This patch cleanups the direction names. It replaces numeric values,
by the associated defines. It ensure the compliance with values found
somwhere else in HAProxy.
It is required by the bugfix patch which is following.
[wt: needs to be backported to 1.6]
Current resolvers section parsing function is permissive on nameserver
id and two nameservers may have the same id.
It's a shame, since we don't know for example, whose statistics belong
to which nameserver...
From now, configuration with duplicated nameserver id in a resolvers
section are considered as broken and returns a fatal error when parsing.
Cyril Bonté reported a reproduceable sequence which can lead to a crash
when using backend connection reuse. The problem comes from the fact that
we systematically add the server connection to an idle pool at the end of
the HTTP transaction regardless of the fact that it might already be there.
This is possible for example when processing a request which doesn't use
a server connection (typically a redirect) after a request which used a
connection. Then after the first request, the connection was already in
the idle queue and we're putting it a second time at the end of the second
request, causing a corruption of the idle pool.
Interestingly, the memory debugger in 1.7 immediately detected a suspicious
double free on the connection, leading to a very early detection of the
cause instead of its consequences.
Thanks to Cyril for quickly providing a working reproducer.
This fix must be backported to 1.6 since connection reuse was introduced
there.
A bug lied in the parsing of DNS CNAME response, leading HAProxy to
think the CNAME was improperly resolved in the response.
This should be backported into 1.6 branch
The status DNS_UPD_NAME_ERROR returned by dns_get_ip_from_response and
which means the queried name can't be found in the response was
improperly processed (fell into the default case).
This lead to a loop where HAProxy simply resend a new query as soon as
it got a response for this status and in the only case where such type
of response is the very first one received by the process.
This should be backported into 1.6 branch
It was accidently discovered that limiting haproxy to 5000 MB leads to
an effective limit of 904 MB. This is because the computation for the
size limit is performed by multiplying rlimit_memmax by 1048576, and
doing so causes the operation to be performed on an int instead of a
long or long long. Just switch to 1048576ULL as is done at other places
to fix this.
This bug affects all supported versions, the backport is desired, though
it rarely affects users since few people apply memory limits.
When DEBUG_MEMORY_POOLS is used, we now use the link pointer at the end
of the pool to store a pointer to the pool, and to control it during
pool_free2() in order to serve four purposes :
- at any instant we can know what pool an object was allocated from
when examining memory, hence how we should possibly decode it ;
- it serves to detect double free when they happen, as the pointer
cannot be valid after the element is linked into the pool ;
- it serves to detect if an element is released in the wrong pool ;
- it serves as a canary, to detect if some buffers experienced an
overflow before being release.
All these elements will definitely help better troubleshoot strange
situations, or at least confirm that certain conditions did not happen.
When debugging a core file, it's sometimes convenient to be able to
visit the released entries in the pools (typically last released
session). Unfortunately the first bytes of these entries are destroyed
by the link elements of the pool. And of course, most structures have
their most accessed elements at the beginning of the structure (typically
flags). Let's add a build-time option DEBUG_MEMORY_POOLS which allocates
an extra pointer in each pool to put the link at the end of each pool
item instead of the beginning.
This commit adds support for setting a per-server maxconn from the stats
socket. The only really notable part of this commit is that we need to
check if maxconn == minconn before changing things, as this indicates
that we are NOT using dynamic maxconn. When we are not using dynamic
maxconn, we should update maxconn/minconn in lockstep.
The function 'EVP_PKEY_get_default_digest_nid()' was introduced in OpenSSL
1.0.0. So for older version of OpenSSL, compiled with the SNI support, the
HAProxy compilation fails with the following error:
src/ssl_sock.c: In function 'ssl_sock_do_create_cert':
src/ssl_sock.c:1096:7: warning: implicit declaration of function 'EVP_PKEY_get_default_digest_nid'
if (EVP_PKEY_get_default_digest_nid(capkey, &nid) <= 0)
[...]
src/ssl_sock.c:1096: undefined reference to `EVP_PKEY_get_default_digest_nid'
collect2: error: ld returned 1 exit status
Makefile:760: recipe for target 'haproxy' failed
make: *** [haproxy] Error 1
So we must add a #ifdef to check the OpenSSL version (>= 1.0.0) to use this
function. It is used to get default signature digest associated to the private
key used to sign generated X509 certificates. It is called when the private key
differs than EVP_PKEY_RSA, EVP_PKEY_DSA and EVP_PKEY_EC. It should be enough for
most of cases.
Basically, it's ill-defined and shouldn't really be used going forward.
We can't guarantee that resolvers will do the 'legwork' for us and
actually resolve CNAMES when we request the ANY query-type. Case in point
(obfuscated, clearly):
PRODUCTION! ahayworth@secret-hostname.com:~$
dig @10.11.12.53 ANY api.somestartup.io
; <<>> DiG 9.8.4-rpz2+rl005.12-P1 <<>> @10.11.12.53 ANY api.somestartup.io
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 62454
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 4, ADDITIONAL: 0
;; QUESTION SECTION:
;api.somestartup.io. IN ANY
;; ANSWER SECTION:
api.somestartup.io. 20 IN CNAME api-somestartup-production.ap-southeast-2.elb.amazonaws.com.
;; AUTHORITY SECTION:
somestartup.io. 166687 IN NS ns-1254.awsdns-28.org.
somestartup.io. 166687 IN NS ns-1884.awsdns-43.co.uk.
somestartup.io. 166687 IN NS ns-440.awsdns-55.com.
somestartup.io. 166687 IN NS ns-577.awsdns-08.net.
;; Query time: 1 msec
;; SERVER: 10.11.12.53#53(10.11.12.53)
;; WHEN: Mon Oct 19 22:02:29 2015
;; MSG SIZE rcvd: 242
HAProxy can't handle that response correctly.
Rather than try to build in support for resolving CNAMEs presented
without an A record in an answer section (which may be a valid
improvement further on), this change just skips ANY record types
altogether. A and AAAA are much more well-defined and predictable.
Notably, this commit preserves the implicit "Prefer IPV6 behavior."
Furthermore, ANY query type by default is a bad idea: (from Robin on
HAProxy's ML):
Using ANY queries for this kind of stuff is considered by most people
to be a bad practice since besides all the things you named it can
lead to incomplete responses. Basically a resolver is allowed to just
return whatever it has in cache when it receives an ANY query instead
of actually doing an ANY query at the authoritative nameserver. Thus
if it only received queries for an A record before you do an ANY query
you will not get an AAAA record even if it is actually available since
the resolver doesn't have it in its cache. Even worse if before it
only got MX queries, you won't get either A or AAAA
Kim Seri reported that haproxy 1.6.0 crashes after a few requests
when a bind line has SSL enabled with more than one certificate. This
was caused by an insufficient condition to free generated certs during
ssl_sock_close() which can also catch other certs.
Christopher Faulet analysed the situation like this :
-------
First the LRU tree is only initialized when the SSL certs generation is
configured on a bind line. So, in the most of cases, it is NULL (it is
not the same thing than empty).
When the SSL certs generation is used, if the cache is not NULL, a such
certificate is pushed in the cache and there is no need to release it
when the connection is closed.
But it can be disabled in the configuration. So in that case, we must
free the generated certificate when the connection is closed.
Then here, we have really a bug. Here is the buggy part:
3125) if (conn->xprt_ctx) {
3126) #ifdef SSL_CTRL_SET_TLSEXT_HOSTNAME
3127) if (!ssl_ctx_lru_tree && objt_listener(conn->target)) {
3128) SSL_CTX *ctx = SSL_get_SSL_CTX(conn->xprt_ctx);
3129) if (ctx != 3130)
SSL_CTX_free(ctx);
3131) }
3133) SSL_free(conn->xprt_ctx);
3134) conn->xprt_ctx = NULL;
3135) sslconns--;
3136) }
The check on the line 3127 is not enough to determine if this is a
generated certificate or not. Because ssl_ctx_lru_tree is NULL,
generated certificates, if any, must be freed. But here ctx should also
be compared to all SNI certificates and not only to default_ctx. Because
of this bug, when a SNI certificate is used for a connection, it is
erroneously freed when this connection is closed.
-------
Christopher provided this reliable reproducer :
----------
global
tune.ssl.default-dh-param 2048
daemon
listen ssl_server
mode tcp
bind 127.0.0.1:4443 ssl crt srv1.test.com.pem crt srv2.test.com.pem
timeout connect 5000
timeout client 30000
timeout server 30000
server srv A.B.C.D:80
You just need to generate 2 SSL certificates with 2 CN (here
srv1.test.com and srv2.test.com).
Then, by doing SSL requests with the first CN, there is no problem. But
with the second CN, it should segfault on the 2nd request.
openssl s_client -connect 127.0.0.1:4443 -servername srv1.test.com // OK
openssl s_client -connect 127.0.0.1:4443 -servername srv1.test.com // OK
But,
openssl s_client -connect 127.0.0.1:4443 -servername srv2.test.com // OK
openssl s_client -connect 127.0.0.1:4443 -servername srv2.test.com // KO
-----------
A long discussion led to the following proposal which this patch implements :
- the cert is generated. It gets a refcount = 1.
- we assign it to the SSL. Its refcount becomes two.
- we try to insert it into the tree. The tree will handle its freeing
using SSL_CTX_free() during eviction.
- if we can't insert into the tree because the tree is disabled, then
we have to call SSL_CTX_free() ourselves, then we'd rather do it
immediately. It will more closely mimmick the case where the cert
is added to the tree and immediately evicted by concurrent activity
on the cache.
- we never have to call SSL_CTX_free() during ssl_sock_close() because
the SSL session only relies on openssl doing the right thing based on
the refcount only.
- thus we never need to know how the cert was created since the
SSL_CTX_free() is either guaranteed or already done for generated
certs, and this protects other ones against any accidental call to
SSL_CTX_free() without having to track where the cert comes from.
This patch also reduces the inter-dependence between the LRU tree and
the SSL stack, so it should cause less sweating to migrate to threads
later.
This bug is specific to 1.6.0, as it was introduced after dev7 by
this fix :
d2cab92 ("BUG/MINOR: ssl: fix management of the cache where forged certificates are stored")
Thus a backport to 1.6 is required, but not to 1.5.
Susheel Jalali reported a confusing bug in namespaces implementation.
If namespaces are enabled at build time (USE_NS=1) and *no* namespace
is used at all in the whole config file, my_socketat() returns -1 and
all socket bindings fail. This is because of a wrong condition in this
function. A possible workaround consists in creating some namespaces.
The function which parses a DNS response buffer did not move properly a
pointer when reading a packet where records does not use DNS "message
compression" techniques.
Thanks to 0yvind Johnsen for the help provided during the troubleshooting
session.
This is equivalent to commit 2af207a ("MEDIUM: tcp: implement tcp-ut
bind option to set TCP_USER_TIMEOUT") except that this time it works
on the server side. The purpose is to detect dead server connections
even when checks are rare, disabled, or after a soft reload (since
checks are disabled there as well), and to ensure client connections
will get killed faster.
Baptiste reported a segfault when the "id" keyword was passed on the
"stats socket" line. The problem is related to the fact that the stats
parser stats_parse_global() passes curpx instead of global.stats_fe to
the keyword parser. Indeed, curpx being a pointer to the proxy in the
current section, it is not correct here since the global section does
not describe a proxy. It's just by pure luck that only bind_parse_id()
uses the proxy since any other keyword parser could use it as well.
The bug has no impact since the id specified here is not usable at all
and can be discarded from a faulty configuration.
This fix must be backported to 1.5.
Lua needs to known the direction of the http data processed (request or
response). It checks the flag SMP_OPT_DIR_REQ, buf this flag is 0. This patch
correctly checks the flags after applying the SMP_OPT_DIR mask.
A previous commit broke the interactive stats cli prompt. Specifically,
it was not clear that we could be in STAT_CLI_PROMPT when we get to
the output functions for the cli handler, and the switch statement did
not handle this case. We would then fall through to the default
statement, which was recently changed to set error flags on the socket.
This in turn causes the socket to be closed, which is not what we wanted
in this specific case.
To fix, we add a case for STAT_CLI_PROMPT, and simply break out of the
switch statement.
Testing:
- Connected to unix stats socket, issued 'prompt', observed that I
could issue multiple consecutive commands.
- Connected to unix stats socket, issued 'prompt', observed that socket
timed out after inactivity expired.
- Connected to unix stats socket, issued 'prompt' then 'set timeout cli
5', observed that socket timed out after 5 seconds expired.
- Connected to unix stats socket, issued invalid commands, received
usage output.
- Connected to unix stats socket, issued 'show info', received info
output and socket disconnected.
- Connected to unix stats socket, issued 'show stat', received stats
output and socket disconnected.
- Repeated above tests with TCP stats socket.
[wt: no backport needed, this was introduced during the applet rework in 1.6]
Now, A callback is defined for generated certificates to set DH parameters for
ephemeral key exchange when required.
In same way, when possible, we also defined Elliptic Curve DH (ECDH) parameters.
This is done by adding EVP_PKEY_EC type in supported types for the CA private
key when we get the message digest used to sign a generated X509 certificate.
So now, we support DSA, RSA and EC private keys.
And to be sure, when the type of the private key is not directly supported, we
get its default message digest using the function
'EVP_PKEY_get_default_digest_nid'.
We also use the key of the default certificate instead of generated it. So we
are sure to use the same key type instead of always using a RSA key.
the file specified by the SSL option 'ca-sign-file' can now contain the CA
certificate used to dynamically generate certificates and its private key in any
order.
Commit d2cab92 ("BUG/MINOR: ssl: fix management of the cache where forged
certificates are stored") removed some needed #ifdefs resulting in ssl not
building on older openssl versions where SSL_CTRL_SET_TLSEXT_HOSTNAME is
not defined :
src/ssl_sock.c: In function 'ssl_sock_load_ca':
src/ssl_sock.c:2504: error: 'ssl_ctx_lru_tree' undeclared (first use in this function)
src/ssl_sock.c:2504: error: (Each undeclared identifier is reported only once
src/ssl_sock.c:2504: error: for each function it appears in.)
src/ssl_sock.c:2505: error: 'ssl_ctx_lru_seed' undeclared (first use in this function)
src/ssl_sock.c: In function 'ssl_sock_close':
src/ssl_sock.c:3095: error: 'ssl_ctx_lru_tree' undeclared (first use in this function)
src/ssl_sock.c: In function '__ssl_sock_deinit':
src/ssl_sock.c:5367: error: 'ssl_ctx_lru_tree' undeclared (first use in this function)
make: *** [src/ssl_sock.o] Error 1
Reintroduce the ifdefs around the faulty areas.
First, the LRU cache must be initialized after the configuration parsing to
correctly set its size.
Next, the function 'ssl_sock_set_generated_cert' returns -1 when an error occurs
(0 if success). In that case, the caller is responsible to free the memory
allocated for the certificate.
Finally, when a SSL certificate is generated by HAProxy but cannot be inserted
in the cache, it must be freed when the SSL connection is closed. This happens
when 'tune.ssl.ssl-ctx-cache-size' is set to 0.
The 'OPTIONS' method was not in the list of supported HTTP methods and
find_http_meth return HTTP_METH_OTHER instead of HTTP_METH_OPTIONS.
[wt: this fix needs to be backported at least to 1.5, 1.4 and 1.3]
When debugging an issue, sometimes it can be useful to be able to use
byte 0 to poison memory areas, resulting in the same effect as a calloc().
This patch changes the default mem_poison_byte to -1 to disable it so that
all positive values are usable.
HAProxy could already support being passed a file list on the command
line, by passing multiple times "-f" followed by a file name. People
have been complaining that it made it hard to pass file lists from init
scripts.
This patch introduces an end of arguments using the common "--" tag,
after which only file names may appear. These files are then added to
the existing list of other files specified using -f and are loaded in
their declaration order. Thus it becomes possible to do something like
this :
haproxy -sf $(pidof haproxy) -- /etc/haproxy/global.cfg /etc/haproxy/customers/*.cfg
Given that all command line arguments start with a '-' and that
no pid number can start with this character, there's no constraint
to make the pid list the last argument. Let's relax this rule.
Thierry reported that keep-alive still didn't cope well with Lua
services. The reason is that for now applets have to be closed at
the end of a transaction so we want to work in server-close mode,
which isn't noticeable by the client since it still sees keep-alive.
Additionally we want to enable the request body transfer analyser
which will be needed to synchronize with the response analyser to
indicate the end of the transfer.
The release handler used to be called twice for some time and just by
pure luck we never ended up double-freeing the data there. Add a NULL
to ensure this can never happen should a future change permit this
situation again.
This commit adds support for dumping all resolver stats. Specifically
if a command 'show stats resolvers' is issued withOUT a resolver section
id, we dump all known resolver sections. If none are configured, a
message is displayed indicating that.
When using the external-check command option HAProxy was failing to
start with a fatal error "'external-check' cannot handle unexpected
argument". When looking at the code it was looking for an incorrect
argument. Also correcting an Alert message text as spotted by by
PiBa-NL.
When first parameter to getaddrinfo() is not NULL (it is always not NULL
in str2ip()), on Linux AI_PASSIVE value for ai_flags is ignored. On
FreeBSD, when AI_PASSIVE is specified and hostname parameter is not NULL,
getaddrinfo() ignores local address selection policy, always returning
AAAA record. Pass zero ai_flags to behave correctly on FreeBSD, this
change should be no-op for Linux.
This fix should be backported to 1.5 as well, after some observation
period.
Michael Ezzell reported a bug causing haproxy to segfault during startup
when trying to send syslog message from Lua. The function __send_log() can
be called with *p that is NULL and/or when the configuration is not fully
parsed, as is the case with Lua.
This patch fixes this problem by using individual vectors instead of the
pre-generated strings log_htp and log_htp_rfc5424.
Also, this patch fixes a problem causing haproxy to write the wrong pid in
the logs -- the log_htp(_rfc5424) strings were generated at the haproxy
start, but "pid" value would be changed after haproxy is started in
daemon/systemd mode.
Only the main execution function can set the run flag, because it is
the last function before the execution time.
This patch removes the flag set by another function. It will be used
by the new lua timeout counter.
Commit e11cfcd ("MINOR: config: new backend directives:
load-server-state-from-file and server-state-file-name") introduced a bug
which can cause haproxy to crash upon startup by sending user-controlled
data in a format string when emitting a warning. Fix the way the warning
message is built to avoid this.
No backport is needed, this was introduced in 1.6-dev6 only.
Commit e11cfcd ("MINOR: config: new backend directives:
load-server-state-from-file and server-state-file-name") caused these
warnings when building with Clang :
src/server.c:1972:21: warning: comparison of unsigned expression < 0 is always false [-Wtautological-compare]
(srv_uweight < 0) || (srv_uweight > SRV_UWGHT_MAX))
~~~~~~~~~~~ ^ ~
src/server.c:1980:21: warning: comparison of unsigned expression < 0 is always false [-Wtautological-compare]
(srv_iweight < 0) || (srv_iweight > SRV_UWGHT_MAX))
~~~~~~~~~~~ ^ ~
Indeed, srv_iweight and srv_uweight are unsigned. Just drop the offending test.
The conn_sock_drain() call is only there to force the system to ACK
pending data in case of TCP_QUICKACK so that the client doesn't retransmit,
otherwise it leads to a real RST making the feature useless. There's no
point in draining the connection when quick ack cannot be disabled, so
let's move the call inside the ifdef part.
The silent-drop action is supposed to close with a TCP reset that is
either not sent or not too far. But since it's on the client-facing
side, the socket's lingering is enabled by default and the RST only
occurs if some pending unread data remain in the queue when closing.
This causes some clean shutdowns to occur with retransmits, which is
not good at all. Force linger_risk on the socket to flush all data
and destroy the socket.
No backport is needed, this was introduced in 1.6-dev6.
req.ssl_st_ext : integer
Returns 0 if the client didn't send a SessionTicket TLS Extension (RFC5077)
Returns 1 if the client sent SessionTicket TLS Extension
Returns 2 if the client also sent non-zero length TLS SessionTicket
