression conversion was completed and didn't
need to be done again. But this is a hack and it becomes cumbersome in
the middle of the other flags which are all relevant to the stats
applet. Let's instead fix it by checking if we're dealing with an
alias of the defaults instance and refrain from compiling this twice.
This allows us to remove the ST_CONVDONE flag.
A typical config requiring this check is :
defaults
mode http
stats auth foo:bar
listen l1
bind :8080
listen l2
bind :8181
Without this (or previous) check it would cmoplain when checking l2's
validity since the rule was already built.
The function http_get_authority() may be used to parse a URI and looks for the
authority, between the scheme and the path. An option may be used to skip the
user info (part before the '@'). Most of time, the user info will be ignored.
The first flag, HTX_SL_F_HAS_AUTHORITY, is set when the uri contains an
authority. For the H1, it happens when a CONNECT request is received or when an
absolute uri is used. For the H2, it happens when the pseudo header ":authority"
is provided.
The second one, HTX_SL_F_NORMALIZED_URI, is set when the received uri is
represented as an absolute uri because of the protocol requirements. For now, it
is only used for h2 requests, when the pseudo headers :authority and :scheme are
found. Internally, the uri is represented as an absolute uri. This flag allows
us to make the difference between an absolute uri in h1 and h2.
This function now dumps info about the HTX message into a buffer, passed as
argument. In addition, it is possible to only dump meta information, without the
message content.
This function now uses the address of the pointer to the htx message where the
copy must be performed. This way, when a zero-copy is performed, there is no
need to refresh the caller's htx message. It is a bit easier to do that way,
especially to add traces in the mux-h1.
William reported that since commit 6b3089856f ("MEDIUM: fd: do not use
the FD_POLL_* flags in the pollers anymore") the master's CLI often
fails to access sub-processes. There are two causes to this. One is
that we did report FD_POLL_ERR on an FD as soon as FD_EV_SHUT_W was
seen, which is automatically inherited from POLLHUP. And since we do
not store the current shutdown state of an FD we can't know if the
poller reports a sudden close resulting from an error or just a
byproduct of a previous shutdown(WR) followed by a read0. The current
patch addresses this by only considering this when the FD was active,
since a shutdown FD is not active. The second issue is that *somewhere*
down the chain, channel data are ignored if an error is reported on a
channel. This results in content truncation, but this cause was not
figured yet.
No backport is needed.
It is now possible to format stats counters as floats. But the stats applet does
not use it.
This patch is required by the Prometheus exporter to send the time averages in
seconds. If the promex change is backported, this patch must be backported
first.
A different engine-id is now generated for each thread. So, it is possible to
enable the async mode with several threads.
This patch may be backported to older versions.
We often need ISO time + microseconds in traces and ring buffers, thus
function does this by calling gettimeofday() and keeping a cached value
of the part representing the tv_sec value, and only rewrites the microsecond
part. The cache is per-thread so it's lockless and safe to use as-is.
Some tests already show that it's easy to see 3-4 events in a single
microsecond, thus it's likely that the nanosecond version will have to
be implemented as well. But certain comments on the net suggest that
some parsers are having trouble beyond microsecond, thus for now let's
stick to the microsecond only.
Now that we can wake tasklet for other threads, make sure that if the thread
is sleeping, we wake it up, or the tasklet won't be executed until it's
done sleeping.
That also means that, before going to sleep, and after we put our bit
in sleeping_thread_mask, we have to check that nobody added a tasklet for
us, just checking for global_tasks_mask isn't enough anymore.
The aim is to rassemble all scheduler information related to the current
thread. It simply points to task_per_thread[tid] without having to perform
the operation at each time. We save around 1.2 kB of code on performance
sensitive paths and increase the request rate by almost 1%.
There are a number of tests there which are enforced on tasklets while
they will never apply (various handlers, destroyed task or not, arguments,
results, ...). Instead let's have a single TASK_IS_TASKLET() test and call
the tasklet processing function directly, skipping all the rest.
It now appears visible that the only unneeded code is the update to
curr_task that is never used for tasklets, except for opportunistic
reporting in the debug handler, which can only catch si_cs_io_cb,
which in practice doesn't appear in any report so the extra cost
incurred there is pointless.
This change alone removes 700 bytes of code, mostly in
process_runnable_tasks() and increases the performance by about
1%.
Now that we can wake up a remote thread's tasklet, it's way more
interesting to use a tasklet than a task in the accept queue, as it
will avoid passing through all the scheduler. Just doing this increases
the accept rate by about 4%, overall recovering the slight loss
introduced by the tasklet change. In addition it makes sure that
even a heavily loaded scheduler (e.g. many very fast checks) will
not delay a connection accept.
Change the tasklet code so that the tasklet list is now a mt_list.
That means that tasklet now do have an associated tid, for the thread it
is expected to run on, and any thread can now call tasklet_wakeup() for
that tasklet.
One can change the associated tid with tasklet_set_tid().
Make it so MT_LIST_ADD and MT_LIST_ADDQ return 1 if it managed to add the
item, 0 (because it was already in a list) otherwise.
Make it so MT_LIST_DEL returns 1 if it managed to remove the item from a
list, or 0 otherwise (because it was in no list).
In srv_add_to_idle_list(), use LIST_DEL_INIT instead of just LIST_DEL.
We're about to add the connection to a mt_list, and MT_LIST_ADD/MT_LIST_ADDQ
will be modified to make sure we're not adding the element if it's already
in a list.
Add a few new macroes to the mt_lists.
MT_LIST_LOCK_ELT()/MT_LIST_UNLOCK_ELT() helps locking/unlocking an element.
This should only be used if you know for sure nobody else will remove the
element from the list in the meanwhile.
mt_list_for_each_entry_safe() is an iterator, similar to
list_for_each_entry_safe().
It takes 5 arguments, item, list_head, member are similar to those of
the non-mt variant, tmpelt is a temporary pointer to a struct mt_list, while
tmpelt2 is a struct mt_list itself.
MT_LIST_DEL_SELF() can be used to delete an item while parsing the list with
mt_list_for_each_entry_safe(). It shouldn't be used outside, and you
shouldn't use MT_LIST_DEL() while using mt_list_for_each_entry_safe().
Instead of using the same type for regular linked lists and "autolocked"
linked lists, use a separate type, "struct mt_list", for the autolocked one,
and introduce a set of macros, similar to the LIST_* macros, with the
MT_ prefix.
When we use the same entry for both regular list and autolocked list, as
is done for the "list" field in struct connection, we know have to explicitely
cast it to struct mt_list when using MT_ macros.
The FCGI application handles all the configuration parameters used to format
requests sent to an application. The configuration of an application is grouped
in a dedicated section (fcgi-app <name>) and referenced in a backend to be used
(use-fcgi-app <name>). To be valid, a FCGI application must at least define a
document root. But it is also possible to set the default index, a regex to
split the script name and the path-info from the request URI, parameters to set
or unset... In addition, this patch also adds a FCGI filter, responsible for
all processing on a stream.
To avoid code duplication in the futur mux FCGI, functions parsing H1 messages
and converting them into HTX have been moved in the file h1_htx.c. Some
specific parts remain in the mux H1. But most of the parsing is now generic.
Application is a generic term here. It is a modules which handle its own log
server list, with no dependency on a proxy. Such applications can now call the
function app_log() to log messages, passing a log server list and a tag as
parameters. Internally, the function __send_log() has been adapted accordingly.
Most of times, when a keyword is added in proxy section or on the server line,
we need to have a post-parser callback to check the config validity for the
proxy or the server which uses this keyword.
It is possible to register a global post-parser callback. But all these
callbacks need to loop on the proxies and servers to do their job. It is neither
handy nor efficient. Instead, it is now possible to register per-proxy and
per-server post-check callbacks.
Most of times, when any allocation is done during configuration parsing because
of a new keyword in proxy section or on the server line, we must add a call in
the deinit() function to release allocated ressources. It is now possible to
register a post-deinit callback because, at this stage, the proxies and the
servers are already releases.
Now, it is possible to register deinit callbacks per-proxy or per-server. These
callbacks will be called for each proxy and server before releasing them.
This new flag may be used to report unexpected error because of not well
formatted HTX messages (not related to a parsing error) or our incapactity to
handle the processing because we reach a limit (ressource exhaustion, too big
headers...). It should result to an error 500 returned to the client when
applicable.
It is now possible to export stats using the JSON format from the HTTP stats
page. Like for the CSV export, to export stats in JSON, you must add the option
";json" on the stats URL. It is also possible to dump the JSON schema with the
option ";json-schema". Corresponding Links have been added on the HTML page.
This patch fixes the issue #263.
This adds two extra fields to the stats, one for the current number of idle
connections and one for the configured limit. A tooltip link now appears on
the HTML page to show these values in front of the active connection values.
This should be backported to 2.0 and 1.9 as it's the only way to monitor
the idle connections behaviour.
When using "http-reuse safe", which is the default, a new incoming connection
does not automatically reuse an existing connection for the first request, as
we don't want to risk to lose the contents if we know the client will not be
able to replay the request. A side effect to this is that when dealing with
mostly http-close traffic, the reuse rate is extremely low and we keep
accumulating server-side connections that may even never be reused. At some
point we're limited to a ratio of file descriptors, but when the system is
configured with very high FD limits, we can still reach the limit of outgoing
source ports and make the system significantly slow down trying to find an
available port for outgoing connections. A simple test on my laptop with
ulimit 100000 and with the following config results in the load immediately
dropping after a few seconds :
listen l1
bind :4445
mode http
server s1 127.0.0.1:8000
As can be seen, the load falls from 38k cps to 400 cps during the first 200ms
(in fact when the source port table is full and connect() takes ages to find
a spare port for a new connection):
$ injectl464 -p 4 -o 1 -u 10 -G 127.0.0.1:4445/ -F -c -w 100
hits ^hits hits/s ^h/s bytes kB/s last errs tout htime sdht ptime
2439 2439 39338 39338 356094 5743 5743 0 0 0.4 0.5 0.4
7637 5198 38185 37666 1115002 5575 5499 0 0 0.7 0.5 0.7
7719 82 25730 820 1127002 3756 120 0 0 21.8 18.8 21.8
7797 78 19492 780 1138446 2846 114 0 0 61.4 2.5 61.4
7877 80 15754 800 1150182 2300 117 0 0 58.6 0.5 58.6
7920 43 13200 430 1156488 1927 63 0 0 58.9 0.3 58.9
At this point, lots of connections are indeed in use, for only 10 connections
on the frontend side:
$ ss -ant state established | wc -l
39022
This patch makes sure we never keep more idle connections than we've ever
had outstanding requests on a server. This way the total number of idle
connections will never exceed the sum of maximum connections. Thus highly
loaded servers will be able to get many connections and slightly loaded
servers will keep less. Ideally we should apply similar limits per process
and the per backend, but in practice this already addresses the issues
pretty well:
$ injectl464 -p 4 -o 1 -u 10 -G 127.0.0.1:4445/ -F -c -w 100
hits ^hits hits/s ^h/s bytes kB/s last errs tout htime sdht ptime
4423 4423 40209 40209 645758 5870 5870 0 0 0.2 0.4 0.2
8020 3597 40100 39966 1170920 5854 5835 0 0 0.2 0.4 0.2
12037 4017 40123 40170 1757402 5858 5864 0 0 0.2 0.4 0.2
16069 4032 40172 40320 2346074 5865 5886 0 0 0.2 0.4 0.2
20047 3978 40013 39386 2926862 5842 5750 0 0 0.3 0.4 0.3
24005 3958 40008 39979 3504730 5841 5837 0 0 0.2 0.4 0.2
$ ss -ant state established | wc -l
234
This patch must be backported to 2.0. It could be useful in 1.9 as well
eventhough pools and reuse are not enabled by default there.
As mentioned in previous commit, these flags do not map well to
modern poller capabilities. Let's use the FD_EV_*_{R,W} flags instead.
This first patch only performs a 1-to-1 mapping making sure that the
previously reported flags are still reported identically while using
the closest possible semantics in the pollers.
It's worth noting that kqueue will now support improvements such as
returning distinctions between shut and errors on each direction,
though this is not exploited for now.
There's currently a big ambiguity on our use of POLLHUP because we
currently map POLLHUP and POLLRDHUP to FD_POLL_HUP. The first one
indicates a close in *both* directions while the second one indicates
a unidirectional close. Since we don't know from the resulting flag
we always have to read when reported. Furthermore kqueue only reports
unidirectional responses which are mapped to FD_POLL_HUP as well, and
their write closes are mapped to a general error.
We could add a new FD_POLL_RDHUP flag to improve the mapping, or
switch only to the POLL* flags, but that further complicates the
portability for operating systems like FreeBSD which do not have
POLLRDHUP but have its semantics.
Let's instead directly use the per-direction flag values we already
have, and it will be a first step in the direction of finer states.
Thus we introduce an ERR and a SHUT status for each direction, that
the pollers will be able to compute and pass to fd_update_events().
It's worth noting that FD_EV_STATUS already sees the two new flags,
but they are harmless since used only by fd_{recv,send}_state() which
are never called. Thus in its current state this patch must be totally
transparent.
These two functions are used to enable recv/send but only if the FD is
not marked as active yet. The purpose is to conditionally mark them as
tentatively usable without interfering with the polling if polling was
already enabled, when it's supposed to be likely true.
Given that all our I/Os are now directed from top to bottom and not the
opposite way around, and the FD cache was removed, it doesn't make sense
anymore to create FDs that are marked not ready since this would prevent
the first accesses unless the caller explicitly does an fd_may_recv()
which is not expected to be its job (which conn_ctrl_init() has to do
by the way). Let's move this into fd_insert() instead, and have a single
atomic operation for both directions via fd_may_both().
Now that we don't have to update FD_EV_POLLED_* at the same time as
FD_EV_ACTIVE_*, we don't need to use a CAS anymore, a bit-test-and-set
operation is enough. Doing so reduces the code size by a bit more than
1 kB. One function was special, fd_done_recv(), whose comments and doc
were inaccurate for the part related to the lack of polling.
Since commit 7ac0e35f2 in 1.9-dev1 ("MAJOR: fd: compute the new fd polling
state out of the fd lock") we've started to update the FD POLLED bit a
bit more aggressively. Lately with the removal of the FD cache, this bit
is always equal to the ACTIVE bit. There's no point continuing to watch
it and update it anymore, all it does is create confusion and complicate
the code. One interesting side effect is that it now becomes visible that
all fd_*_{send,recv}() operations systematically call updt_fd_polling(),
except fd_cant_recv()/fd_cant_send() which never saw it change.
Now by prefixing a log server with "ring@<name>" it's possible to send
the logs to a ring buffer. One nice thing is that it allows multiple
sessions to consult the logs in real time in parallel over the CLI, and
without requiring file system access. At the moment, ring0 is created as
a default sink for tracing purposes and is available. No option is
provided to create new rings though this is trivial to add to the global
section.
Instead of detecting an AF_UNSPEC address family for a log server and
to deduce a file descriptor, let's create a target type field and
explicitly mention that the socket is of type FD.
The purpose is to be able to remember that initialization was already
done for a file descriptor. This will allow to get rid of some dirty
hacks performed in the logs or fd sinks where the init state of the
fd has to be guessed.
The "cache" entry was still present in the fdtab struct and it was
reported in "show sess". Removing it broke the cache-line alignment
on 64-bit machines which is important for threads, so it was fixed
by adding an attribute(aligned()) when threads are in use. Doing it
only in this case allows 32-bit thread-less platforms to see the
struct fit into 32 bytes.
Now it is possible for a reader to subscribe and wait for new events
sent to a ring buffer. When new events are written to a ring buffer,
the applets that are subscribed are woken up to display new events.
For now we only support this with the CLI applet called by "show events"
since the I/O handler is indeed a CLI I/O handler. But it's not
complicated to add other mechanisms to consume events and forward them
to external log servers for example. The wait mode is enabled by adding
"-w" after "show events <sink>". An extra "-n" was added to directly
seek to new events only.
Some CLI parsers are currently abusing the CLI context types such as
pointers to stuff longs into them by lack of room. But the context is
80 bytes while cli is only 48, thus there's some room left. This patch
adds a list element and two size_t usable as various offsets. The list
element is initialized.
The detail level initially based on syslog levels is not used, while
something related is missing, trace verbosity, to indicate whether or
not we want to call the decoding callback and what level of decoding
we want (raw captures etc). Let's change the field to "verbosity" for
this. A verbosity of zero means that the decoding callback is not
called, and all other levels are handled by this callback and are
source-specific. The source is now prompted to list the levels that
are proposed to the user. When the source doesn't define anything,
"quiet" and "default" are available.
Working on adding traces to mux-h2 revealed that the function names are
manually copied a lot in developer traces. The reason is that they are
not preprocessor macros and as such cannot be concatenated. Let's
slightly adjust the trace() function call to take a function name just
after the file:line argument. This argument is only added for the
TRACE_DEVEL and 3 new TRACE_ENTER, TRACE_LEAVE, and TRACE_POINT macros
and left NULL for others. This way the function name is only reported
for traces aimed at the developers. The pretty-print callback was also
extended to benefit from this. This will also significantly shrink the
data segment as the "entering" and "leaving" strings will now be merged.
One technical point worth mentioning is that the function name is *not*
passed as an ist to the inline function because it's not considered as
a builtin constant by the compiler, and would lead to strlen() being
run on it from all call places before calling the inline function. Thus
instead we pass the const char * (that the compiler knows where to find)
and it's the __trace() function that converts it to an ist for internal
consumption and for the pretty-print callback. Doing this avoids losing
5-10% peak performance.
The "payload" trace level was ambigous because its initial purpose was
to be able to dump received data. But it doesn't make sense to force to
report data transfers just to be able to report state changes. For
example, all snd_buf()/rcv_buf() operations coming from the application
layer should be tagged at this level. So here we move this payload level
above the state transitions and rename it to avoid the ambiguity making
one think it's only about request/response payload. Now it clearly is
about any data transfer and is thus just below the developer level. The
help messages on the CLI and the doc were slightly reworded to help
remove this ambiguity.
In prompts on the CLI we now commonly need to propose a keyword name
and a description and it doesn't make sense to define a new struct for
each such pairs. Let's simply have a generic "name_desc" for this.
Save the authority TLV in a PROXYv2 header from the client connection,
if present, and make it available as fc_pp_authority.
The fetch can be used, for example, to set the SNI for a backend TLS
connection.
Previously the callback was almost mandatory so it made sense to have it
before the message. Now that it can default to the one declared in the
trace source, most TRACE() calls contain series of empty args and callbacks,
which make them suitable for being at the end and being totally omitted.
This patch thus reverses the TRACE arguments so that the message appears
first, then the mask, then arg1..arg4, then the callback. In practice
we'll mostly see 1 arg, or 2 args and nothing else, and it will not be
needed anymore to pass long series of commas in the middle of the
arguments. However if a source is enforced, the empty commas will still
be needed for all omitted arguments.
It becomes apparent that most traces will use a single trace pretty
print callback, so let's allow the trace source to declare a default
one so that it can be omitted from trace calls, and will be used if
no other one is specified.
The principle is that when emitting a message, if some dropped events
were logged, we first attempt to report this counter before going
further. This is done under an exclusive lock while all logs are
produced under a shared lock. This ensures that the dropped line is
accurately reported and doesn't accidently arrive after a later
event.
This now provides sink_new_buf() which allocates a ring buffer. One such
ring ("buf0") of 1 MB is created already, and may be used by sink_write().
The sink's creation should probably be moved somewhere else later.
The three functions (attach, IO handler, and release) are meant to be
called by any CLI command which requires to dump the contents of a ring
buffer. We do not implement anything generic to dump any ring buffer on
the CLI since it's meant to be used by other functionalities above.
However these functions deal with locking and everything so it's trivial
to embed them in other code.
This function tries to write to the ring buffer, possibly removing enough
old messages to make room for the new one. It takes two arrays of fragments
on input to ease the insertion of prefixes by the caller. It atomically
writes the message, possibly truncating it if desired, and returns the
operation's status.
Our circular buffers are well suited for being used as ring buffers for
not-so-structured data. The machanism here consists in making room in a
buffer before inserting a new record which is prefixed by its size, and
looking up next record based on the previous one's offset and size. We
can have up to 255 consumers watching for data (dump in progress, tail)
which guarantee that entrees are not recycled while they're being dumped.
The complete representation is described in the header file. For now only
ring_new(), ring_resize() and ring_free() are created.
Currently both logs and event sinks may use a file descriptor to
atomically emit some output contents. The two may use the same FD though
nothing is done to make sure they use the same lock. Also there is quite
some redundancy between the two. Better make a specific function to send
a fragmented message to a file descriptor which will take care of the
locking via the fd's lock. The function is also able to truncate a
message and to enforce addition of a trailing LF when building the
output message.
The new functions are :
__b_put_varint() : inserts a varint when it's known that it fits
b_put_varint() : tries to insert a varint at the tail
b_get_varint() : tries to get a varint from the head
b_peek_varint() : tries to peek a varint at a specific offset
Wrapping is supported so that they are expected to be safe to use to
manipulate varints with buffers anywhere.
It will sometimes be useful to encode varints to know the output size in
advance. Two versions are provided, one inline using a switch/case construct
which will be trivial for use with constants (and will be very fast albeit
huge) and one function iterating on the number which is 5 times smaller,
for use with variables.
I forgot to fix this one before pushing, despite my tests. lockon_ptr is
only used to compare pointers, it doesn't need to point to a writable
location. Without threads the atomic store is turned into an assignment
and rightfully complains.
Given that we can pass typed arguments to the trace() function, let's
add provisions for tracking them. They are source-specific so we need
to let the source fill their name and description. Only those with a
non-null name will be proposed.
With a few macros it's possible for a trace source to commit to only
using a certain type for a given argument (or set of). This will be
particularly useful to let the trace subsystem retrieve some precious
information such as a connection, session, listener, source address or
so, and enable/disable filtering and/or locking.
The new TRACE_<level>() macros take a mask, 4 args, a callback and a
static message. From this they also inherit the TRACE_SOURCE macro from
the caller, which contains the pointer to the trace source (so that it's
not required to paste it everywhere), and an ist string is also made by
the concatenation of the file name and the line number. This uses string
concatenation by the preprocessor, and turns it into an ist by the compiler
so that there is no operation at all to perform to adjust the data length
as the compiler knows where to cut during the optimization phase. Last,
the message is also automatically turned into an ist so that it's trivial
to put it into an iovec without having to run strlen() on it.
All arguments and the callback may be empty and will then automatically
be replaced with a NULL pointer. This makes the TRACE calls slightly
lighter especially since arguments are not always used. Several other
options were considered to use variadic macros but there's no outstanding
rule that justifies to place an argument before another one, and it still
looks convenient to have the message be the last one to encourage copy-
pasting of the trace statements.
A generic TRACE() macro takes TRACE_LEVEL in from the source file as the
trace level instead of taking it from its name. This may slightly simplify
the production of traces that always run at the same level (internal core
parts may probably only be called at developer level).
The trace() call will support an optional decoding callback and 4
arguments that this function is supposed to know how to use to provide
extra information. The output remains unchanged when the function is
NULL. Otherwise, the message is pre-filled into the thread-local
trace_buf, and the function is called with all arguments so that it
completes the buffer in a readable form depending on the expected
level of detail.
This new "level" argument will allow the trace sources to label the
traces for different purposes, and filter out some of them if they
are not relevant to the current target. Right now we have 5 different
levels:
- USER : the least verbose one, only a few functional information
- PAYLOAD: like user but also displays some payload-related information
- PROTO: focuses on the protocol's framing
- STATE: also indicate state internal transitions or non-transitions
- DEVELOPER: adds extra info about branches taken in the code (break
points, return points)
We now pass an extra argument "where" to the trace() call, which
is supposed to be an ist made of the concatenation of the filename
and the line number. We only keep the last 10 chars from this string
since the end of file names is most often easy to recognize. This
gives developers useful information at very low cost.
For now it remains quite basic. It performs a few state checks, calls
the source's sink if defined, and performs the transitions between
RUNNING, STOPPED and WAITING when the configured events match.
For now it lists the sources if one is not provided, and checks
for the source's existence. It lists the events if not provided,
checks for their existence if provided, and adjusts reported
events/start/stop/pause events, and performs state transitions.
It lists sinks and adjusts them as well. Filters, lock, and
level are not implemented yet.
The principle of this subsystem will be to support taking live traces
at various places in the code with conditional triggers, filters, and
ability to lock on some elements. The traces will support typed events
and will be sent into sinks made of ring buffers, file descriptors or
remote servers.
This is the most basic type of sink. It pre-registers "stdout" and
"stderr", and is able to use writev() on them. The writev() operation
is locked to avoid mixing outputs. It's likely that the registration
should move somewhere else to take into account the fact that stdout
and stderr are still opened or are closed.
The principle will be to be able to dispatch events to various destinations
called "sinks". This is already done in part in logs where log servers can
be either a UDP socket or a file descriptor. This will be needed with the
new trace subsystem where we may also want to add ring buffers. And it turns
out that all such destinations make sense at all places. Logs may need to be
sent to a TCP server via a ring buffer, or consulted from the CLI. Trace
events may need to be sent to stdout/stderr as well as to remote log servers.
This patch creates a new structure "sink" aiming at addressing these similar
needs. The goal is to merge together what is common to all of them, such as
the output format, the dropped events count, etc, and also keep separately
the target identification (network address, file descriptor). Provisions
were made to have a "waiter" on the sink. For a TCP log server it will be
the task to wake up after writing to the log buffer. For a ring buffer, it
could be the list of watchers on the CLI running a "tail" operation and
waiting for new events. A lock was also placed in the struct since many
operations will require some locking, including the FD ones. The output
formats covers those in use by logs and two extra ones prepending the ISO
time in front of the message (convenient for stdio/buffer).
For now only the generic infrastructure is present, no type-specific
output is implemented. There's the sink_write() function which prepares
and formats a message to be sent, trying hard to avoid copies and only
using pointer manipulation, where the type-specific code just has to be
added. Dropped messages are already counted (for now 100% drop). The
message is put into an iovec array as it will be trivial to use with
file descriptors and sockets.
The function call tracing code is a quite old and was never ported to
support threads. It's not even sure whether it still works well, but
at least its presence creates confusion for future work so let's rename
it to calltrace.c and add a comment about its lack of thread-safety.
It's sometimes convenient for debugging macros not to be forced to
explicitly pass NULL in an unused argument. This macro does this, it
replaces a missing arg with NULL.
The current functions are seen outside from the debugging code and are
convenient to export so that we can improve the thread dump output :
void hlua_applet_tcp_fct(struct appctx *ctx);
void hlua_applet_http_fct(struct appctx *ctx);
struct task *hlua_process_task(struct task *task, void *context, unsigned short state);
Of course they are only available when USE_LUA is defined.
This is somewhat related to indent_msg() except that this one places a
known prefix at the beginning of each line, allows to replace the EOL
character, and not to insert a prefix on the first line if not desired.
It works with a normal output buffer/chunk so it doesn't need to allocate
anything nor to modify the input string. It is suitable for use in multi-
line backtraces.
When I/O events are being processed, we want to make sure to mark the
thread as not stuck. The reason is that some pollers (like poll()) which
do not limit the number of FDs they report could possibly report a huge
amount of FD all having to perform moderately expensive operations in
the I/O callback (e.g. via mux-pt which forwards to the upper layers),
making the watchdog think the thread is stuck since it does not schedule.
Of course this must never happen but if it ever does we must be liberal
about it.
This should be backported to 2.0, where the situation may happen more
easily due to the FD cache which can start to collect a large amount of
events. It may be related to the report in issue #201 though nothing is
certain about it.
These functions perform all the boring filling of the appctx's
cli struct needed by CLI parsers to return a message or an error,
and they return 1 so that they can be used as a single-line return
statement. They may be used for const messages or dynamic messages.
Right now we used to have extremely inconsistent states to report output,
one is CLI_ST_PRINT which prints constant message cli->msg with the
assigned severity, and CLI_ST_PRINT_FREE which prints dynamically
allocated cli->err with severity LOG_ERR, and nothing in between,
eventhough it's useful to be able to report dynamically allocated
messages as well as constant error messages.
This patch adds two extra states, which are not particularly well named
given the constraints imposed by existing ones. One is CLI_ST_PRINT_ERR
which prints a constant error message. The other one is CLI_ST_PRINT_DYN
which prints a dynamically allocated message. By doing so we maintain
the compatibility with current code.
It is important to keep in mind that we cannot pre-initialize pointers
and automatically detect what message type it is based on the assigned
fields, because the CLI's context is in a union shared with all other
users, thus unused fields contain anything upon return. This is why we
have no choice but using 4 states. Keeping the two fields <msg> and
<err> remains useful because one is const and not the other one, and
this catches may copy-paste mistakes. It's just that <err> is pretty
confusing here, it should be renamed.
The CPU time accounting field called "cpu_time" is used only by tasks
and not tasklets, yet it used to be stored into the TASK_COMMON part,
which doesn't make sense and wastes tasklet memory. In addition, moving
it to tasks also helps better group the various parts in cache lines.
Since last commit there's no point anymore in having two variants of the
same function, let's switch to b_free() only. __b_drop() was renamed to
__b_free() for obvious consistency reasons.
A small race exists in buffers with "show sess all". This one wants to show
some information grabbed from the buffer (especially in HTX mode). But the
thread owning this buffer might just be releasing its area, right after a
free() or munmap() call, resulting in a head that is not seen as empty yet
though the area was released. It may then be dereferenced by "show sess all"
causing a crash. Note that in practice it only happens in debug mode with
UAF enabled, but it's tricky enough to fix it right now.
This should be backported to stable versions which support threads and a
store barrier. It's worth noting that by performing the clearing first,
b_free() and b_drop() now become two exact equivalent.
Commit 85b2cae63 ("MINOR: pools: make the thread harmless during the
mmap/munmap syscalls") was used to relax the pressure experienced by
other threads when running in debug mode with UAF enabled. It places
a pair of thread_harmless_now()/thread_harmless_end() around the call
to mmap(), assuming callers are not sensitive to parallel activity.
But there are a few cases like "show sess all" where this happens in
isolated threads, and marking the thread as harmless there is a very
bad idea, even worse when arriving to thread_harmless_end() which loops
forever.
Let's only do that when the thread is not isolated. No backport is
needed as the patch above was only in 2.1-dev.
When parsing references to stick-tables declared as backends, they are added to
a list of proxies (they are proxies!) which refer to this stick-tables.
Before this patch we added them to these list without checking they were already
present, making the silly hypothesis the actions/sample were checked/resolved in the same
order the proxies are parsed.
This patch implement a simple inline function to in_proxies_list() to test
the presence of a proxy in a list of proxies. We use this function when resolving
/checking samples/actions.
This bug was introduced by 015e4d7 commit.
Must be backported to 2.0.
In stream_set_backend(), if we have a TCP stream, and we want to upgrade it
to H2 instead of attempting ot reuse the stream, just destroy the
conn_stream, make sure we don't log anything about the stream, and pretend
we failed setting the backend, so that the stream will get destroyed.
New streams will then be created by the mux, as if the connection just
happened.
This fixes a crash when upgrading from TCP to H2, as the H2 mux totally
ignored the conn_stream provided by the upgrade, as reported in github
issue #196.
This should be backported to 2.0.
It happens that upon looping threads the watchdog fires, starts a dump,
and other threads expire their budget while waiting for the other threads
to get dumped and trigger a watchdog event again, adding some confusion
to the traces. With this patch the situation becomes clearer as we export
the list of threads being dumped so that the watchdog can check it before
deciding to trigger. This way such threads in queue for being dumped are
not attempted to be reported in turn.
This should be backported to 2.0 as it helps understand stack traces.
In the poller code, instead of just remembering if we're currently polling
a fd or not, remember if we're polling it for writing and/or for reading, that
way, we can avoid to modify the polling if it's already polled as needed.
Now that the architecture was changed so that attempts to receive/send data
always come from the upper layers, instead of them only trying to do so when
the lower layer let them know they could try, we can finally get rid of the
fd cache. We don't really need it anymore, and removing it gives us a small
performance boost.
A problem involving server slowstart was reported by @max2k1 in issue #197.
The problem is that pendconn_grab_from_px() takes the proxy lock while
already under the server's lock while process_srv_queue() first takes the
proxy's lock then the server's lock.
While the latter seems more natural, it is fundamentally incompatible with
mayn other operations performed on servers, namely state change propagation,
where the proxy is only known after the server and cannot be locked around
the servers. Howwever reversing the lock in process_srv_queue() is trivial
and only the few functions related to dynamic cookies need to be adjusted
for this so that the proxy's lock is taken for each server operation. This
is possible because the proxy's server list is built once at boot time and
remains stable. So this is what this patch does.
The comments in the proxy and server structs were updated to mention this
rule that the server's lock may not be taken under the proxy's lock but
may enclose it.
Another approach could consist in using a second lock for the proxy's queue
which would be different from the regular proxy's lock, but given that the
operations above are rare and operate on small servers list, there is no
reason for overdesigning a solution.
This fix was successfully tested with 10000 servers in a backend where
adjusting the dyncookies in loops over the CLI didn't have a measurable
impact on the traffic.
The only workaround without the fix is to disable any occurrence of
"slowstart" on server lines, or to disable threads using "nbthread 1".
This must be backported as far as 1.8.
When a lua action or a lua sample fetch is called, a lua transaction is
created. It is an entry in the stack containing the class TXN. Thanks to it, we
can know the direction (request or response) of the call. But, for some
functions, it is also necessary to know if the buffer is "HTTP ready" for the
given direction. "HTTP ready" means there is a valid HTTP message in the
channel's buffer. So, when a lua action or a lua sample fetch is called, the
flag HLUA_TXN_HTTP_RDY is set if it is appropriate.
This one was added by commit daacf3664 ("BUG/MEDIUM: protocols: add a
global lock for the init/deinit stuff") but I forgot to add it to the
include file, breaking DEBUG_THREAD.
There is no standard case for HTTP header names because, as stated in the
RFC7230, they are case-insensitive. So applications must handle them in a
case-insensitive manner. But some bogus applications erroneously rely on the
case used by most browsers. This problem becomes critical with HTTP/2
because all header names must be exchanged in lowercase. And HAProxy uses the
same convention. All header names are sent in lowercase to clients and servers,
regardless of the HTTP version.
This design choice is linked to the HTX implementation. So, for previous
versions (2.0 and 1.9), a workaround is to disable the HTX mode to fall
back to the legacy HTTP mode.
Since the legacy HTTP mode was removed, some users reported interoperability
issues because their application was not able anymore to handle HTTP/1 message
received from HAProxy. So, we've decided to add a way to change the case of some
headers before sending them. It is now possible to define a "mapping" between a
lowercase header name and a version supported by the bogus application. To do
so, you must use the global directives "h1-case-adjust" and
"h1-case-adjust-file". Then options "h1-case-adjust-bogus-client" and
"h1-case-adjust-bogus-server" may be used in proxy sections to enable the
conversion. See the configuration manual for more info.
Of course, our advice is to urgently upgrade these applications for
interoperability concerns and because they may be vulnerable to various types of
content smuggling attacks. But, if your are really forced to use an unmaintained
bogus application, you may use these directive, at your own risks.
If it is relevant, this feature may be backported to 2.0.
Dragan Dosen found that the listeners lock is not sufficient to protect
the listeners list when proxies are stopping because the listeners are
also unlinked from the protocol list, and under certain situations like
bombing with soft-stop signals or shutting down many frontends in parallel
from multiple CLI connections, it could be possible to provoke multiple
instances of delete_listener() to be called in parallel for different
listeners, thus corrupting the protocol lists.
Such operations are pretty rare, they are performed once per proxy upon
startup and once per proxy on shut down. Thus there is no point trying
to optimize anything and we can use a global lock to protect the protocol
lists during these manipulations.
This fix (or a variant) will have to be backported as far as 1.8.
Empty error files may be used to disable the sending of any message for specific
error codes. A common use-case is to use the file "/dev/null". This way the
default error message is overridden and no message is returned to the client. It
was supported in the legacy HTTP mode, but not in HTX. Because of a bug, such
messages triggered an error.
This patch must be backported to 2.0 and 1.9. However, the patch will have to be
adapted.
When forcing the outgoing address of a connection, till now we used to
allocate this outgoing connection and set the address into it, then set
SF_ADDR_SET. With connection reuse this causes a whole lot of issues and
difficulties in the code.
Thanks to the previous changes, it is now possible to store the target
address into the stream instead, and copy the address from the stream to
the connection when initializing the connection. assign_server_address()
does this and as a result SF_ADDR_SET now reflects the presence of the
target address in the stream, not in the connection. The http_proxy mode,
the peers and the master's CLI now use the same mechanism. For now the
existing connection code was not removed to limit the amount of tricky
changes, but the allocated connection is not used anymore.
This change also revealed a latent issue that we've been having around
option http_proxy : the address was set in the connection but neither the
SF_ADDR_SET nor the SF_ASSIGNED flags were set. It looks like the connection
could establish only due to the fact that it existed with a non-null
destination address.
The purpose will be to store the target address there and not to
allocate a connection just for this anymore. For now it's only placed
in the struct, a few fields were moved to plug some holes, and the
entry is freed on release (never allocated yet for now). This must
have no impact. Note that in order to fit, the store_count which
previously was an int was turned into a short, which is way more
than enough given that the hard-coded limit is 8.
Now addresses are dynamically allocated when needed. Each connection is
created with src=dst=NULL, these entries are allocated on the fly, and
released when the connection is released.
This commit places calls to sockaddr_alloc() at the places where an address
is needed, and makes sure that the allocation is properly tested. This does
not add too many error paths since connection allocations are already in the
vicinity and share the same error paths. For the two cases where a
clear_addr() was called, instead the address was not allocated.
This pool will be used to allocate storage for source and destination
addresses used in connections. Two functions sockaddr_{alloc,free}()
were added and will have to be used everywhere an address is needed.
These ones are safe for progressive replacement as they check that the
existing pointer is set before replacing it. The pool is not yet used
during allocation nor freeing. Also they operate on pointers to pointers
so they will perform checks and replace values. The free one nulls the
pointer.
This is in preparation for the switch to dynamic address allocation,
let's migrate the code using the old fields to the pointers instead.
Note that no extra check was added for now, the purpose is only to
get the code to use the pointers and still work.
In the proxy protocol message handling we make sure the addresses are
properly allocated before declaring them unset.
At the moment we're facing difficulties with connection reuse based on
the fact that connections may be allocated very early only to set a
target address in transparent mode. With the imminent removal of the
legacy mode, the connection reuse by a same stream will not exist
anymore and all this awful complexity is not justified anymore. However
we still need to be able to assign addresses somewhere.
Thus instead of allocating a connection, we'll only place addresses where
needed in the stream during operations. But this takes quite some room
(typically 128 bytes). This is a nice opportunity for cleaning all this
up and dynamically allocatating the addresses fields, which will result
in actually saving memory from connection structs since most of the time
the client's "to" address is not used and the server's "from" is not used
either, thus saving ~256 bytes per end-to-end connection.
For now these new "src" and "dst" pointers point to addr.from and addr.to.
This will allow us to smoothly update the whole code to use these pointers
prior to going further and switching them to pools.
These functions are not used anymore. They didn't report failures and
as such were often misused. conn_get_src() and conn_get_dst() now
replaced them everywhere.
The backend connect code uses conn_get_{from,to}_addr to forward addresses
in transparent mode and to map server ports, without really checking if the
operation succeeds. In preparation of future changes, let's switch to
conn_get_{src,dst}() and integrate status check for possible failures.
These functions currently are the same as conn_get_from_addr() and
conn_get_to_addr() respectively except that they return a status for
the operation that the caller can test.
Default HTTP error messages are stored in an array of chunks. And since the HTX
was added, these messages are also converted in HTX and stored in another
array. But now, the first array is not used anymore because the legacy HTTP mode
was removed.
So now, only the array with the HTX messages are kept. The other one was
removed.
The keywords req* and rsp* are now unsupported. So the corresponding lists are
now unused. It is safe to remove them from the structure proxy.
As a result, the code dealing with these rules in HTTP analyzers was also
removed.
It was announced for the 2.1. Following keywords are now unsupported:
* reqadd, reqallow, reqiallow, reqdel, reqidel, reqdeny, reqideny, reqpass,
reqipass, reqrep, reqirep reqtarpit, reqitarpit
* rspadd, rspdel, rspidel, rspdeny, rspideny, rsprep, rspirep
a fatal error is emitted if one of these keyword is found during the
configuraion parsing.
The option 'http-tunnel' is deprecated and it was only used in the legacy HTTP
mode. So this option is now totally ignored and a warning is emitted during
HAProxy startup if it is found in a configuration file.
The old module proto_http does not exist anymore. All code dedicated to the HTTP
analysis is now grouped in the file proto_htx.c. So, to finish the polishing
after removing the legacy HTTP code, proto_htx.{c,h} files have been moved in
http_ana.{c,h} files.
In addition, all HTX analyzers and related functions prefixed with "htx_" have
been renamed to start with "http_" instead.
Many flags of the HTTP transction (TX_*) are now unused and useless. So the
flags TX_WAIT_CLEANUP, TX_HDR_CONN_*, TX_CON_CLO_SET and TX_CON_KAL_SET were
removed. Most of TX_CON_WANT_* were also removed. Only TX_CON_WANT_TUN has been
kept.
First of all, all legacy HTTP analyzers and all functions exclusively used by
them were removed. So the most of the functions in proto_http.{c,h} were
removed. Only functions to deal with the HTTP transaction have been kept. Then,
http_msg and hdr_idx modules were entirely removed. And finally the structure
http_msg was lightened of all its useless information about the legacy HTTP. The
structure hdr_ctx was also removed because unused now, just like unused states
in the enum h1_state. Note that the memory pool "hdr_idx" was removed and
"http_txn" is now smaller.
This commit breaks the compatibility with filters still relying on the legacy
HTTP code. The legacy callbacks were removed (http_data, http_chunk_trailers and
http_forward_data).
For now, the filters must still set the flag FLT_CFG_FL_HTX to be used on HTX
streams.
Since the legacy HTTP mode is disbabled, all HTTP sample fetches work on HTX
streams. So it is safe to remove all code relying on HTTP legacy mode. Among
other things, the function smp_prefetch_http() was removed with the associated
macros CHECK_HTTP_MESSAGE_FIRST() and CHECK_HTTP_MESSAGE_FIRST_PERM().
Since the legacy HTTP mode is disabled and no multiplexer relies on it anymore,
there is no reason to have 2 multiplexer protocols for the HTTP. So the protocol
PROTO_MODE_HTX was removed and all HTTP multiplexers use now PROTO_MODE_HTTP.
Because the h2 multiplexer only uses the HTX mode, following H2 functions were
removed :
* h2_prepare_h1_reqline
* h2_make_h1_request()
* h2_make_h1_trailers()
Instead of using a array of (struct block), it is more natural and intuitive to
use an array of char. Indeed, not only (struct block) are stored in this array,
but also their payload.
<head> and <tail> fields are now signed 32-bits integers. For an empty HTX
message, these fields are set to -1. So the field <used> is now useless and can
safely be removed. To know if an HTX message is empty or not, we just compare
<head> against -1 (it also works with <tail>). The function htx_nbblks() has
been added to get the number of used blocks.
A long time ago, applets were seen as an alternative to connections,
and since their respective sizes were roughly equal it appeared wise
to share the same pool. Nowadays, connections got significantly larger
but applets are not that often used, except for the cache. However
applets are mostly complementary and not alternatives anymore, as
it's very possible not to have a back connection or to share one with
other streams.
The connections will soon lose their addresses and their size will
shrink so much that appctx won't fit anymore. Given that the old
benefits of sharing these pools have long disappeared, let's stop
doing this and have a dedicated pool for appctx.
Since commit 81492c989 ("MINOR: threads: flatten the per-thread cpu-map"),
we don't keep the proc*thread matrix anymore to represent the full binding
possibilities, but only the proc and thread ones. The problem is that the
per-process binding is not the same for each thread and for the process,
and the proc[] array was assumed to store the per-proc first thread value
when doing this change. Worse, the logic present there tries to deal with
thread ranges and process ranges in a way which automatically exclused the
other possibility (since ranges cannot be used on both) but as such fails
to apply changes if neither the process nor the thread is expressed as a
range.
The real problem comes from the fact that specifying cpu-map 1/1 doesn't
yet reveal if the per-process mask or the per-thread mask needs to be
updated. In practice it's the thread one but then the current storage
doesn't allow to store the binding of the first thread of each other
process in nbproc>1 configurations.
When removing the proc*thread matrix, what ought to have been kept was
both the thread column for process 1 and the process line for threads 1,
but instead only the thread column was kept. This patch reintroduces the
storage of the configuration for the first thread of each process so that
it is again possible to store either the per-thread or per-process
configuration.
As a partial workaround for existing configurations, it is possible to
systematically indicate at least two processes or two threads at once
and map them by pairs or more so that at least two values are present
in the range. E.g :
# set processes 1-4 to cpus 0-3 :
cpu-map auto:1-4/1 0 1 2 3
# or:
cpu-map 1-2/1 0 1
cpu-map 2-3/1 2 3
# set threads 1-4 to cpus 0-3 :
cpu-map auto:1/1-4 0 1 2 3
# or :
cpu-map 1/1-2 0 1
cpu-map 3/3-4 2 3
This fix must be backported to 2.0.
Move the logic to decide if we redispatch to a new server from
sess_update_st_cer() to a new inline function, stream_choose_redispatch(), and
use it in do_l7_retry() instead of just setting the state to SI_ST_REQ.
That way, when using L7 retries, we won't redispatch the request to another
server except if "option redispatch" is used.
This should be backported to 2.0.
Sometimes we need to delegate some list processing to a function running
on another thread. In this case the list element will simply be queued
into a dedicated self-locked list and the task responsible for this list
will be woken up, calling the associated function which will run over the
list.
This is what work_list does. Such lists will be dedicated to a limited
type of work but will significantly ease such remote handling. A function
is provided to create these per-thread lists, their tasks and to properly
bind each task to a distinct thread, so that the caller only has to store
the resulting pointer to the start of the structure.
These structures should not be abused though as each head will consume
4 pointers per thread, hence 32 bytes per thread or 2 kB for 64 threads.
When we're purging idle connections, there's a race condition, when we're
removing the connection from the idle list, to add it to the list of
connections to free, if the thread owning the connection tries to free it
at the same time.
To fix this, simply add a per-thread lock, that has to be hold before
removing the connection from the idle list, and when, in conn_free(), we're
about to remove the connection from every list. That way, we know for sure
the connection will stay valid while we remove it from the idle list, to add
it to the list of connections to free.
This should happen rarely enough that it shouldn't have any impact on
performances.
This has not been reported yet, but could provoke random segfaults.
This should be backported to 2.0.
The maximum number of idle connections for a server can be configured by setting
the server option "pool-max-conn". But when we try to add a connection in its
idle list, because of a wrong comparison, it may be rejected because there are
already "pool-max-conn - 1" idle connections.
This patch must be backported to 2.0 and 1.9.
While experimenting with potentially improved fairness and latency using
ticket locks on a Ryzen 16-thread/8-core, a very strange situation happened
a lot for some levels of traffic. Around 300k connections per second, no
more connections would be accepted on the multi-threaded listener but all
others would continue to work fine. All attempts to trace showed that the
threads were all in the trylock in the fd cache, or in the spinlock of
fd_update_events(), or in the one of fd_may_recv(). But as indicated this
was not a deadlock since the process continues to work fine.
After quite some investigation it appeared that the issue is caused by a
lack of fairness between the fdcache's trylock and these functions' spin
locks above. In fact, regardless of the success or failure of the fdcache's
attempt at grabbing the lock, the poller was calling fd_update_events()
which locks the FD once for something that can be done with a CAS, and
then calls fd_may_recv() with another lock for something that most often
didn't change. The high contention on these spinlocks leaves no chance to
any other thread to grab the lock using trylock(), and once this happens,
there is no thread left to process incoming connection events nor to stop
polling on the FD, leaving all threads at 100% CPU but partially operational.
This patch addresses the issue by using bit-test-and-set instead of the OR
in fd_may_recv() / fd_may_send() so that nothing is done if the FD was
already configured as expected. It does the same in fd_update_events()
using a CAS to check if the FD's events need to be changed at all or not.
With this patch applied, it became impossible to reproduce the issue, and
now there's no way to saturate all 16 CPUs with the load used for testing,
as no more than 1350-1400 were noticed at 300+kcps vs 1600.
Ideally this patch should go further and try to remove the remaining
incarnations of the fdlock as this seems possible, but it's difficult
enough to be done in a distinct patch that will not have to be backported.
It is possible that workloads involving a high connection rate may slightly
benefit from this patch and observe a slightly lower CPU usage even when
the service doesn't misbehave.
This patch must be backported to 2.0 and 1.9.
These calls can take quite some time and leave the thread harmless so
it's better to mark it as such. This makes "show sess" respond way
faster during high loads running on processes build with DEBUG_UAF
since these calls are stressed a lot.
When calling mmap(), in general the system gives us a page but does not
really allocate it until we first dereference it. And it turns out that
this time is much longer than the time to perform the mmap() syscall.
Unfortunately, when running with memory debugging enabled, we mmap/munmap()
each object resulting in lots of such calls and a high contention on the
allocator. And the first accesses to the page being done under the pool
lock is extremely damaging to other threads.
The simple fact of writing a 0 at the beginning of the page after
allocating it and placing the POOL_LINK pointer outside of the lock is
enough to boost the performance by 8x in debug mode and to save the
watchdog from triggering on lock contention. This is what this patch
does.
The malloc and free calls and especially the underlying mmap/munmap()
can occasionally take a huge amount of time and even cause the thread
to sleep. This is visible when haproxy is compiled with DEBUG_UAF which
causes every single pool allocation/free to allocate and release pages.
In this case, when using the locked pools, the watchdog can occasionally
fire under high contention (typically requesting 40000 1M objects in
parallel over 8 threads). Then, "perf top" shows that 50% of the CPU
time is spent in mmap() and munmap(). The reason the watchdog fires is
because some threads spin on the pool lock which is held by other threads
waiting on mmap() or munmap().
This patch modifies this so that the pool lock is released during these
syscalls. Not only this allows other threads to request try to allocate
their data in parallel, but it also considerably reduces the lock
contention.
Note that the locked pools are only used on small architectures where
high thread counts would not make sense, so this will not provide any
benefit in the general case. However it makes the debugging versions
way more stable, which is always appreciated.
In the function stream_int_notify(), when the opposite stream-interface is
blocked because there is no more room into the input buffer, if the flag
CF_WRITE_PARTIAL is set on this buffer, it is unblocked. It is a way to unblock
the reads on the other side because some data was sent.
But it is a problem during the fast-forwarding because only the stream is able
to remove the flag CF_WRITE_PARTIAL. So it is possible to have this flag because
of a previous send while the input buffer of the opposite stream-interface is
now full. In such case, the opposite stream-interface will be woken up for
nothing because its input buffer is full. If the same happens on the opposite
side, we will have a loop consumming all the CPU.
To fix the bug, the opposite side is now only notify if there is some available
room in its input buffer in the function si_cs_send(), so only if some data was
sent.
This patch must be backported to 2.0 and 1.9.
This code should be now used by action to stop at the same time the rules
processing and the possible following processings. And from its side, the return
code ACT_RET_STOP should be used to only stop rules processing.
So concretely, for TCP rules, there is no changes. ACT_RET_STOP and ACT_RET_DONE
are handled the same way. However, for HTTP rules, ACT_RET_STOP should now be
mapped on HTTP_RULE_RES_STOP and ACT_RET_DONE on HTTP_RULE_RES_DONE. So this
way, a action will have the possibilty to stop all processing or only rules
processing.
Note that changes about the TCP is done in this commit but changes about the
HTTP will be done in another one because it will fix a bug in the same time.
This patch must be backported to 2.0 because a bugfix depends on it.
When deciding if we keep an idle connection in the session, check if
the number of connections currently in the session is >= the max allowed,
not >, or we'll keep an extra connection.
This should be backported to 1.9 and 2.0.
Just calling conn_force_unsubscribe() from conn_upgrade_mux_fe() is not
enough, as there may be multiple XPRT involved. Instead, require that
any user of conn_upgrade_mux_fe() unsubscribe itself before calling it.
This should fix upgrading a TCP connection to HTX when using SSL.
This should be backported to 2.0.
The receive limit of an HTX channel must be calculated against the total size of
the HTX message. Otherwise, the buffer may never be seen as full whereas the
receive limit is 0. Indeed, the function channel_htx_full() already takes care
to add a block size to the buffer's reserve (8 bytes). So if the function
channel_htx_recv_limit() also keep a block size free in addition to the buffer's
reserve, it means that at least 2 block size will be kept free but only one will
be taken into account, freezing the stream if the option http-buffer-request is
enabled.
This patch fixes the Github issue #136. It should be backported to 2.0 and
1.9. Thanks jaroslawr (Jarosław Rzeszótko) for his help.
Revert commit fe4abe62c7.
The goal was to make sure for health-checks, we would not get sockets in
TIME_WAIT. To do so, we would not call shutdown() if linger_risk is set.
However that is wrong, and that means shutw would never be forwarded to
the server, and thus we could get connection that are never properly closed.
Instead, to fix the original problem as described here :
https://www.mail-archive.com/haproxy@formilux.org/msg34080.html
Just make sure the checks code call cs_shutr() before calling cs_shutw().
If shutr has been called, conn_sock_shutw() will make no attempt to call
shutdown(), as it knows close() will be called.
We should really review and revamp the shutr/shutw code, as described in
github issue #142.
This should be backported to 1.9 and 2.0.
When using a level lower than admin on the master CLI, a \n is output
before the response, this is caused by the response of the "operator" or
"user" that are sent before the actual command.
To fix this problem we introduce the flag APPCTX_CLI_ST1_NOLF which ask
a command response to not be followed by the final \n.
This patch made a special case with the command operator and user
followed by a - so they are not followed by \n.
This patch must be backported to 2.0 and 1.9.
As its name suggest, this function change the value length of a block. But it
also update the HTX message accordingly. It simplifies the HTX API. The function
htx_set_blk_value_len() is still available and must be used with caution because
this one does not update the HTX message. It just updates the HTX block. It
should be considered as an internal function. When possible,
htx_change_blk_value_len() should be used instead.
This function is used to fix a bug affecting the 2.0. So, this patch must be
backported to 2.0.
Server states can be recovered from either a "global" file (all backends)
or a "local" file (per backend).
The way the algorithm to parse the state file was first implemented was good
enough for a low number of backends and servers per backend.
Basically, for each backend the state file (global or local) is opened,
parsed entirely and for each line we check if it contains data related to
a server from the backend we're currently processing.
We must read the file entirely, just in case some lines for the current
backend are stored at the end of the file.
This does not scale at all!
This patch changes the behavior above for the "global" file only. Now,
the global file is read and parsed once and all lines it contains are
stored in a tree, for faster discovery.
This result in way much less fopen, fgets, and strcmp calls, which make
loading of very big state files very quick now.
In commit 86eded6c6 ("CLEANUP: tasks: rename task_remove_from_tasklet_list()
to tasklet_remove_*") which consisted in removing the casts between tasks
and tasklet, I was a bit too fast to believe that we only saw tasklets in
this function since process_runnable_tasks() also uses it with tasks under
a cast. So removing the bookkeeping on task_list_size was not appropriate.
Bah, the joy of casts which hide the real thing...
This patch does two things at once to address this mess once for all:
- it restores the decrement of task_list_size when it's a real task,
but moves it to process_runnable_task() since it's the only place
where it's allowed to call it with a task
- it moves the increment there as well and renames
task_insert_into_tasklet_list() to tasklet_insert_into_tasklet_list()
of obvious consistency reasons.
This way the increment/decrement of task_list_size is made at the only
places where the cast is enforced, so it has less risks to be missed.
The comments on top of these functions were updated to reflect that they
are only supposed to be used with tasklets and that the caller is responsible
for keeping task_list_size up to date if it decides to enforce a task there.
Now we don't have to worry anymore about how these functions work outside
of the scheduler, which is better longterm-wise. Thanks to Christopher for
spotting this mistake.
No backport is needed.
In conn_sock_shutw(), avoid calling shutdown() if linger_risk is set. Not
doing so will result in getting sockets in TIME_WAIT for some time.
This is particularly observable with health checks.
This should be backported to 1.9.
The function really only operates on tasklets, its arguments are always
tasklets cast as tasks to match the function's type, to be cast back to
a struct tasklet. Let's rename it to tasklet_remove_from_tasklet_list(),
take a struct tasklet, and get rid of the undesired task casts.
It's really confusing to call it a task because it's a tasklet and used
in places where tasks and tasklets are used together. Let's rename it
to tasklet to remove this confusion.
The first one, HTX_SL_F_HAS_SCHM, will be used to know the request has an
explicit scheme. So, in H2, it is always true because the pseudo-header
":scheme" is mandatory. In H1, it is only true when an absolute URI is found on
the start-line. The other flags, HTX_SL_F_SCHM_HTTP and HTX_SL_F_SCHM_HTTPS,
will be used to know which scheme the request have. For now, other protocols are
not handled.
The aim of these flags is to pass this information to the backend side in
general, and to the H2 mux in particular. So the multiplexer will have a chance
to use this information to send the right scheme to the server.
In the HTX structure, the field <first> is used to know where to (re)start the
analysis. It may differ from the message's head. It is especially important to
update it to handle 1xx messages, to be sure to restart the analysis on the next
message (another 1xx message or the final one). It is also updated when some
data are forwarded (the headers or part of the body). But this update is an
error and must never be done at the analysis level. It is a bug, because some
sample fetches may be used after the data forwarding (but before the first send
of course). At this stage, if the first block position does not point on the
start-line, most of HTTP sample fetches fail.
So now, when something is forwarding by HTX analyzers, the first block position
is not update anymore.
This issue was reported on Github. See #119. No backport needed.
When channel_full() is called for an HTX stream, we fall back on the HTX
version. This function is called, among other, from tcp_inspect_request(). With
this patch, the inspect delay is respected again.
This patch must be backported to 1.9.
With both I/O and tasks in the same tasklet list, we now have a very
smooth and responsive scheduler, providing a good fairness between I/O
activities. With the lower layers relying on tasklet a lot (I/O wakeup,
subscribe, etc), there may often be a large number of totally autonomous
tasklets doing their business such as forwarding data between two muxes.
But the task scheduler historically refrained from picking tasks from the
priority-ordered run queue to put them into the tasklet list until this
later had less than max_runqueue_depth entries. This was to make sure that
low-latency, high-priority tasks would have an opportunity to be dequeued
before others even if they arrive late. But the counter used for this is
still the tasklet list size, which contains countless I/O events. This
causes an unfairness between unbounded I/Os and bounded tasks, resulting
for example in the CLI responding slower when forwarding 40 Gbps of HTTP
traffic spread over a thousand of connections.
A good solution consists in sticking to the initial intent of
max_runqueue_depth which is to limit the number of tasks in the list
(to maintain fairness between them) and not to limit the number of these
tasks among tasklets. It just turns out that the task_list_size initially
was this task counter and changed over time to be a tasklet list size.
Let's simply refrain from updating it for pure tasklets so that it takes
back its original role of counting real tasks as its name implies. With
this change the CLI becomes instantly responsive under load again.
This patch may possibly be backported to 1.9 though it requires some
careful checks.
The function htx_add_data_before() was removed because it was buggy. The
function htx_move_blk_before() may be used if necessary to do something
equivalent, except it just moves blocks. It doesn't handle the adding.
In an HTX message, it may have 2 available rooms to store a new block. The first
one is between the blocks and their payload. Blocks are added starting from the
end of the buffer and their payloads are added starting from the begining. So
the first free room is between these 2 edges. The second one is at the begining
of the buffer, when we start to wrap to add new payloads. Once we start to use
this one, the other one is ignored until the next defragmentation of the HTX
message.
In theory, there is no problem. But in practice, some lacks in the HTX structure
force us to defragment too often HTX messages to always be in a known state. The
second free room is not tracked as it should do and the first one may be easily
corrupted when rewrites happen.
So to fix the problem and avoid unecessary defragmentation, the HTX structure
has been refactored. The front (the block's position of the first payload before
the blocks) is no more stored. Instead we keep the relative addresses of 3 edges:
* tail_addr : The start address of the free space in front of the the blocks
table
* head_addr : The start address of the free space at the beginning
* end_addr : The end address of the free space at the beginning
Here is the general view of the HTX message now:
head_addr end_addr tail_addr
| | |
V V V
+------------+------------+------------+------------+------------------+
| | | | | |
| PAYLOAD | Free space | PAYLOAD | Free space | Blocks area |
| ==> | 1 | ==> | 2 | <== |
+------------+------------+------------+------------+------------------+
<head_addr> is always lower or equal to <end_addr> and <tail_addr>. <end_addr>
is always lower or equal to <tail_addr>.
In addition;, to simplify everything, the blocks area are now contiguous. It
doesn't wrap anymore. So the head is always the block with the lowest position,
and the tail is always the one with the highest position.
The function htx_add_data_before() is buggy and cannot work. It first add a data
block and then move it before another one, passed in argument. The problem
happens when a defragmentation is done to add the new block. In this case, the
reference is no longer valid, because the blocks are rearranged. So, instead of
moving the new block before the reference, it is moved at the head of the HTX
message.
So this function has been removed. It was only used by the compression filter to
add a last data block before a TLR, EOT or EOM block. Now, the new function
htx_add_last_data() is used. It adds a last data block, after all others and
before any TLR, EOT or EOM block. Then, the next bock is get. It is the first
non-data block after data in the HTX message. The compression loop continues
with it.
This patch must be backported to 1.9.
This function provides an alternate way to leave a critical section run
under thread_isolate(). Currently, a thread may remain in thread_release()
without having the time to notice that the rdv mask was released and taken
again by another thread entering thread_isolate() (often the same that just
released it). This is because threads wait in harmless mode in the loop,
which is compatible with the conditions to enter thread_isolate(). It's
not possible to make them wait with the harmless bit off or we cannot know
when the job is finished for the next thread to start in thread_isolate(),
and if we don't clear the rdv bit when going there, we create another
race on the start point of thread_isolate().
This new synchronous variant of thread_release() makes use of an extra
mask to indicate the threads that want to be synchronously released. In
this case, they will be marked harmless before releasing their sync bit,
and will wait for others to release their bit as well, guaranteeing that
thread_isolate() cannot be started by any of them before they all left
thread_sync_release(). This allows to construct synchronized blocks like
this :
thread_isolate()
/* optionally do something alone here */
thread_sync_release()
/* do something together here */
thread_isolate()
/* optionally do something alone here */
thread_sync_release()
And so on. This is particularly useful during initialization where several
steps have to be respected and no thread must start a step before the
previous one is completed by other threads.
This one must not be placed after any call to thread_release() or it would
risk to block an earlier call to thread_isolate() which the current thread
managed to leave without waiting for others to complete, and end up here
with the thread's harmless bit cleared, blocking others. This might be
improved in the future.
As reported in GH issue #109 and in discourse issue
https://discourse.haproxy.org/t/haproxy-returns-408-or-504-error-when-timeout-client-value-is-every-25d
the time parser doesn't error on overflows nor underflows. This is a
recurring problem which additionally has the bad taste of taking a long
time before hitting the user.
This patch makes parse_time_err() return special error codes for overflows
and underflows, and adds the control in the call places to report suitable
errors depending on the requested unit. In practice, underflows are almost
never returned as the parsing function takes care of rounding values up,
so this might possibly happen on 64-bit overflows returning exactly zero
after rounding though. It is not really possible to cut the patch into
pieces as it changes the function's API, hence all callers.
Tests were run on about every relevant part (cookie maxlife/maxidle,
server inter, stats timeout, timeout*, cli's set timeout command,
tcp-request/response inspect-delay).
When we look up an dictionary entry in the cache used upon transmission
we store the last result in ->prev_lookup of struct dcache_tx so that
to compare it with the subsequent entries to look up and save performances.
When allocating new dictionary entries we store the length of the strings.
May be useful so that not to have to call strlen() too much often at runing
time.
We store pointers to server names dictionary entries in a pre-allocated array of
ebpt_node's (->entries member of struct dcache_tx) to cache those sent to remote
peers. Consequently the ID used to identify the server name dictionary entry is
also used as index for this array. There is no need to implement a lookup by key
for this dictionary cache.
The fd_sets we've been using in the log encoding functions are not portable
and were shown to break at least under Cygwin. This patch gets rid of them
in favor of the new bitmap functions. It was verified with the config below
that the log output was exactly the same before and after the change :
defaults
mode http
option httplog
log stdout local0
timeout client 1s
timeout server 1s
timeout connect 1s
frontend foo
bind :8001
capture request header chars len 255
backend bar
option httpchk "GET" "/" "HTTP/1.0\r\nchars: \x01\x02\x03\x04\x05\x06\x07\x08\x09\x0b\x0c\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20\x21\x22\x23\x24\x25\x26\x27\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f\x30\x31\x32\x33\x34\x35\x36\x37\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f\x40\x41\x42\x43\x44\x45\x46\x47\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f\x50\x51\x52\x53\x54\x55\x56\x57\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f\x60\x61\x62\x63\x64\x65\x66\x67\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f\x70\x71\x72\x73\x74\x75\x76\x77\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
server foo 127.0.0.1:8001 check
We now have ha_bit_{set,clr,flip,test} to manipulate bitfields made
of arrays of longs. The goal is to get rid of the remaining non-portable
FD_{SET,CLR,ISSET} that still exist at a few places.
Just like we have a synchronous recv() function for the stream interface,
let's have a synchronous send function that we'll be able to call from
different places. For now this only moves the code, nothing more.
We should not update the two directions at once, in fact we should update
the Rx path after recv() and the Tx path after send(). Let's start by
splitting the update function in two for this.
The purpose of making idle-conns switch to SI_ST_CON was to make the
transition detectable and the operation retryable in case of connection
error. Now we have the RDY state for this which is much more suitable
since it indicates a validated connection on which we didn't necessarily
send anything yet. This will still lead to a transition to EST while not
requiring unnatural write polling nor connect timeouts.
The main reason for all the trouble we're facing with stream interface
error or timeout reports during the connection phase is that we currently
can't make the difference between a connection attempt and a validated
connection attempt. It is problematic because we tend to switch early
to SI_ST_EST but can't always do what we want in this state since it's
supposed to be set when we don't need to visit sess_establish() again.
This patch introduces a new state betwen SI_ST_CON and SI_ST_EST, which
is SI_ST_RDY. It indicates that we've verified that the connection is
ready. It's a transient state, like SI_ST_DIS, that cannot persist when
leaving process_stream(). For now it is not set, only verified in various
tests where SI_ST_CON was used or SI_ST_EST depending on the cases.
The stream-int state diagram was minimally updated to reflect the new
state, though it is largely obsolete and would need to be seriously
updated.
The stream interface state checks involving ranges were replaced with
checks on a set of states, already revealing some issues. No issue was
fixed, all was replaced in a one-to-one mapping for easier control. Some
checks involving a strict difference were also replaced with fields to
be clearer. At this stage, the result must be strictly equivalent. A few
tests were also turned to their bit-field equivalent for better readability
or in preparation for upcoming changes.
The test performed in the SPOE filter was swapped so that the closed and
error states are evicted first and that the established vs conn state is
tested second.
At some places we do check for ranges of stream-int states but those
are confusing as states ordering is not well known (e.g. it's not obvious
that CER is between CON and EST). Let's create a bit field from states so
that we can match multiple states at once instead. The new enum si_state_bit
contains SI_SB_* which are state bits instead of state values. The function
si_state_in() indicates if the state in argument is one of those represented
by the bit mask in second argument.
Now that the various handshakes come with their own XPRT, there's no
need for the CONN_FL_SOCK* flags, and the conn_sock_want|stop functions,
so garbage-collect them.
Add a new XPRT that is used when using non-SSL handshakes, such as proxy
protocol or Netscaler, instead of taking care of it in conn_fd_handler().
This XPRT is installed when any of those is used, and it removes itself once
the handshake is done.
This should allow us to remove the distinction between CO_FL_SOCK* and
CO_FL_XPRT*.
Add a new method to xprt_ops, remove_xprt. When called, if the provided
xprt_ctx is the same as the xprt's underlying xprt_ctx, it then uses the
new xprt provided, otherwise it calls the remove_xprt method of the next
xprt.
The goal is to be able to add a temporary xprt, that removes itself from
the chain when it did what it had to do. This will be used to implement
a pseudo-xprt for anything that just requires a handshake (such as the
proxy protocol).
As the SSL code may have different needs than the upper layer, ie it may want
to receive when the upper layer wants to right, instead of directly forwarding
the subscribe to the underlying xprt, handle it ourself. The SSL code will
know remember any subscribe call, and wake the tasklet when it is ready
for more I/O.
This type of blocks is useless because transition between data and trailers is
obvious. And when there is no trailers, the end-of-message is still there to
know when data end for chunked messages.
HTTP trailers are now parsed in the same way headers are. It means trailers are
converted to K/V blocks followed by an end-of-trailer marker. For now, to make
things simple, the type for trailer blocks are not the same than for header
blocks. But the aim is to make no difference between headers and trailers by
using the same type. Probably for the end-of marker too.
It was only done for the headers (including the EOH marker). data were prefixed
by the info field of these blocks. The payload and the trailers of the messages
were stored in raw. The total size of headers and payload were kept in the
cached object state to help output formatting.
Now, info about each HTX block is store in the cache. Only data are allowed to
be splitted. Otherwise, all blocks of an HTX message are handled the same way,
both when storing a message in the cache and when delivering it from the
cache. This will help the cache implementation to be more robust to internal
changes in the HTX. Especially for the upcoming parsing of trailers. There is
also no more need to keep extra info in the cached object state.
In channel_htx_forward() and channel_htx_forward_forever(), if the HTX message
is empty, the underlying buffer may be really empty too. And we have no warranty
the caller will call htx_to_buf() later. And in practice, it is almost never
done. So the channel's buffer must not be altered. Otherwise, the buffer may be
considered as full (data == size) for an empty HTX message and no outgoing data.
This patch must be backported to 1.9.
Make usage of the APIs implemented for dictionaries (dict.c) and their LRU caches (struct dcache)
so that to send/receive server names used for the server by name stickiness. These
names are sent over the network as follows:
- in every case we send the encode length of the data (STD_T_DICT), then
- if the server names is not present in the cache used upon transmission (struct dcache_tx)
we cache it and we the ID of this TX cache entry followed the encode length of the
server name, and finally the sever name itseft (non NULL terminated string).
- if the server name is present, we repead these operations but we only send the TX cache
entry ID.
Upon receipt, the couple of (cache IDs, server name) are stored the LRU cache used
only upon receipt (struct dcache_rx). As the peers protocol is symetrical, the fact
that the server name is present in the received data (resp. or not) denotes if
the entry is absent (resp. or not).
This simple patch only adds definitions to create a new stick-table
data type ID and a new standard type to store information in relation
wich dictionary entries (STD_T_DICT).
We want to send some stick-table data fields stored as strings in dictionaries
without consuming too much memory and CPU. To do so we implement with this patch
a cache for send/received dictionaries entries. These dictionary of strings entries are
stored in others real dictionary entries with an identifier as key (unsigned int)
and a pointer to the dictionary of strings entries as values.
This patch adds minimalistic definitions to implement dictionary new data structure
which is an ebtree of ebpt_node structs with strings as keys. Note that this has nothing
to see with real dictionary data structure (maps of keys in association with values).
As reported in GH issue #99, when hard-stop-after triggers and threads
are in use, the chance that any thread releases the resources in use by
the other ones is non-null. Thus no thread should be allowed to deinit()
nor exit by itself.
Here we take a different approach. We simply use a 3rd possible value
for the "killed" variable so that all threads know they must break out
of the run-poll-loop and immediately stop.
This patch was tested by commenting the stream_shutdown() calls in
hard_stop() to increase the chances to see a stream use released
resources. With this fix applied, it never crashes anymore.
This fix should be backported to 1.9 and 1.8.
Have "socks4" and "check-via-socks4" server keyword added.
Implement handshake with SOCKS4 proxy server for tcp stream connection.
See issue #82.
I have the "SOCKS: A protocol for TCP proxy across firewalls" doc found
at "https://www.openssh.com/txt/socks4.protocol". Please reference to it.
[wt: for now connecting to the SOCKS4 proxy over unix sockets is not
supported, and mixing IPv4/IPv6 is discouraged; indeed, the control
layer is unique for a connection and will be used both for connecting
and for target address manipulation. As such it may for example report
incorrect destination addresses in logs if the proxy is reached over
IPv6]
Remove the active_tasks_mask variable, we can deduce if we've work to do
by other means, and it is costly to maintain. Instead, introduce a new
function, thread_has_tasks(), that returns non-zero if there's tasks
scheduled for the thread, zero otherwise.
Add session flags, and add a new flag, SESS_FL_PREFER_LAST, to be set when
we use NTLM authentication, and we should reuse the last connection. This
should fix using NTLM with HTX. This totally replaces TX_PREFER_LAST.
This should be backported to 1.9.
In lock profiles it's visible that there is a huge contention on the
buffer lock. The reason is that when offer_buffers() is called, it
systematically takes the lock before verifying if there is any
waiter. However doing so doesn't protect against races since a
waiter can happen just after we release the lock as well. Similarly
in h2 we take the lock every time an h2c is going to be released,
even without checking that the h2c belongs to a wait list. These
two have now been addressed by verifying non-emptiness of the list
prior to taking the lock.
Haproxy is designed to be able to continue to run even under very low
memory conditions. However this can sometimes have a serious impact on
performance that it hard to diagnose. Let's report counters of failed
pool and buffer allocations per thread in show activity.
We have been abusing the do_poll()'s timeout for a while, making it zero
whenever there is some known activity. The problem this poses is that it
complicates activity diagnostic by incrementing the poll_exp field for
each known activity. It also requires extra computations that could be
avoided.
This change passes a "wake" argument to say that the poller must not
sleep. This simplifies the operations and allows one to differenciate
expirations from activity.
In order to later allow htx_add_data() to transmit partial blocks and
avoid defragmenting the buffer, we'll need to return the number of bytes
consumed. This first modification makes the function do this and its
callers take this into account. At the moment the function still works
atomically so it returns either the block size or zero. However all
call places have been adapted to consider any value between zero and
the block size.
