Since pool_get_from_cache() and pool_put_to_cache() were now only wrappers
to the local cache versions which do all the job, let's merge them together
so that there is no more local-cache specific function.
Now when pool_get_from_local_cache() fails, it automatically falls back
to pool_get_from_shared_cache(), which used to always be done in
pool_get_from_cache(). Thus now the API is simpler as we always allocate
and free from/to the local caches.
Till now it used to call it only if there were not too many objects into
the local cache otherwise would send the latest one directly into the
shared cache. Now it always sends to the local cache and it's up to the
local cache to free its oldest objects. From a cache freshness perspective
it's better this way since we always evict cold objects instead of hot
ones. From an API perspective it's better because it will help make the
shared cache invisible to the public API.
These two functions are now responsible for allocating directly from
the cache and releasing to the cache.
Now the pool_alloc() function simply does this:
if cache enabled
return pool_alloc_from_cache() if no NULL
return pool_alloc_nocache() otherwise
and the pool_free() function does this:
if cache enabled
pool_put_to_cache()
else
pool_free_nocache()
For now this only introduces these two functions without changing anything
else, but the goal is to soon allow to make them implementation-specific.
Continuing the unification of local and shared pools, now the usage of
pools is governed by CONFIG_HAP_POOLS without which allocations and
releases are performed directly from the OS using pool_alloc_nocache()
and pool_free_nocache().
There are two levels of freeing to the OS:
- code that wants to keep the pool's usage counters updated uses
pool_free_area() and handles the counters itself. That's what
pool_put_to_shared_cache() does in the no-global-pools case.
- code that does not want to update the counters because they were
already updated only calls pool_free_area().
Let's extract these calls to establish the symmetry with pool_get_from_os()
and pool_alloc_nocache(), resulting in pool_put_to_os() (which only updates
the allocated counter) and pool_free_nocache() (which also updates the used
counter). This will later allow to simplify the generic code.
Calling pool_free_area() inside a lock in pool_put_to_shared_cache() is
a very bad idea. Fortunately this only happens on the lowest end platforms
which almost never use threads or in very small counts.
This change consists in zeroing the pointer once already released to the
cache in the first test so that the second stage knows if it needs to
pass it to the OS or not. This has slightly reduced the length of the
A part of the code cannot be factored out because it still uses non-atomic
inc/dec for pool->used and pool->allocated as these are located under the
pool's lock. While it can make sense in terms of bus cycles, it does not
make sense in terms of code normalization. Further, some operations were
still performed under a lock that could be totally removed via the use of
atomic ops.
There is still one occurrence in pool_put_to_shared_cache() in the locked
code where pool_free_area() is called under the lock, which must absolutely
be fixed.
Now there's one part dealing with the allocation itself and keeping
counters up to date, and another one on top of it to return such an
allocated pointer to the user and update the use count and stats.
This is in anticipation for being able to group cache-related parts.
The release code is still done at once.
Till now it was limited to objects allocated from the OS which means
it had little use as soon as pools were enabled. Let's move it upper
in the layers so that any code can benefit from fault injection. In
addition this allows to pass a new flag POOL_F_NO_FAIL to disable it
if some callers prefer a no-failure approach.
Now the multi-level cache becomes more visible:
pool_get_from_local_cache()
pool_put_to_local_cache()
pool_get_from_shared_cache()
pool_put_to_shared_cache()
The functions were rightfully called from/to_cache when the thread-local
cache was considered as the only cache, but this is getting terribly
confusing. Let's call them from/to local_cache to make it clear that
it is not related with the shared cache.
As a side note, since pool_evict_from_cache() used not to work for a
particular pool but for all of them at once, it was renamed to
pool_evict_from_local_caches() (plural form).
This is exactly what it is, the entry is retrieved from the shared
cache when it is defined. The implementation that is enabled with
CONFIG_HAP_NO_GLOBAL_POOLS continues to return NULL.
Now that __pool_alloc() only surrounds __pool_get_first() with the lock,
let's move it to the only variant that requires it and remove the ugly
ifdefs from the function. This is safe because nobody else calls this
function.
In __pool_alloc(), historically we used to use factor out the
pool's lock between __pool_get_first() and __pool_refill_alloc(),
resulting in real malloc() or mmap() calls being performed under
the pool lock (for platforms using the locked shared pools).
As this is not needed anymore, let's move the call out of the
lock, it may improve allocation patterns on some platforms. This
also makes __pool_alloc() cleaner as we see a first attempt to
allocate from the local cache, then a second from the shared
cache then a reall allocation.
They were strictly equivalent, let's remerge them and rename them to
pool_alloc_nocache() as it's the call which performs a real allocation
which does not check nor update the cache. The only difference in the
past was the former taking the lock and not the second but now the lock
is not needed anymore at this stage since the pool's list is not touched.
In addition, given that the "avail" argument is no longer used by the
function nor by its callers, let's drop it.
This is a first step towards unifying all the fallback code. Right now
these two functions are the only ones which do not update the needed_avg
rate counter since there's currently no shared pool kept when using them.
But their code is similar to what could be used everywhere except for
this one, so let's make them capable of maintaining usage statistics.
As a side effect the needed field in "show pools" will now be populated.
The mem_should_fail() call enabled by DEBUG_FAIL_ALLOC used to be placed
only in the no-cache version of the allocator. Now we can generalize it
to all modes and remove the exclusive test on CONFIG_HAP_NO_GLOBAL_POOLS.
We're going to make the local pool always present unless pools are
completely disabled. This means that pools are always enabled by
default, regardless of the use of threads. Let's drop this notion
of "local" pools and make it just "pool". The equivalent debug
option becomes DEBUG_NO_POOLS instead of DEBUG_NO_LOCAL_POOLS.
For now this changes nothing except the option and dropping the
dependency on USE_THREAD.
Initially per-thread pool caches were stored into a fixed-size array.
But this was a bit ugly because the last allocated pools were not able
to benefit from the cache at all. As a work around to preserve
performance, a size of 64 cacheable pools was set by default (there
are 51 pools at the moment, excluding any addon and debugging code),
so all in-tree pools were covered, at the expense of higher memory
usage.
In addition an index had to be calculated for each pool, and was used
to acces the pool cache head into that array. The pool index was not
even stored into the pools so it was required to determine it to access
the cache when the pool was already known.
This patch changes this by moving the pool cache head into the pool
head itself. This way it is certain that each pool will have its own
cache. This removes the need for index calculation.
The pool cache head is 32 bytes long so it was aligned to 64B to avoid
false sharing between threads. The extra cost is not huge (~2kB more
per pool than before), and we'll make better use of that space soon.
The pool cache head contains the size, which should probably be removed
since it's already in the pool's head.
In commit fb117e6a8 ("MEDIUM: memory: don't let pool_put_to_cache() free
the objects itself") pool_evict_from_cache() was introduced with no
argument, yet the only call place passes it the pool, the pointer and
the index number!
Let's remove these as they even let the reader think that the function
does something specific to the current pool while it's not the case.
This patch renames all existing uri-normalizers into a more consistent naming
scheme:
1. The part of the URI that is being touched.
2. The modification being performed as an explicit verb.
This normalizer merges `../` path segments with the predecing segment, removing
both the preceding segment and the `../`.
Empty segments do not receive special treatment. The `merge-slashes` normalizer
should be executed first.
See GitHub Issue #714.
When a thread ends its harmeless period, we must only consider running
threads when testing threads_want_rdv_mask mask. To do so, we reintroduce
all_threads_mask mask in the bitwise operation (It was removed to fix a
deadlock).
Note that for now it is useless because there is no way to stop threads or
to have threads reserved for another task. But it is safer this way to avoid
bugs in the future.
A previous patch was pushed to fix a deadlock when an isolated thread ends
its harmless period (a9a9e9aac ["BUG/MEDIUM: thread: Fix a deadlock if an
isolated thread is marked as harmless"]). But, unfortunately, the fix is
incomplete. The same must be done in the outer loop, in
thread_harmless_end() function. The current thread must be ignored when
threads_want_rdv_mask mask is tested.
This patch must also be backported as far as 2.0.
This library is required for the subsequent patch which adds
the JSON query possibility.
It is necessary to change the include statement in "src/mjson.c"
because the imported includes in haproxy are in "include/import"
orig: #include "mjson.h"
new: #include <import/mjson.h>
ub64dec and ub64enc are the base64url equivalent of b64dec and base64
converters. base64url encoding is the "URL and Filename Safe Alphabet"
variant of base64 encoding. It is also used in in JWT (JSON Web Token)
standard.
RFC1421 mention in base64.c file is deprecated so it was replaced with
RFC4648 to which existing converters, base64/b64dec, still apply.
Example:
HAProxy:
http-request return content-type text/plain lf-string %[req.hdr(Authorization),word(2,.),ub64dec]
Client:
Token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VyIjoiZm9vIiwia2V5IjoiY2hhZTZBaFhhaTZlIn0.5VsVj7mdxVvo1wP5c0dVHnr-S_khnIdFkThqvwukmdg
$ curl -H "Authorization: Bearer ${TOKEN}" http://haproxy.local
{"user":"foo","key":"chae6AhXai6e"}
Allocation error are now handled in bind_conf_alloc() functions. Thus
callers, when not already done, are also updated to catch NULL return value.
This patch may be backported (at least partially) to all stable
versions. However, it only fix errors durung configuration parsing. Thus it
is not mandatory.
Add the trace support for the checks. Only tcp-check based health-checks are
supported, including the agent-check.
In traces, the first argument is always a check object. So it is easy to get
all info related to the check. The tcp-check ruleset, the conn-stream and
the connection, the server state...
Olivier spotted that I messed up during a rebase of commit 92c059c2a
("MINOR: atomic: implement native BTS/BTR for x86"), losing the x86
version of the BTS/BTR and leaving the generic version for it instead
of having this block in the #else. Since this variant is not used for
now it was easy to overlook it. Let's re-implement it here.
The time initialization was made a bit complex because we rely on a
dummy negative argument to reset all fields, leaving no distinction
between process-level initialization and thread-level initialization.
This patch changes this by introducing two functions, one for the
process and the second one for the threads. This removes ambigous
test and makes sure that the relevant fields are always initialized
exactly once. This also offers a better solution to the bug fixed in
commit b48e7c001 ("BUG/MEDIUM: time: make sure to always initialize
the global tick") as there is no more special values for global_now_ms.
It's simple enough to be backported if any other time-related issues
are encountered in stable versions in the future.
It was only used by freq_ctr and is not used anymore. In addition the
local curr_sec_ms was removed, as well as the equivalent extern
definitions which did not exist anymore either.
update_freq_ctr_period() was still not very clean and didn't wait for
the rotation lock to be dropped before trying again, thus maintaining
the contention at a high level. In addition, the rotation update was
made in three steps, which are not very efficient in terms of bus
cycles.
Here the wait loop was reworked so that the fast path remains short
and that the contended path waits for the lock to be dropped before
attempting another write, but it only waits a relax cycle before
attempting a read. The rotation block was simplified to remove a
test that was already validated by the first loop, and so that the
retrieval of the current period, its reset and its increment are all
performed in a single atomic op and the store to the previous period
is performed immediately after.
All this results in significantly smaller code for the inline function
(~1kB total) and a shorter critical path.
When counters are rotated, there is contention between the threads which
can slow down the operation of the thread performing the rotation. Let's
apply a cpu_relax there to let the first thread finish faster.
It remains cumbersome to preserve two versions of the freq counters and
two different internal clocks just for this. In addition, the savings
from using two different mechanisms are not that important as the only
saving is a divide that is replaced by a multiply, but now thanks to
the freq_ctr_total() unificaiton the code could also be simplified to
optimize it in case of constants.
This patch turns all non-period freq_ctr functions to static inlines
which call the period-based ones with a period of 1 second. A direct
benefit is that a single internal clock is now needed for any counter
and that they now all rely on ticks.
These 1-second counters are essentially used to report request rates
and to enforce a connection rate limitation in listeners. It was
verified that these continue to work like before.
Both structures are identical except the name of the field starting
the period and its description. Let's call them all freq_ctr and the
period's start "curr_tick" which is generic.
This is only a temporary change and fields are expected to remain
the same with no code change (verified).
There was still no function to compute a wait time for periods, let's
implement it on top of freq_ctr_total() as we'll soon need it for the
per-second one. The divide here is applied on the frequency so that it
will be replaced with a reciprocal multiply when constant.
This one is the easiest to implement, it just requires a call and a
divide of the result. Anti-flapping correction for low-rates was
preserved.
Now calls using a constant period will be able to use a reciprocal
multiply for the period instead of a divide.
Most of the functions designed to read a counter over a period go through
the same complex loop and only differ in the way they use the returned
values, so it was worth implementing all this into freq_ctr_total() which
returns the total number of events over a period so that the caller can
finish its operation using a divide or a remaining time calculation. As
a special case, read_freq_ctr_period() doesn't take pending events but
requires to enable an anti-flapping correction at very low frequencies.
Thus the function implements it when pend<0.
Thanks to this function it will be possible to reimplement the other ones
as inline and merge the per-second ones with the arbitrary period ones
without always adding the cost of a 64 bit divide.
Some variables are mostly read (mostly pointers) but they tend to be
merged with other ones in the same cache line, slowing their access down
in multi-thread setups. This patch declares an empty, aligned variable
in a section called "read_mostly". This will force a cache-line alignment
on this section so that any variable declared in it will be certain to
avoid false sharing with other ones. The section will be eliminated at
link time if not used.
A __read_mostly attribute was added to compiler.h to ease use of this
section.
HA_SECTION() is used as an attribute to force a section name. This is
required because OSX prepends "__DATA, " in front of the declaration.
HA_SECTION_START() and HA_SECTION_STOP() are used as post-attribute on
variable declaration to designate the section start/end (needed only on
OSX, empty on others).
For platforms with an obsolete linker, all macros are left empty. It would
possibly still work on some of them but this will not be needed anyway.
Due to length restrictions on OSX the initcall sections are called "i_"
there while they're called "init_" on other OSes. However the start and
end of sections are still called "__start_init_" and "__stop_init_",
which forces to have distinct code between the OSes. Let's switch everyone
to "i_" and rename the symbols accordingly.
The trace() function is convenient to avoid calling trace() when traces
are not enabled, but there starts to be some callers which place complex
expressions in their trace calls, which results in all of them to be
evaluated before being passed as arguments to the trace() function. This
needlessly wastes precious CPU cycles.
Let's change the function for a macro, so that the arguments are now only
evaluated when the surce has traces enabled. However having a generic
macro being called "trace()" can easily cause conflicts with innocent
code so we rename it "_trace".
Just doing this has resulted in a 2.5% increase of the HTTP/1 request rate.
Interestingly, all arrays used to declare patterns were read-write while
only hard-coded. Let's mark them const so that they move from data to
rodata and don't risk to experience false sharing.
This argument is not being used inside the function (and the functions
themselves are unused as well) and not documented. Its purpose is not clear.
Just remove it.
With latest commit f50906519 ("MEDIUM: fd: merge fdtab[].ev and state
for FD_EV_* and FD_POLL_* into state") one occurrence of a pair of
chars was missed in fd_stop_both(), resulting in the operation to
fail if the upper flags were set. Interestingly it managed to fail
2 tests in all setups in the CI while all used to work fine on my
local machines. Probably that the reason is that the chars had enough
room above them for the CAS to fail then refill "old" overwriting the
upper parts of the stack, and that thanks to this the subsequent tests
worked. With ASAN being used on lots of tests, it very likely caught
it but used to only report failed tests with no more info.
No backport is needed, as this was never released nor backported.
The current BTS/BTR operations on x86 are ugly because they rely on a
CAS, so they may be unfair and take time to converge. Fortunately,
where they are currently used (mostly FDs) the contention is expected
to be rare (mostly listeners). But this also limits their use to such
few low-load cases.
On x86 there is a set of BTS/BTR instructions which help for this,
but before the FD's state migrated to 32 bits there was little use of
them since they do not exist in 8 bits.
Now at least it makes sense to use them, at the very least in order
to significantly reduce the code size (one BTS instead of a CMPXCHG
loop). The implementation relies on modern gcc's ability to return
condition flags and limit code inflation and register spilling. The
fall back is retained on the old implementation for all other situations
(inappropriate target size or non-capable compiler). The code shrank
by 1.6 kB on the fast path.
As expected, for now on up to 4 threads there is no measurable difference
of performance.
Probably due to the result of an old copy-paste, HA_ATOMIC_BTS/BTR were
still implemented using the __sync_* builtins instead of the more
modern __atomic_* which allow to specify the memory model. Let's update
this to use the newer there and also implement the relaxed variants
(which are not used for now).
This patch replaces roughly all occurrences of an HA_ATOMIC_ADD(&foo, 1)
or HA_ATOMIC_SUB(&foo, 1) with the equivalent HA_ATOMIC_INC(&foo) and
HA_ATOMIC_DEC(&foo) respectively. These are 507 changes over 45 files.
Most ADD/SUB callers use them for a single unit (e.g. refcounts) and
it's a pain to always pass ",1". Let's add them to simplify the API.
However we currently don't add any return value. If needed in the future
better report zero/non-zero than a real value for the sake of efficiency
at the instruction level.
The fetch_and_xxx variant is often missing for add/sub/and/or. In fact
it was only provided for ADD under the name XADD which corresponds to
the x86 instruction name. But for destructive operations like AND and
OR it's missing even more as it's not possible to know the value before
modifying it.
This patch explicitly adds HA_ATOMIC_FETCH_{OR,AND,ADD,SUB} which
cover these standard operations, and renames XADD to FETCH_ADD (there
were only 6 call places).
In the future, backport of fixes involving such operations could simply
remap FETCH_ADD(x) to XADD(x), FETCH_SUB(x) to XADD(-x), and for the
OR/AND if needed, these could possibly be done using BTS/BTR.
It's worth noting that xchg could have been renamed to fetch_and_store()
but xchg already has well understood semantics and it wasn't needed to
go further.
In order to make sure these ones will not be used anymore in an expression,
let's make them always void. New callers will now be forced to use the
explicit _FETCH variant if required.
Currently our atomic ops return a value but it's never known whether
the fetch is done before or after the operation, which causes some
confusion each time the value is desired. Let's create an explicit
variant of these operations suffixed with _FETCH to explicitly mention
that the fetch occurs after the operation, and make use of it at the
few call places.
Gcc 10.2 implements outline atomics on aarch64. The replace all inline
atomic ops with a function call that checks if the machine supports LSE
atomics. This comes with a small cost but allows modern machines to scale
much better than with the old LL/SC ones even when built for full 8.0
compatibility.
This patch enables the use of the __atomic_compare_exchange() builtin
for the double-word CAS when detected as available instead of using the
hand-written LL/SC version. The extra cost is negligible because we do
very few DWCAS operations (essentially FD migrations and shared pools)
so the cost is low but under high contention it can still be beneficial.
As expected no performance difference was measured in either direction
on 4-core machines with this change.
This could be backported to 2.3 if it was shown that FD migrations were
representing a significant source of contention, but for now it does
not appear to be needed.
There is a function called fd_write_frag_line() that's essentially used
by loggers and that is used to write an atomic message line over a file
descriptor using writev(). However a lock is required around the writev()
call to prevent messages from multiple threads from being interleaved.
Till now a SPIN_TRYLOCK was used on a dedicated lock that was common to
all FDs. This is quite not pretty as if there are multiple output pipes
to collect logs, there will be quite some contention. Now that there
are empty flags left in the FD state and that we can finally use atomic
ops on them, let's add a flag to indicate the FD is locked for exclusive
access by a syscall. At least the locking will now be on an FD basis and
not the whole process, so we can remove the log_lock.
No need to keep this flag apart any more, let's merge it into the global
state. The bit was not cleared in fd_insert() because the only user is
the function used to create and atomically send a log message to a pipe
FD, which never registers the fd. Here we clear it nevertheless for the
sake of clarity.
Note that with an extra cleaning pass we could have a bit number
here and simply use a BTS to test and set it.
No need to keep this flag apart any more, let's merge it into the global
state. The CLI's output state was extended to 6 digits and the linger/cloned
flags moved inside the parenthesis.
For a long time we've had fdtab[].ev and fdtab[].state which contain two
arbitrary sets of information, one is mostly the configuration plus some
shutdown reports and the other one is the latest polling status report
which also contains some sticky error and shutdown reports.
These ones used to be stored into distinct chars, complicating certain
operations and not even allowing to clearly see concurrent accesses (e.g.
fd_delete_orphan() would set the state to zero while fd_insert() would
only set the event to zero).
This patch creates a single uint with the two sets in it, still delimited
at the byte level for better readability. The original FD_EV_* values
remained at the lowest bit levels as they are also known by their bit
value. The next step will consist in merging the remaining bits into it.
The whole bits are now cleared both in fd_insert() and _fd_delete_orphan()
because after a complete check, it is certain that in both cases these
functions are the only ones touching these areas. Indeed, for
_fd_delete_orphan(), the thread_mask has already been zeroed before a
poller can call fd_update_event() which would touch the state, so it
is certain that _fd_delete_orphan() is alone. Regarding fd_insert(),
only one thread will get an FD at any moment, and it as this FD has
already been released by _fd_delete_orphan() by definition it is certain
that previous users have definitely stopped touching it.
Strictly speaking there's no need for clearing the state again in
fd_insert() but it's cheap and will remove some doubts during some
troubleshooting sessions.
In preparation of merging FD_POLL* and FD_EV*, this only changes the
value of FD_POLL_* to use bits 8-15 (the second byte). The size of the
field has been temporarily extended to 32 bits already, as well as
the temporary variables that carry the new composite value inside
fd_update_events(). The resulting fdtab entry becomes temporarily
unaligned. All places making access to .ev or FD_POLL_* were carefully
inspected to make sure they were safe regarding this change. Only one
temporary update was needed for the "show fd" code. The code was only
slightly inflated at this step.
The former was not used and the second was used only as a positive mask
of the flags to keep instead of having the flags that are updated. Both
were removed in favor of a new FD_POLL_UPDT_MASK that only mentions the
updated flags. This will ease merging of state and ev later.
This patch registers the parsed file and the line where a log server
is declared to make those information available in configuration
post check.
Those new informations were added on error messages probed resolving
ring names on post configuration check.
Define MODE_DIAG which is used to run haproxy in diagnostic mode. This
mode is used to output extra warnings about possible configuration
blunder or sub-optimal usage. It can be activated with argument '-dD'.
A new output function ha_diag_warning is implemented reserved for
diagnostic output. It serves to standardize the format of diagnostic
messages.
A macro HA_DIAG_WARN_COND is also available to automatically check if
diagnostic mode is on before executing the diagnostic check.
Historically, an option was added to wait for the request payload (option
http-buffer-request). This option has 2 drawbacks. First, it is an ON/OFF
option for the whole proxy. It cannot be enabled on demand depending on the
message. Then, as its name suggests, it only works on the request side. The
only option to wait for the response payload was to write a dedicated
filter. While it is an acceptable solution for complex applications, it is a
bit overkill to simply match strings in the body.
To make everyone happy, this patch adds a dedicated HTTP action to wait for
the message payload, for the request or the response depending it is used in
an http-request or an http-response ruleset. The time to wait is
configurable and, optionally, the minimum payload size to have before stop
to wait.
Both the http action and the old http analyzer rely on the same internal
function.
L6 sample fetches are now ignored when called from an HTTP proxy. Thus, a
warning is emitted during the startup if such usage is detected. It is true
for most ACLs and for log-format strings. Unfortunately, it is a bit painful
to do so for sample expressions.
This patch relies on the commit "MINOR: action: Use a generic function to
check validity of an action rule list".
The check_action_rules() function is now used to check the validity of an
action rule list. It is used from check_config_validity() function to check
L5/6/7 rulesets.
It is now possible to perform HTTP upgrades on a TCP stream from the
frontend side. To do so, a tcp-request content rule must be defined with the
switch-mode action, specifying the mode (for now, only http is supported)
and optionnaly the proto (h1 or h2).
This way it could be possible to set HTTP directives on a TCP frontend which
will only be evaluated if an upgrade is performed. This new way to perform
HTTP upgrades should replace progressively the old way, consisting to route
the request to an HTTP backend. And it should be also a good start to remove
all HTTP processing from tcp-request content rules.
This action is terminal, it stops the ruleset evaluation. It is only
available on proxy with the frontend capability.
The configuration manual has been updated accordingly.
The code responsible to perform an HTTP upgrade from a TCP stream is moved
in a dedicated function, stream_set_http_mode().
The stream_set_backend() function is slightly updated, especially to
correctly set the request analysers.
Now allocation and initialization of HTTP transactions are performed in a
unique function. Historically, there were two functions because the same TXN
was reset for K/A connections in the legacy HTTP mode. Now, in HTX, K/A
connections are handled at the mux level. A new stream, and thus a new TXN,
is created for each request. In addition, the function responsible to end
the TXN is now also reponsible to release it.
So, now, http_create_txn() and http_destroy_txn() must be used to create and
destroy an HTTP transaction.
MX_FL_NO_UPG flag may now be set on a multiplexer to explicitly disable
upgrades from this mux. For now, it is set on the FCGI multiplexer because
it is not supported and there is no upgrade on backend-only multiplexers. It
is also set on the H2 multiplexer because it is clearly not supported.
Historically we've used SOL_IP/SOL_IPV6/SOL_TCP everywhere as the socket
level value in getsockopt() and setsockopt() but as we've seen over time
it regularly broke the build and required to have them defined to their
IPPROTO_* equivalent. The Linux ip(7) man page says:
Using the SOL_IP socket options level isn't portable; BSD-based
stacks use the IPPROTO_IP level.
And it indeed looks like a pure linuxism inherited from old examples and
documentation. strace also reports SOL_* instead of IPPROTO_*, which does
not help... A check to linux/in.h shows they have the same values. Only
SOL_SOCKET and other non-IP values make sense since there is no IPPROTO
equivalent.
Let's get rid of this annoying confusion by removing all redefinitions of
SOL_IP/IPV6/TCP and using IPPROTO_* instead, just like any other operating
system. This also removes duplicated tests for the same value.
Note that this should not result in exposing syscalls to other OSes
as the only ones that were still conditionned to SOL_IPV6 were for
IPV6_UNICAST_HOPS which already had an IPPROTO_IPV6 equivalent, and
IPV6_TRANSPARENT which is Linux-specific.
Very often we use "int" where negative numbers are not needed (and can
further cause trouble) just because it's painful to type "unsigned int"
or "unsigned", or ugly to use in function arguments. Similarly sometimes
chars would absolutely need to be signed but nobody types "signed char".
Let's add a few aliases for such types and make them part of the standard
internal API so that over time we can get used to them and get rid of
horrible definitions. A comment also reminds some commonly available
types and their properties regarding other types.
In order to process samples from the command line interface we'll need
rules as well, and these rules will have to be marked as coming from
the CLI parser. This new origin is used for this.
In order to prepare for supporting calling sample expressions from the
CLI, let's create a new CLI_PARSER parsing context. This one supports
constants and internal samples only.
In order to process samples from the config file we'll need rules as
well, and these rules will have to be marked as coming from the
config parser. This new origin is used for this.
We'd sometimes like to be able to process samples while parsing
the configuration based on purely internal thing but that's not
possible right now. Let's add a new CFG_PARSER context for samples
which only permits constant samples (i.e. those which do not change
in the process' life and which are stable during config parsing).
This level indicates that everything it constant in the expression during
the whole process' life and that it may safely be used at config parsing
time.
For now smp_resolve_args() complains on stderr via ha_alert(), but if we
want to make it a bit more dynamic, we need it to return errors in an
allocated message. Let's pass it an error pointer and have it fill it.
On return we indent the output if it contains more than one line.
Define a new cap PR_CAP_LUA. It can be used to allocate the internal
proxy for lua Socket class. This cap overrides default settings for
preferable values in the lua context.
Move all liberation code related to a proxy in a dedicated function
free_proxy in proxy.c. For now, this function is only called in
haproxy.c. In the future, it will be used to free the lua proxy.
This helps to clean up haproxy.c.
Create a new function parse_new_proxy specifically designed to allocate
a new proxy from the configuration file and copy settings from the
default proxy.
The function alloc_new_proxy is reduced to a minimal allocation. It is
used for default proxy allocation and could also be used for internal
proxies such as the lua Socket proxy.
Move deinit_acl_cond and deinit_act_rules from haproxy.c respectively in
acl.c and action.c. The name of the functions has been slightly altered,
replacing the prefix deinit_* by free_* to reflect their purpose more
clearly.
This change has been made in preparation to the implementation of a free
proxy function. As a side-effect, it helps to clean up haproxy.c.
Create a new module init which contains code related to REGISTER_*
macros for initcalls. init.h is included in api.h to make init code
available to all modules.
It's a step to clean up a bit haproxy.c/global.h.
The default SSL_CTX used by a specific frontend is the one of the first
ckch instance created for this frontend. If this instance has SNIs, then
the SSL context is linked to the instance through the list of SNIs
contained in it. If the instance does not have any SNIs though, then the
SSL_CTX is only referenced by the bind_conf structure and the instance
itself has no link to it.
When trying to update a certificate used by the default instance through
a cli command, a new version of the default instance was rebuilt but the
default SSL context referenced in the bind_conf structure would not be
changed, resulting in a buggy behavior in which depending on the SNI
used by the client, he could either use the new version of the updated
certificate or the original one.
This patch adds a reference to the default SSL context in the default
ckch instances so that it can be hot swapped during a certificate
update.
This should fix GitHub issue #1143.
It can be backported as far as 2.2.
Christopher discovered an issue mostly affecting 2.2 and to a less extent
2.3 and above, which is that it's possible to deadlock a soft-stop when
several threads are using a same listener:
thread1 thread2
unbind_listener() fd_set_running()
lock(listener) listener_accept()
fd_delete() lock(listener)
while (running_mask); -----> deadlock
unlock(listener)
This simple case disappeared from 2.3 due to the removal of some locked
operations at the end of listener_accept() on the regular path, but the
architectural problem is still here and caused by a lock inversion built
around the loop on running_mask in fd_clr_running_excl(), because there
are situations where the caller of fd_delete() may hold a lock that is
preventing other threads from dropping their bit in running_mask.
The real need here is to make sure the last user deletes the FD. We have
all we need to know the last one, it's the one calling fd_clr_running()
last, or entering fd_delete() last, both of which can be summed up as
the last one calling fd_clr_running() if fd_delete() calls fd_clr_running()
at the end. And we can prevent new threads from appearing in running_mask
by removing their bits in thread_mask.
So what this patch does is that it sets the running_mask for the thread
in fd_delete(), clears the thread_mask, thus marking the FD as orphaned,
then clears the running mask again, and completes the deletion if it was
the last one. If it was not, another thread will pass through fd_clr_running
and will complete the deletion of the FD.
The bug is easily reproducible in 2.2 under high connection rates during
soft close. When the old process stops its listener, occasionally two
threads will deadlock and the old process will then be killed by the
watchdog. It's strongly believed that similar situations do exist in 2.3
and 2.4 (e.g. if the removal attempt happens during resume_listener()
called from listener_accept()) but if so, they should be much harder to
trigger.
This should be backported to 2.2 as the issue appeared with the FD
migration. It requires previous patches "fd: make fd_clr_running() return
the remaining running mask" and "MINOR: fd: remove the unneeded running
bit from fd_insert()".
Notes for backport: in 2.2, the fd_dodelete() function requires an extra
argument "do_close" indicating whether we want to remove and close the FD
(fd_delete) or just delete it (fd_remove). While this information is not
conveyed along the chain, we know that late calls always imply do_close=1
become do_close=0 exclusively results from fd_remove() which is only used
by the config parser and the master, both of which are single-threaded,
hence are always the last ones in the running_mask. Thus it is safe to
assume that a postponed FD deletion always implies do_close=1.
Thanks to Olivier for his help in designing this optimal solution.
There's no point taking the running bit in fd_insert() since by
definition there will never be more than one thread inserting the FD,
and that fd_insert() may only be done after the fd was allocated by
the system, indicating the end of use by any other thread.
This will need to be backported to 2.2 to fix an issue.
We'll need to know that a thread is the last one to use an fd, so let's
make fd_clr_running() return the remaining bits after removal. Note that
in practice we're only interested in knowing if it's zero but the compiler
doesn't make use of the clags after the AND and emits a CMPXCHG anyway :-/
This will need to be backported to 2.2 to fix an issue.
The commit reverts following commits:
* 83926a04 BUG/MEDIUM: debug/lua: Don't dump the lua stack if not dumpable
* a61789a1 MEDIUM: lua: Use a per-thread counter to track some non-reentrant parts of lua
Instead of relying on a Lua function to print the lua traceback into the
debugger, we are now using our own internal function (hlua_traceback()).
This one does not allocate memory and use a chunk instead. This avoids any
issue with a possible deadlock in the memory allocator because the thread
processing was interrupted during a memory allocation.
This patch relies on the commit "BUG/MEDIUM: debug/lua: Use internal hlua
function to dump the lua traceback". Both must be backported wherever the
patches above are backported, thus as far as 2.0
The separator string is now configurable, passing it as parameter when the
function is called. In addition, the message have been slightly changed to
be a bit more readable.
If an unknown CA file was first mentioned in an "add ssl crt-list" CLI
command, it would result in a call to X509_STORE_load_locations which
performs a disk access which is forbidden during runtime. The same would
happen if a "ca-verify-file" or "crl-file" was specified. This was due
to the fact that the crt-list file parsing and the crt-list related CLI
commands parsing use the same functions.
The patch simply adds a new parameter to all the ssl_bind parsing
functions so that they know if the call is made during init or by the
CLI, and the ssl_store_load_locations function can then reject any new
cafile_entry creation coming from a CLI call.
It can be backported as far as 2.2.
str2sa_range function options PA_O_DGRAM and PA_O_STREAM are used to
define the supported address types but also to set the default type
if it is not explicit. If the used address support both STREAM and DGRAM,
the default was always set to STREAM.
This patch introduce a new option PA_O_DEFAULT_DGRAM to force the
default to DGRAM type if it is not explicit in the address field
and both STREAM and DGRAM are supported. If only DGRAM or only STREAM
is supported, it continues to be considered as the default.
In commit a1ecbca0a ("BUG/MINOR: freq_ctr/threads: make use of the last
updated global time"), for period-based counters, the millisecond part
of the global_now variable was used as the date for the new period. But
it's wrong, it only works with sub-second periods as it wraps every
second, and for other periods the counters never rotate anymore.
Let's make use of the newly introduced global_now_ms variable instead,
which contains the global monotonic time expressed in milliseconds.
This patch needs to be backported wherever the patch above is backported.
It depends on previous commit "MINOR: time: also provide a global,
monotonic global_now_ms timer".
The period-based freq counters need the global date in milliseconds,
so better calculate it and expose it rather than letting all call
places incorrectly retrieve it.
Here what we do is that we maintain a new globally monotonic timer,
global_now_ms, which ought to be very close to the global_now one,
but maintains the monotonic approach of now_ms between all threads
in that global_now_ms is always ahead of any now_ms.
This patch is made simple to ease backporting (it will be needed for
a subsequent fix), but it also opens the way to some simplifications
on the time handling: instead of computing the local time and trying
to force it to the global one, we should soon be able to proceed in
the opposite way, that is computing the new global time an making the
local one just the latest snapshot of it. This will bring the benefit
of making sure that the global time is always ahead of the local one.
The pool_alloc_dirty() function was renamed to __pool_alloc() and now
takes a set of flags indicating whether poisonning is permitted or not
and whether zeroing the area is needed or not. The pool_alloc() function
is now just a wrapper calling __pool_alloc(pool, 0).
This one used to maintain a shortcut in the pools allocation path that
was only justified by b_alloc_fast() which was not used! Let's get rid
of it as well so that the allocator becomes a bit more straight forward.
It is never used anymore since 1.7 where it was used by b_alloc_margin()
then replaced by direct calls to the pools function, and it maintains a
dependency on the exposed pools functions. It's time to get rid of it,
as it's not even certain it still works.
The channel's buffer allocator, channel_alloc_buffer(), was still relying
on the principle of a margin for the request and not for the response.
But this margin stopped working around 1.7 with the introduction of the
content filters such as SPOE, and was completely anihilated with the
local pools that came with threads. Let's simplify this and just use
b_alloc().
Right now there is a discrepancy beteween b_alloc() and b_allow_margin():
the former forcefully overwrites the target pointer while the latter tests
it and returns it as-is if already allocated.
As a matter of fact, all callers of b_alloc() either preliminary test the
buffer, or assume it's already null.
Let's remove this pain and make the function test the buffer's allocation
before doing it again, and match call places' expectations.
Some parts of the Lua are non-reentrant. We must be sure to carefully track
these parts to not dump the lua stack when it is interrupted inside such
parts. For now, we only identified the custom lua allocator. If the thread
is interrupted during the memory allocation, we must not try to print the
lua stack wich also allocate memory. Indeed, realloc() is not
async-signal-safe.
In this patch we introduce a thread-local counter. It is incremented before
entering in a non-reentrant part and decremented when exiting. It is only
performed in hlua_alloc() for now.
Now that connections aren't being reused when they failed, remove the
reset() method. It was unimplemented anywhere, except for H1 where it did
nothing, anyway.
Introduce a new XPRT method, start(). The init() method will now only
initialize whatever is needed for the XPRT to run, but any action the XPRT
has to do before being ready, such as handshakes, will be done in the new
start() method. That way, we will be sure the full stack of xprt will be
initialized before attempting to do anything.
The init() call is also moved to conn_prepare(). There's no longer any reason
to wait for the ctrl to be ready, any action will be deferred until start(),
anyway. This means conn_xprt_init() is no longer needed.
Remove static qualifier on stats_allocate_proxy_counters_internal. This
function will be used to allocate extra counters at runtime for dynamic
servers.
Prepare the server parsing API to support dynamic servers.
- define a new parsing flag to be used for dynamic servers
- each keyword contains a new field dynamic_ok to indicate if it can be
used for a dynamic server. For now, no keyword are supported.
- do not copy settings from the default server for a new dynamic server.
- a dynamic server is created in a maintenance mode and requires an
explicit 'enable server' command.
- a new server flag named SRV_F_DYNAMIC is created. This flag is set for
all servers created at runtime. It might be useful later, for example
to know if a server can be purged.
Modify the API of parse_server function. Use flags to describe the type
of the parsed server instead of discrete arguments. These flags can be
used to specify if a server/default-server/server-template is parsed.
Additional parameters are also specified (parsing of the address
required, resolve of a name must be done immediately).
It is now unneeded to use strcmp on args[0] in parse_server. Also, the
calls to parse_server are more explicit thanks to the flags.
The freq counters were using the thread's own time as the start of the
current period. The problem is that in case of contention, it was
occasionally possible to perform non-monotonic updates on the edge of
the next second, because if the upfront thread updates a counter first,
it causes a rotation, then the second thread loses the race from its
older time, and tries again, and detects a different time again, but
in the past so it only updates the counter, then a third thread on the
new date would detect a change again, thus provoking a rotation again.
The effect was triple:
- rare loss of stored values during certain transitions from one
period to the next one, causing counters to report 0
- half of the threads forced to go through the slow path every second
- difficult convergence when using many threads where the CAS can fail
a lot and we can observe N(N-1) attempts for N threads to complete
This patch fixes this issue in two ways:
- first, it now makes use og the monotonic global_now value which also
happens to be volatile and to carry the latest known time; this way
time will never jump backwards anymore and only the first thread
updates it on transition, the other ones do not need to.
- second, re-read the time in the loop after each failure, because
if the date changed in the counter, it means that one thread knows
a more recent one and we need to update. In this case if it matches
the new current second, the fast path is usable.
This patch relies on previous patch "MINOR: time: export the global_now
variable" and must be backported as far as 1.8.
This is the process-wide monotonic time that is used to update each
thread's own time. It may be required at a few places where a strictly
monotonic clock is required such as freq_ctr. It will be have to be
backported as a dependency of a forthcoming fix.
Some are not always easy to spot with "chk" vs "check" or hyphens at
some places and not at others. Now entering "option http-close" properly
suggests "httpclose" and "option tcp-chk" suggests "tcp-check". There's
no need to consider the proxy's capabilities, what matters is to figure
what related word the user tried to spell, and there are not that many
options anyway.
The distance between two words can be high due to a sub-word being missing
and in this case it happens that other totally unrealted words are proposed
because their average score looks lower thanks to being shorter. Here we're
introducing the notion of presence of each character so that word sequences
that contain existing sub-words are favored against the shorter ones having
nothing in common. In addition we do not distinguish being/end from a
regular delimitor anymore. That made it harder to spot inverted words.
When tasklets were derived from tasks, there was no immediate need for
the scheduler to know their status after execution, and in a spirit of
simplicity they just started to always return NULL. The problem is that
it simply prevents the scheduler from 1) accounting their execution time,
and 2) keeping track of their current execution status. Indeed, a remote
wake-up could very well end up manipulating a tasklet that's currently
being executed. And this is the reason why those handlers have to take
the idle lock before checking their context.
In 2.5 we'll take care of making tasklets and tasks work more similarly,
but trouble is to be expected if we continue to propagate the trend of
returning NULL everywhere, especially if some fixes relying on a stricter
model later need to be backported. For this reason this patch updates all
known tasklet handlers to make them return NULL only when the tasklet was
freed. It has no effect for now and isn't even guaranteed to always be
100% safe but it puts the code into the right direction for this.
There were still a very small list of functions, variables and fields
called "stats_" while they were really purely CLI-centric. There's the
frontend called "stats_fe" in the global section, which instantiates a
"cli_applet" called "<CLI>" so it was renamed "cli_fe".
The "alloc_stats_fe" function cas renamed to "cli_alloc_fe" which also
better matches the naming convention of all cli-specific functions.
Finally the "stats_permission_denied_msg" used to return an error on
the CLI was renamed "cli_permission_denied_msg".
Now there's no more "stats_something" that designates the CLI.
This is the number of args accepted on a command received on the CLI,
is has long been totally independent of stats and should not carry
this misleading "stats" name anymore.
Instead of making a new one from scratch, let's support not wiping the
existing fingerprint and updating it, and to do the same char by char.
The word-by-word one will still result in multiple beginnings and ends,
but that will accurately translate word boundaries. The char-based one
has more flexibility and requires that the caller maintains the previous
char to indicate the transition, which also allows to insert delimiters
for example.
Entering "show tls" would still emit 35 entries. By measuring the distance
between all unknown words and the candidates, we can sort them and pick the
10 most likely candidates. This works reasonably well, as now "show tls"
only proposes "show tls-keys", "show threads", "show pools" and "show tasks".
If the distance is still too high or if a word is missing, the whole
prefix list continues to be dumped, thus "show" alone will still report
the entire list of commands beginning with "show".
It's still impossible to skip a word, for example "show conn" will not
propose "show servers conn" because the distance is calculated for each
word individually. Some changes to the distance calculation to support
updating an existing map could easily address this. But this is already
a great improvement.
It was mentioned that ACCESS_MASTER_ONLY as for workers only instead of
master-only. And it wasn't clear that all ACCESS_* would belong to the
same thing.
The last time when an item was seen in a resolver responses is now stored in
milliseconds instead of seconds. This avoid some corner-cases at the
edges. This also simplifies time comparisons.
Another way to say it: "Safely unlink requester from a requester callbacks".
Requester callbacks must never try to unlink a requester from a resolution, for
the current requester or another one. First, these callback functions are called
in a loop on a request list, not necessarily safe. Thus unlink resolution at
this place, may be unsafe. And it is useless to try to make these loops safe
because, all this stuff is placed in a loop on a resolution list. Unlink a
requester may lead to release a resolution if it is the last requester.
However, the unkink is necessary because we cannot reset the server state
(hostname and IP) with some pending DNS resolution on it. So, to workaround
this issue, we introduce the "safe" unlink. It is only performed from a
requester callback. In this case, the unlink function never releases the
resolution, it only reset it if necessary. And when a resolution is found
with an empty requester list, it is released.
This patch depends on the following commits :
* MINOR: resolvers: Purge answer items when a SRV resolution triggers an error
* MINOR: resolvers: Use a function to remove answers attached to a resolution
* MINOR: resolvers: Directly call srvrq_update_srv_state() when possible
* MINOR: resolvers: Add function to change the srv status based on SRV resolution
All the series must be backported as far as 2.2. It fixes a regression
introduced by the commit b4badf720 ("BUG/MINOR: resolvers: new callback to
properly handle SRV record errors").
don't release resolution from requester cb
srvrq_update_srv_status() update the server status based on result of SRV
resolution. For now, it is only used from snr_update_srv_status() when
appropriate.
resolv_purge_resolution_answer_records() must be used to removed all answers
attached to a resolution. For now, it is only used when a resolution is
released.
This function search for a SRV answer item associated to a requester
whose type is server.
This is mainly useful to "link" a server to its SRV record when no
additional record were found to configure the IP address.
This patch is required by a bug fix.
action_suggest() will return a pointer to an action whose keyword more or
less ressembles the passed argument. It also accepts to be more tolerant
against prefixes (since actions taking arguments are handled as prefixes).
This will be used to suggest approaching words.
Just like with the server keywords, now's the turn of "bind" keywords.
The difference is that 100% of the bind keywords are registered, thus
we do not need the list of extra keywords.
There are multiple bind line parsers today, all were updated:
- peers
- log
- dgram-bind
- cli
$ printf "listen f\nbind :8000 tcut\n" | ./haproxy -c -f /dev/stdin
[NOTICE] 070/101358 (25146) : haproxy version is 2.4-dev11-7b8787-26
[NOTICE] 070/101358 (25146) : path to executable is ./haproxy
[ALERT] 070/101358 (25146) : parsing [/dev/stdin:2] : 'bind :8000' unknown keyword 'tcut'; did you mean 'tcp-ut' maybe ?
[ALERT] 070/101358 (25146) : Error(s) found in configuration file : /dev/stdin
[ALERT] 070/101358 (25146) : Fatal errors found in configuration.
Instead of just reporting "unknown keyword", let's provide a function which
will look through a list of registered keywords for a similar-looking word
to the one that wasn't matched. This will help callers suggest correct
spelling. Also, given that a large part of the config parser still relies
on a long chain of strcmp(), we'll need to be able to pass extra candidates.
Thus the function supports an optional extra list for this purpose.
This introduces two functions, one which creates a fingerprint of a word,
and one which computes a distance between two words fingerprints. The
fingerprint is made by counting the transitions between one character and
another one. Here we consider the 26 alphabetic letters regardless of
their case, then any digit as a digit, and anything else as "other". We
also consider the first and last locations as transitions from begin to
first char, and last char to end. The distance is simply the sum of the
squares of the differences between two fingerprints. This way, doubling/
missing a letter has the same cost, however some repeated transitions
such as "e"->"r" like in "server" are very unlikely to match against
situations where they do not exist. This is a naive approach but it seems
to work sufficiently well for now. It may be refined in the future if
needed.
It is possible to have a session without a listener. It happens for applets
on the client side. Thus all accesses to the listener info from the session
must be guarded. It was the purpose of the commit 36119de18 ("BUG/MEDIUM:
session: NULL dereference possible when accessing the listener"). However,
some tests on the session's listener existence are missing in proxy_inc_*
functions.
This patch should fix the issues #1171, #1172, #1173, #1174 and #1175. It
must be backported with the above commit as far as 1.8.
Since commit f8fb4f75f ("MINOR: atomic: implement a more efficient arm64
__ha_cas_dw() using pairs"), on some modern arm64 (armv8.1+) compiled
with -march=armv8.1-a under gcc-7.5.0, a build error may appear on ev_poll.o :
/tmp/ccHD2lN8.s:1771: Error: reg pair must start from even reg at operand 1 -- `casp x27,x28,x22,x23,[x12]'
Makefile:927: recipe for target 'src/ev_poll.o' failed
It appears that the compiler cannot always assign register pairs there
for a structure made of two u64. It was possibly later addressed since
gcc-9.3 never caused this, but there's no trivially available info on
the subject in the changelogs. Unsuprizingly, using a u128 instead does
fix this, but it significantly inflates the code (+4kB for just 6 places,
very likely that it loaded some extra stubs) and the comparison is ugly,
involving two slower conditional jumps instead of a single one and a
conditional comparison. For example, ha_random64() grew from 144 bytes
to 232.
However, simply forcing the base register does work pretty well, and
makes the code even cleaner and more efficient by further reducing it
by about 4.5kB, possibly because it helps the compiler to pick suitable
registers for the pair there. And the perf on 64-cores looks steadily
0.5% above the previous one, so let's do this.
Note that the commit above was backported to 2.3 to fix scalability
issues on AWS Graviton2 platform, so this one will need to be as well.
The QUIC connection struct connection member was not initialized. This may
make randomly haproxy handle TLS connections as QUIC ones only when QUIC support
is enabled leading to such OpenSSL errors (captured from a reg test output, TLS
Client-Hello callback failed):
OpenSSL error[0x10000085] OPENSSL_internal: CONNECTION_REJECTED
OpenSSL error[0x10000410] OPENSSL_internal: SSLV3_ALERT_HANDSHAKE_FAILURE
OpenSSL error[0x1000009a] OPENSSL_internal: HANDSHAKE_FAILURE_ON_CLIENT_HELLO
This patch should fix#1168 github issue.
The recent default runqueue size reduction appeared to have significantly
lowered performance on low-thread count configs. Testing various values
runqueue values on different workloads under thread counts ranging from
1 to 64, it appeared that lower values are more optimal for high thread
counts and conversely. It could even be drawn that the optimal value for
various workloads sits around 280/sqrt(nbthread), and probably has to do
with both the L3 cache usage and how to optimally interlace the threads'
activity to minimize contention. This is much easier to optimally
configure, so let's do this by default now.
The recently introduced Financial Information eXchange (FIX)
converters have some hard coded tags based on the specification that
were misspelled. Specifically, SenderComID and TargetComID should
be SenderCompID and TargetCompID according to the specification [1][2].
This patch updates all references, which includes the converters
themselves, the regression test, and the documentation.
[1] https://fiximate.fixtrading.org/en/FIX.5.0SP2_EP264/tag49.html
[2] https://fiximate.fixtrading.org/en/FIX.5.0SP2_EP264/tag56.html
Parameter "accepted_payload_size" is currently considered regardless
the used nameserver is using TCP or UDP. It remains mandatory to annouce
such capability to support e-dns, so a value have to be announced also
in TCP. Maximum DNS message size in TCP is limited by protocol to 65535
and so for UDP (65507) if system supports such UDP messages. But
the maximum value for this option was arbitrary forced to 8192.
This patch change this maximum to 65535 to allow user to set bigger value
for UDP if its system supports. It also sets accepted_payload_size
in TCP allowing to retrieve huge responses if the configuration uses
TCP nameservers.
The request announcing the accepted_payload_size capability is currently
built at resolvers level and is common to all used nameservers of the
section regardess transport protocol used. A further patch should be
made to at least specify a different payload size depending of the
transport, and perhaps could be forced to 65535 in case of TCP and
maximum would be forced back to 65507 matching UDP max.
This patch is appliable since 2.4 version
These are some leftovers from the ancient code where they were still
called sessions, but these areas in the code remain confusing due to
this naming. They were now called "strm" which will not even affect
indenting nor alignment.
It was brought by commit c44b8de99 ("CLEANUP: connection: Use `VAR_ARRAY`
in `struct tlv` definition") but breaks the build with clang. Actually it
had already been done 6 months ago by commit 4987a4744 ("CLEANUP: tree-wide:
use VAR_ARRAY instead of [0] in various definitions") then reverted by
commit 441b6c31e ("BUILD: connection: fix build on clang after the VAR_ARRAY
cleanup") which explained the same thing but didn't place a comment in the
code to justify this (in short it's just an end of struct marker).
The default proxy was passed as a variable to all parsers instead of a
const, which is not without risk, especially when some timeout parsers used
to make some int pointers point to the default values for comparisons. We
want to be certain that none of these parsers will modify the defaults
sections by accident, so it's important to mark this proxy as const.
This patch touches all occurrences found (89).
TCC happens to define __OPTIMIZE__ at -O2 but doesn't proceed with dead
code elimination, resulting in ha_free() to always reference the link
error symbol. Let's condition this test on __GCC__ which others like
Clang also define.
ha_free() uses code that attempts to set a non-existant variable to provoke
a link-time error, with the expectation that the compiler will not omit that
if the code is unreachable. However, clang will emit it when compiling with
no optimization, so only do that if __OPTIMIZE__ is defined.
There's currently quite some thread contention in the server struct
because frequently fields accessed fields are mixed with those being
often written to by any thread. Let's split this a little bit to
separate a few areas:
- pure config / admin / operating status (almost never changes)
- idle and queuing (fast changes, done almost together)
- LB (fast changes, not necessarily dependent on the above)
- counters (fast changes, at a different instant again)
The actconns list creates massive contention on low server counts because
it's in fact a list of streams using a server, all threads compete on the
list's head and it's still possible to see some watchdog panics on 48
threads under extreme contention with 47 threads trying to add and one
thread trying to delete.
Moving this list per thread is trivial because it's only used by
srv_shutdown_streams(), which simply required to iterate over the list.
The field was renamed to "streams" as it's really a list of streams
rather than a list of connections.
There are multiple per-thread lists in the listeners, which isn't the
most efficient in terms of cache, and doesn't easily allow to store all
the per-thread stuff.
Now we introduce an srv_per_thread structure which the servers will have an
array of, and place the idle/safe/avail conns tree heads into. Overall this
was a fairly mechanical change, and the array is now always initialized for
all servers since we'll put more stuff there. It's worth noting that the Lua
code still has to deal with its own deinit by itself despite being in a
global list, because its server is not dynamically allocated.
It's a real pain not to have access to the list of all registered servers,
because whenever there is a need to late adjust their configuration, only
those attached to regular proxies are seen, but not the peers, lua, logs
nor DNS.
What this patch does is that new_server() will automatically add the newly
created server to a global list, and it does so as well for the 1 or 2
statically allocated servers created for Lua. This way it will be possible
to iterate over all of them.
Some entries are atomically updated by various threads, such as the
global counters, and they're mixed with others which are read all the
time like the mode. This explains why "perf" was seeing a huge access
cost on global.mode in process_stream()! Let's reorder them so that
the static config stuff is at the beginning and the live stuff is at
the end.
These functions are used on the mux layer to indicate that the connection
is becoming idle and that the xprt ought to be careful before checking the
context or that it's not idle anymore and that the context is safe. The
purpose is to allow a mux which is going to release a connection to tell
the xprt to be careful when touching it. At the moment, the xprt are
always careful and that's costly so we want to have the ability to relax
this a bit.
No xprt layer uses this yet.
This flag will be usable by any application. It will be preserved across
wakeups so the application can use it to do various stuff. Some I/O
handlers will soon benefit from this.
It's been too short for quite a while now and is now full. It's still
time to extend it to 32-bits since we have room for this without
wasting any space, so we now gained 16 new bits for future flags.
The values were not reassigned just in case there would be a few
hidden u16 or short somewhere in which these flags are placed (as
it used to be the case with stream->pending_events).
The patch is tagged MEDIUM because this required to update the task's
process() prototype to use an int instead of a short, that's quite a
bunch of places.
The nice field isn't needed anymore for the tasklet so we can move it
from the TASK_COMMON area into the struct task which already has a
hole around the expire entry.
It's cleaner to use a flag from the task's state to detect a tasklet
and it's even cheaper. One of the best benefits is that this will
allow to get the nice field out of the common part since the tasklet
doesn't need it anymore. This commit uses the last task bit available
but that's temporary as the purpose of the change is to extend this.
We frequently need to access a simple and fast PRNG for statistical
purposes. The debug_prng() function did exactly this using a xorshift
generator but its use was limited to debug only. Let's move this to
tools.h and tools.c to make it accessible everywhere. Since it needs to
be fast, its state is thread-local. An initialization function starts a
different initial value for each thread for better distribution.
In stream_add_srv_conn() MT_LIST_ADD() is used instead of MT_LIST_ADDQ(),
resulting in the stream being queued at the end of the server list. This
has no particular effect since we cannot dump the streams on a server,
and this is only used by "shutdown sessions" on a server. But it also
turns out to be significantly faster due to the shorter recovery from
the conflict with an adjacent MT_LIST_DEL(), thus it remains desirable
to use it, but at least it deserves a comment.
In addition to this, it's worth mentioning that this list should creates
extreme contention with threads while almost never used. It should be
made per-thread just like the global streams list.
The reason is that H2 can already require 32 16kB buffers for the mux
output at once, which will deplete the local cache. Thus it makes sense
to go further to leave some time to other connection to release theirs.
In addition, the L2 cache on modern CPUs is already 1 MB, so this change
is welcome in any case.
We've reached a point where the global pools represent a significant
bottleneck with threads. On a 64-core machine, the performance was
divided by 8 between 32 and 64 H2 connections only because there were
not enough entries in the local caches to avoid picking from the global
pools, and the contention on the list there was very high. It becomes
obvious that we need to have an array of lists, but that will require
more changes.
In parallel, standard memory allocators have improved, with tcmalloc
and jemalloc finding their ways through mainstream systems, and glibc
having upgraded to a thread-aware ptmalloc variant, keeping this level
of contention here isn't justified anymore when we have both the local
per-thread pool caches and a fast process-wide allocator.
For these reasons, this patch introduces a new compile time setting
CONFIG_HAP_NO_GLOBAL_POOLS which is set by default when threads are
enabled with thread local pool caches, and we know we have a fast
thread-aware memory allocator (currently set for glibc>=2.26). In this
case we entirely bypass the global pool and directly use the standard
memory allocator when missing objects from the local pools. It is also
possible to force it at compile time when a good allocator is used with
another setup.
It is still possible to re-enable the global pools using
CONFIG_HAP_GLOBAL_POOLS, if a corner case is discovered regarding the
operating system's default allocator, or when building with a recent
libc but a different allocator which provides other benefits but does
not scale well with threads.
There finally is a way to support register pairs on aarch64 assembly
under gcc, it's just undocumented, like many of the options there :-(
As indicated below, it's possible to pass "%H" to mention the high
part of a register pair (e.g. "%H0" to go with "%0"):
https://patchwork.ozlabs.org/project/gcc/patch/59368A74.2060908@foss.arm.com/
By making local variables from pairs of registers via a struct (as
is used in IST for example), we can let gcc choose the correct register
pairs and avoid a few moves in certain situations. The code is now slightly
more efficient than the previous one on AWS' Graviton2 platform, and
noticeably smaller (by 4.5kB approx). A few tests on older releases show
that even Linaro's gcc-4.7 used to support such register pairs and %H,
and by then ATOMICS were not supported so this should not cause build
issues, and as such this patch replaces the earlier implementation.
This variant uses the CASP instruction available on armv8.1-a CPU cores,
which is detected when __ARM_FEATURE_ATOMICS is set (gcc-linaro >= 7,
mainline >= 9). This one was tested on cortex-A55 (S905D3) and on AWS'
Graviton2 CPUs.
The instruction performs way better on high thread counts since it
guarantees some forward progress when facing extreme contention while
the original LL/SC approach is light on low-thread counts but doesn't
guarantee progress.
The implementation is not the most optimal possible. In particular since
the instruction requires to work on register pairs and there doesn't seem
to be a way to force gcc to emit register pairs, we have to decide to force
to use the pair (x0,x1) to store the old value, and (x2,x3) to store the
new one, and this necessarily involves some extra moves. But at least it
does improve the situation with 16 threads and more. See issue #958 for
more context.
Note, a first implementation of this function was making use of an
input/output constraint passed using "+Q"(*(void**)target), which was
resulting in smaller overall code than passing "target" as an input
register only. It turned out that the cause was directly related to
whether the function was inlined or not, hence the "forceinline"
attribute. Any changes to this code should still pay attention to this
important factor.
On highly threaded machines it is possible to occasionally trigger the
watchdog on certain contended areas like the server's connection list,
because while the mechanism inherently cannot guarantee a constant
progress, it lacks CPU relax calls which are absolutely necessary in
this situation to let a thread finish its job.
The loop's "while (1)" was changed to use a "for" statement calling
__ha_cpu_relax() as its continuation expression. This way the "continue"
statements jump to the unique place containing the pause without
excessively inflating the code.
This was sufficient to definitely fix the problem on 64-core ARM Graviton2
machines. This patch should probably be backported once it's confirmed it
also helps on many-cores x86 machines since some people are facing
contention in these environments. This patch depends on previous commit
"REORG: atomic: reimplement pl_cpu_relax() from atomic-ops.h".
An attempt was made to first read the value before exchanging, and it
significantly degraded the performance. It's very likely that this caused
other cores to lose exclusive ownership on their line and slow down their
next xchg operation.
In addition it was found that MT_LIST_ADD is significantly faster than
MT_LIST_ADDQ under high contention, because it fails one step earlier
when conflicting with an adjacent MT_LIST_DEL(). It might be worth
switching some operations' order to favor MT_LIST_ADDQ() instead.
There is some confusion here as we need to place some cpu_relax statements
in some loops where it's not easily possible to condition them on the use
of threads. That's what atomic.h already does. So let's take the various
pl_cpu_relax() implementations from there and place them in atomic.h under
the name __ha_cpu_relax() and let them adapt to the presence or absence of
threads and to the architecture (currently only x86 and aarch64 use a barrier
instruction), though it's very likely that arm would work well with a cache
flushing ISB instruction as well).
This time they were implemented as expressions returning 1 rather than
statements, in order to ease their placement as the loop condition or the
continuation expression inside "for" loops. We should probably do the same
with barriers and a few such other ones.
If dispatch mode or transparent backend is used, the backend connection
target is a proxy instead of a server. In these cases, the reuse of
backend connections is not consistent.
With the default behavior, no reuse is done and every new request uses a
new connection. However, if http-reuse is set to never, the connection
are stored by the mux in the session and can be reused for future
requests in the same session.
As no server is used for these connections, no reuse can be made outside
of the session, similarly to http-reuse never mode. A different
http-reuse config value should not have an impact. To achieve this, mark
these connections as private to have a defined behavior.
For this feature to properly work, the connection hash has been slightly
adjusted. The server pointer as an input as been replaced by a generic
target pointer to refer to the server or proxy instance. The hash is
always calculated on connect_server even if the connection target is not
a server. This also requires to allocate the connection hash node for
every backend connections, not just the one with a server target.
Fix such compilation issues:
include/haproxy/quic_tls.h:157:10: error: implicit conversion from
'enum ssl_encryption_level_t' to 'enum quic_tls_enc_level'
[-Werror=enum-conversion]
157 | return ssl_encryption_application;
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
src/xprt_quic.c: In function 'quic_conn_enc_level_init':
src/xprt_quic.c:2358:13: error: implicit conversion from
'enum quic_tls_enc_level' to 'enum ssl_encryption_level_t'
[-Werror=enum-conversion]
2358 | qel->level = quic_to_ssl_enc_level(level);
| ^
Not detected by all the compilators.
This makes the code more readable and less prone to copy-paste errors.
In addition, it allows to place some __builtin_constant_p() predicates
to trigger a link-time error in case the compiler knows that the freed
area is constant. It will also produce compile-time error if trying to
free something that is not a regular pointer (e.g. a function).
The DEBUG_MEM_STATS macro now also defines an instance for ha_free()
so that all these calls can be checked.
178 occurrences were converted. The vast majority of them were handled
by the following Coccinelle script, some slightly refined to better deal
with "&*x" or with long lines:
@ rule @
expression E;
@@
- free(E);
- E = NULL;
+ ha_free(&E);
It was verified that the resulting code is the same, more or less a
handful of cases where the compiler optimized slightly differently
the temporary variable that holds the copy of the pointer.
A non-negligible amount of {free(str);str=NULL;str_len=0;} are still
present in the config part (mostly header names in proxies). These
ones should also be cleaned for the same reasons, and probably be
turned into ist strings.
A network may be specified to avoid header addition for "forwardfor" and
"orignialto" option via the "except" parameter. However, only IPv4
networks/addresses are supported. This patch adds the support of IPv6.
To do so, the net_addr structure is used to store the parameter value in the
proxy structure. And ipcmp2net() function is used to perform the comparison.
This patch should fix the issue #1145. It depends on the following commit:
* c6ce0ab MINOR: tools: Add function to compare an address to a network address
* 5587287 MINOR: tools: Add net_addr structure describing a network addess
ipcmp2net() function may be used to compare an addres (struct
sockaddr_storage) to a network address (struct net_addr). Among other
things, this function will be used to add support of IPv6 for "except"
parameter of "forwardfor" and "originalto" options.
The net_addr structure describes a IPv4 or IPv6 address. Its ip and mask are
represented. Among other things, this structure will be used to add support
of IPv6 for "except" parameter of "forwardfor" and "originalto" options.
This class will be used exclusively for heavy processing tasklets. It
will be cleaner than mixing them with the bulk ones. For now it's
allocated ~1% of the CPU bandwidth.
The largest part of the patch consists in re-arranging the fields in the
task_per_thread structure to preserve a clean alignment with one more
list head. Since we're now forced to increase the struct past a second
cache line, it now uses 4 cache lines (for easy multiplying) with the
first two ones being exclusively used by local operations and the third
one mostly by atomic operations. Interestingly, this better arrangement
causes less stress and reduced the response time by 8 microseconds at
1 million requests per second.
These function names are unbearably long, they don't even fit into the
screen in "show profiling", let's trim the "_connections" to "_conns",
which happens to match the name of the lists there.
While the scheduler is priority-aware and class-aware, and consistently
tries to maintain fairness between all classes, it doesn't make use of a
fine execution budget to compensate for high-latency tasks such as TLS
handshakes. This can result in many subsequent calls adding multiple
milliseconds of latency between the various steps of other tasklets that
don't even depend on this.
An ideal solution would be to add a 4th queue, have all tasks announce
their estimated cost upfront and let the scheduler maintain an auto-
refilling budget to pick from the most suitable queue.
But it turns out that a very simplified version of this already provides
impressive gains with very tiny changes and could easily be backported.
The principle is to reserve a new task flag "TASK_HEAVY" that indicates
that a task is expected to take a lot of time without yielding (e.g. an
SSL handshake typically takes 700 microseconds of crypto computation).
When the scheduler sees this flag when queuing a tasklet, it will place
it into the bulk queue. And during dequeuing, we accept only one of
these in a full round. This means that the first one will be accepted,
will not prevent other lower priority tasks from running, but if a new
one arrives, then the queue stops here and goes back to the polling.
This will allow to collect more important updates for other tasks that
will be batched before the next call of a heavy task.
Preliminary tests consisting in placing this flag on the SSL handshake
tasklet show that response times under SSL stress fell from 14 ms
before the patch to 3.0 ms with the patch, and even 1.8 ms if
tune.sched.low-latency is set to "on".
Some static functions are now exported and renamed to follow the same
pattern of other exported functions. Here is the list :
* update_server_fqdn: Renamed to srv_update_fqdn and exported
* update_server_check_addr_port: renamed to srv_update_check_addr_port and exported
* update_server_agent_addr_port: renamed to srv_update_agent_addr_port and exported
* update_server_addr: renamed to srv_update_addr
* update_server_addr_potr: renamed to srv_update_addr_port
* srv_prepare_for_resolution: exported
This change is mandatory to move all functions dealing with the server-state
files in a separate file.
Parsed parameters are now stored in the tree of server-state lines. This
way, a line from the global server-state file is only parsed once. Before,
it was parsed a first time to store it in the tree and one more time to load
the server state. To do so, the server-state line object must be allocated
before parsing a line. This means its size must no longer depend on the
length of first parsed parameters (backend and server names). Thus the node
type was changed to use a hashed key instead of a string.
The structure used to store a server-state line in an eb-tree has a too
generic name. Instead of state_line, the structure is renamed as
server_state_line.
<state_line.name_name> field is a node in an eb-tree. Thus, instead of
"name_name", we now use "node" to name this field. If is a more explicit
name and not too strange.
It is extremely useful to be able to observe the wakeup latency of some
important I/O operations, so let's accept to inflate the tasklet struct
by 8 extra bytes when DEBUG_TASK is set. With just this we have enough
to get live reports like this:
$ socat - /tmp/sock1 <<< "show profiling"
Per-task CPU profiling : on # set profiling tasks {on|auto|off}
Tasks activity:
function calls cpu_tot cpu_avg lat_tot lat_avg
si_cs_io_cb 8099492 4.833s 596.0ns 8.974m 66.48us
h1_io_cb 7460365 11.55s 1.548us 2.477m 19.92us
process_stream 7383828 22.79s 3.086us 18.39m 149.5us
h1_timeout_task 4157 - - 348.4ms 83.81us
srv_cleanup_toremove_connections751 39.70ms 52.86us 10.54ms 14.04us
srv_cleanup_idle_connections 21 1.405ms 66.89us 30.82us 1.467us
task_run_applet 16 1.058ms 66.13us 446.2us 27.89us
accept_queue_process 7 34.53us 4.933us 333.1us 47.58us
Instead of decrementing grq_total once per task picked from the global
run queue, let's do it at once after the loop like we do for other
counters. This simplifies the code everywhere. It is not expected to
bring noticeable improvements however, since global tasks tend to be
less common nowadays.
The function htx_reserve_max_data() should be used to get an HTX DATA block
with the max possible size. A current block may be extended or a new one
created, depending on the HTX message state. But the idea is to let the
caller to copy a bunch of data without requesting many new blocks. It is its
responsibility to resize the block at the end, to set the final block size.
This function will be used to parse messages with small chunks. Indeed, we
can have more than 2700 1-byte chunks in a 16Kb of input data. So it is easy
to understand how this function may help to improve the parsing of chunk
messages.
When a SRV record was created, it used to register the regular server name
resolution callbacks. That said, SRV records and regular server name
resolution don't work the same way, furthermore on error management.
This patch introduces a new call back to manage DNS errors related to
the SRV queries.
this fixes github issue #50.
Backport status: 2.3, 2.2, 2.1, 2.0
In fd_set_running_excl() we don't reset the old mask in the CAS loop,
so if we fail on the first round, we'll forcefully take the FD on the
next one.
In practice it's used bu fd_insert() and fd_delete() only, none of which
is supposed to be passed an FD which is still in use since in practice,
given that for now only listeners may be enabled on multiple threads at
once.
This can be backported to 2.2 but shouldn't result in fixing any user
visible bug for now.
This function has become large with the multi-queue scheduler. We need
to keep the fast path and the debugging parts inlined, but the rest now
moves to task.c just like was done for task_wakeup(). This has reduced
the code size by 6kB due to less inlining of large parts that are always
context-dependent, and as a side effect, has increased the overall
performance by 1%.
The nb_tasks counter was still global and gets incremented and decremented
for each task_new()/task_free(), and was read in process_runnable_tasks().
But it's only used for stats reporting, so doing this this often is
pointless and expensive. Let's move it to the task_per_thread struct and
have the stats sum it when needed.
Historically we used to call __task_wakeup() with a known tree root but
this is not the case and the code has remained needlessly complicated
with the root calculation in task_wakeup() passed in argument to
__task_wakeup() which compares it again.
Let's get rid of this and just move the detection code there. This
eliminates some ifdefs and allows to simplify the test conditions quite
a bit.
This one is systematically misunderstood due to its unclear name. It
is in fact the number of tasks in the local tasklet list. Let's call
it "tasks_in_list" to remove some of the confusion.
This one is exclusively used as a boolean nowadays and is non-zero only
when the thread-local run queue is not empty. Better check the root tree's
pointer and avoid updating this counter all the time.
This counter is solely used for reporting in the stats and is the hottest
thread contention point to date. Moving it to the scheduler and having a
separate one for the global run queue dramatically improves the performance,
showing a 12% boost on the request rate on 16 threads!
In addition, the thread debugging output which used to rely on rqueue_size
was not totally accurate as it would only report task counts. Now we can
return the exact thread's run queue length.
It is also interesting to note that there are still a few other task/tasklet
counters in the scheduler that are not efficiently updated because some cover
a single area and others cover multiple areas. It looks like having a distinct
counter for each of the following entries would help and would keep the code
a bit cleaner:
- global run queue (tree)
- per-thread run queue (tree)
- per-thread shared tasklets list
- per-thread local lists
Maybe even splitting the shared tasklets lists between pure tasklets and
tasks instead of having the whole and tasks would simplify the code because
there remain a number of places where several counters have to be updated.
The lock was still used exclusively to deal with the concurrency between
the "show sess" release handler and a stream_new() or stream_free() on
another thread. All other accesses made by "show sess" are already done
under thread isolation. The release handler only requires to unlink its
node when stopping in the middle of a dump (error, timeout etc). Let's
just isolate the thread to deal with this case so that it's compatible
with the dump conditions, and remove all remaining locking on the streams.
This effectively kills the streams lock. The measured gain here is around
1.6% with 4 threads (374krps -> 380k).
The global streams list is exclusively used for "show sess", to look up
a stream to shut down, and for the hard-stop. Having all of them in a
single list is extremely expensive in terms of locking when using threads,
with performance losses as high as 7% having been observed just due to
this.
This patch makes the list per-thread, since there's no need to have a
global one in this situation. All call places just iterate over all
threads. The most "invasive" changes was in "show sess" where the end
of list needs to go back to the beginning of next thread's list until
the last thread is seen. For now the lock was maintained to keep the
code auditable but a next commit should get rid of it.
The observed performance gain here with only 4 threads is already 7%
(350krps -> 374krps).
The "show sess" CLI command currently lists all streams and needs to
stop at a given position to avoid dumping forever. Since 2.2 with
commit c6e7a1b8e ("MINOR: cli: make "show sess" stop at the last known
session"), a hack consists in unlinking the stream running the applet
and linking it again at the current end of the list, in order to serve
as a delimiter. But this forces the stream list to be global, which
affects scalability.
This patch introduces an epoch, which is a global 32-bit counter that
is incremented by the "show sess" command, and which is copied by newly
created streams. This way any stream can know whether any other one is
newer or older than itself.
For now it's only stored and not exploited.
RAND_keep_random_devices_open is OpenSSL specific function, not
implemented in LibreSSL and BoringSSL. Let us define guard
HAVE_SSL_RAND_KEEP_RANDOM_DEVICES_OPEN in include/haproxy/openssl-compat.h
That guard does not depend anymore on HA_OPENSSL_VERSION
The runqueue_ticks counts the number of task wakeups and is used to
position new tasks in the run queue, but since we've had per-thread
run queues, the values there are not very relevant anymore and the
nice value doesn't apply well if some threads are more loaded than
others. In addition, letting all threads compete over a shared counter
is not smart as this may cause some excessive contention.
Let's move this index close to the run queues themselves, i.e. one per
thread and a global one. In addition to improving fairness, this has
increased global performance by 2% on 16 threads thanks to the lower
contention on rqueue_ticks.
Fairness issues were not observed, but if any were to be, this patch
could be backported as far as 2.0 to address them.
Historically this function would try to wake the most accurate number of
process_stream() waiters. But since the introduction of filters which could
also require buffers (e.g. for compression), things started not to be as
accurate anymore. Nowadays muxes and transport layers also use buffers, so
the runqueue size has nothing to do anymore with the number of supposed
users to come.
In addition to this, the threshold was compared to the number of free buffer
calculated as allocated minus used, but this didn't work anymore with local
pools since these counts are not updated upon alloc/free!
Let's clean this up and pass the number of released buffers instead, and
consider that each waiter successfully called counts as one buffer. This
is not rocket science and will not suddenly fix everything, but at least
it cannot be as wrong as it is today.
This could have been marked as a bug given that the current situation is
totally broken regarding this, but this probably doesn't completely fix
it, it only goes in a better direction. It is possible however that it
makes sense in the future to backport this as part of a larger series if
the situation significantly improves.
The buffer wait queue used to be global historically but this doest not
make any sense anymore given that the most common use case is to have
thread-local pools. Thus there's no point waking up waiters of other
threads after releasing an entry, as they won't benefit from it.
Let's move the queue head to the thread_info structure and use
ti->buffer_wq from now on.
This revert the commit 63e6cba12 ("MEDIUM: server: add server-states version
2"), but keeping all recent features added to the server-sate file. Instead
of adding a 2nd version for the server-state file format to handle the 5 new
fields added during the 2.4 development, these fields are considered as
optionnal during the parsing. So it is possible to load a server-state file
from HAProxy 2.3. However, from 2.4, these new fields are always dumped in
the server-state file. But it should not be a problem to load it on the 2.3.
This patch seems a bit huge but the diff ignoring the space is much smaller.
The version 2 of the server-state file format is reserved for a real
refactoring to address all issues of the current format.
Remove ebmb_node entry from struct connection and create a dedicated
struct conn_hash_node. struct connection contains now only a pointer to
a conn_hash_node, allocated only for connections where target is of type
OBJ_TYPE_SERVER. This will reduce memory footprints for every
connections that does not need http-reuse such as frontend connections.
In MT_LIST_TRY_ADDQ(), deal with the "prev" field of the element before the
"next". If the element is the first in the list, then its next will
already have been locked when we locked list->prev->next, so locking it
again will fail, and we'll start over and over.
This should be backported to 2.3.
The maximum number of connections accepted at once by a thread for a single
listener used to default to 64 divided by the number of processes but the
tasklet-based model is much more scalable and benefits from smaller values.
Experimentation has shown that 4 gives the highest accept rate for all
thread values, and that 3 and 5 come very close, as shown below (HTTP/1
connections forwarded per second at multi-accept 4 and 64):
ac\thr| 1 2 4 8 16
------+------------------------------
4| 80k 106k 168k 270k 336k
64| 63k 89k 145k 230k 274k
Some tests were also conducted on SSL and absolutely no change was observed.
The value was placed into a define because it used to be spread all over the
code.
It might be useful at some point to backport this to 2.3 and 2.2 to help
those who observed some performance regressions from 1.6.
Since a lot of internal callbacks were turned to tasklets, the runqueue
depth had not been readjusted from the default 200 which was initially
used to favor batched processing. But nowadays it appears too large
already based on the following tests conducted on a 8c16t machine with
a simple config involving "balance leastconn" and one server. The setup
always involved the two threads of a same CPU core except for 1 thread,
and the client was running over 1000 concurrent H1 connections. The
number of requests per second is reported for each (runqueue-depth,
nbthread) couple:
rq\thr| 1 2 4 8 16
------+------------------------------
32| 120k 159k 276k 477k 698k
40| 122k 160k 276k 478k 722k
48| 121k 159k 274k 482k 720k
64| 121k 160k 274k 469k 710k
200| 114k 150k 247k 415k 613k <-- default
It's possible to save up to about 18% performance by lowering the
default value to 40. One possible explanation to this is that checking
I/Os more frequently allows to flush buffers faster and to smooth the
I/O wait time over multiple operations instead of alternating phases
of processing, waiting for locks and waiting for new I/Os.
The total round trip time also fell from 1.62ms to 1.40ms on average,
among which at least 0.5ms is attributed to the testing tools since
this is the minimum attainable on the loopback.
After some observation it would be nice to backport this to 2.3 and
2.2 which observe similar improvements, since some users have already
observed some perf regressions between 1.6 and 2.2.
SCTL (signed certificate timestamp list) specified in RFC6962
was implemented in c74ce24cd22e8c683ba0e5353c0762f8616e597d, let
us introduce macro HAVE_SSL_SCTL for the HAVE_SSL_SCTL sake,
which in turn is based on SN_ct_cert_scts, which comes in the same commit
smp_is_safe() function is used to be sure a sample may be safely
modified. For string samples, a test is performed to verify if there is a
null-terminated byte. If not, one is added, if possible. It means if the
sample is not const and if there is some free space in the buffer, after
data. However, we must not try to read the null-terminated byte if the
string sample is too long (data >= size) or if the size is equal to
zero. This last test was not performed. Thus it was possible to consider a
string sample as safe by testing a byte outside the buffer.
Now, a zero size string sample is always considered as unsafe and is
duplicated when smp_make_safe() is called.
This patch must be backported in all stable versions.
It's pretty easy to pre-initialize the index, change it on free() and check
it during the wakeup, so let's do this to ease detection of any accidental
task_wakeup() after a task_free() or tasklet_wakeup() after a tasklet_free().
If this would ever happen we'd then get a backtrace and a core now. The
index's parity is respected so that the call history remains exploitable.
The idea is to know who woke a task up, by recording the last two
callers in a rotating mode. For now it's trivial with task_wakeup()
but tasklet_wakeup_on() will require quite some more changes.
This typically gives this from the debugger:
(gdb) p t->debug
$2 = {
caller_file = {0x0, 0x8c0d80 "src/task.c"},
caller_line = {0, 260},
caller_idx = 1
}
or this:
(gdb) p t->debug
$6 = {
caller_file = {0x7fffe40329e0 "", 0x885feb "src/stream.c"},
caller_line = {284, 284},
caller_idx = 1
}
But it also provides a trivial macro allowing to simply place a call in
a task/tasklet handler that needs to be observed:
DEBUG_TASK_PRINT_CALLER(t);
Then starting haproxy this way would trivially yield such info:
$ ./haproxy -db -f test.cfg | sort | uniq -c | sort -nr
199992 h1_io_cb woken up from src/sock.c:797
51764 h1_io_cb woken up from src/mux_h1.c:3634
65 h1_io_cb woken up from src/connection.c:169
45 h1_io_cb woken up from src/sock.c:777
The two algos defining these functions (first and leastconn) do not need the
server's lock. However it's already present in pendconn_process_next_strm()
so the API must be updated so that the functions may take it if needed and
that the callers indicate whether they already own it.
As such, the call places (backend.c and stream.c) now do not take it
anymore, queue.c was unchanged since it's already held, and both "first"
and "leastconn" were updated to take it if not already held.
A quick test on the "first" algo showed a jump from 432 to 565k rps by
just dropping the lock in stream.c!
This reverts commit 8f1f177ed0.
Repeated tests have shown a small perforamnce degradation of ~1.8%
caused by this patch at high request rates on 16 threads. The exact
cause is not yet perfectly known but it probably stems in slower
accesses for non-64-bit aligned atomic accesses.
The remaining contention on the server lock solely comes from
sess_change_server() which takes the lock to add and remove a
stream from the server's actconn list. This is both expensive
and pointless since we have mt-lists, and this list is only
used by the CLI's "shutdown server sessions" command!
Let's migrate to an mt-list and remove the need for this costly
lock. By doing so, the request rate increased by ~1.8%.
Since OTHER_LOCK is commonly used it's become much more difficult to
profile lock contention by temporarily changing a lock label. Let's
add DEBUG1..5 to serve only for debugging. These ones must not be
used in committed code. We could decide to only define them when
DEBUG_THREAD is set but that would complicate attempts at measuring
performance with debugging turned off.
The server lock was taken preventively for anything in health_adjust(),
including the static config checks needed to detect that the lock was not
needed, while the function is always called on the response path to update
a server's status. This was responsible for huge contention causing a
performance drop of about 17% on 16 threads. Let's move the lock only
where it should be, i.e. inside the function around the critical sections
only. By doing this, a 16-thread process jumped back from 575 to 675 krps.
This should be backported to 2.3 as the situation degraded there, and
maybe later to 2.2.
Always try to remove a connexion from its toremove_list in conn_free.
This prevents a double-free in case the connection is freed but was
already added in toremove_list.
This bug was easily reproduced by running 4-5 runs of inject on a
single-thread instance of haproxy :
$ inject -u 10000 -d 10 -G 127.0.0.1:20080
A crash would soon be triggered in srv_cleanup_toremove_connections.
This does not need to be backported.
move listen status to a helper, defining both status enum and string
definition.
this will be helpful to be reused in prometheus code. It also removes
this hard-to-read nested ternary.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
prometheus approach requires to output all values for a given metric
name; meaning we iterate through all metrics, and then iterate in the
inner loop on all objects for this metric.
In order to allow more code reuse, adapt the stats API to be able to
select one field or fill them all otherwise.
From this patch it should be possible to add support for listen stats in
prometheus.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
This patch introduce the "dns_stream_nameserver" to use DNS over
TCP on strict nameservers. For the upper layer it is analog to
the api used with udp nameservers except that the user que switch
the name server in "stream" mode at the init using "dns_stream_init".
The fallback from UDP to TCP is not handled and this is not the
purpose of this feature. This is done to choose the transport layer
during the initialization.
Currently there is a hardcoded limit of 4 pipelined transactions
per TCP connections. A batch of idle connections is expired every 5s.
This code is designed to support a maximum DNS message size on TCP: 64k.
Note: this code won't perform retry on unanswered queries this
should be handled by the upper layer
This patch splits current dns.c into two files:
The first dns.c contains code related to DNS message exchange over UDP
and in future other TCP. We try to remove depencies to resolving
to make it usable by other stuff as DNS load balancing.
The new resolvers.c inherit of the code specific to the actual
resolvers.
Note:
It was really difficult to obtain a clean diff dur to the amount
of moved code.
Note2:
Counters and stuff related to stats is not cleany separated because
currently counters for both layers are merged and hard to separate
for now.
This patch splits recv and send functions in two layers. the
lowest is responsible of DNS message transactions over
the network. Doing this we could use DNS message layer
for something else than resolving. Load balancing for instance.
This patch also re-works the way to init a nameserver and
introduce the new struct dns_dgram_server to prepare the arrival
of dns_stream_server and the support of DNS over TCP.
The way to retry a send failure of a request because of EAGAIN
was re-worked. Previously there was no control and all "pending"
queries were re-played each time it reaches a EAGAIN. This
patch introduce a ring to stack messages in case of sent
failure. This patch is emptied if poller shows that the
socket is ready again to push messages.
Counters are currently stored into lowlevel nameservers struct but
most of them are resolving layer data and increased in the upper layer
So this patch renames the prototype used to allocate/dump them with prefix
'resolv' waiting for a clean split.
Some types are specific to resolver code and a renamed using
the 'resolv' prefix instead 'dns'.
-struct dns_query_item {
+struct resolv_query_item {
-struct dns_answer_item {
+struct resolv_answer_item {
-struct dns_response_packet {
+struct resolv_response {
This patch adds the attribute packed on struct dns_question
because it is directly memcpy to network building a response.
This patch also removes the commented line:
// struct list options; /* list of option records */
because it is also used directly using memcpy to build a request
and must not contain host data.
Resolv callbacks are also updated to rely on counters and not on
nameservers.
"show stat domain dns" will now show the parent id (i.e. resolvers
section name).
The "show peers" output has become huge due to the dictionaries making it
less readable. Now this feature has reached a certain level of maturity
which doesn't warrant to dump it all the time, given that it was essentially
needed by developers. Let's make it optional, and disabled by default, only
when "show peers dict" is requested. The default output reminds about the
command. The output has been divided by 5 :
$ socat - /tmp/sock1 <<< "show peers dict" | wc -l
125
$ socat - /tmp/sock1 <<< "show peers" | wc -l
26
It could be useful to backport this to recent stable versions.
Now default proxies are stored into a dedicated tree, sorted by name.
Only unnamed entries are not kept upon new section creation. The very
first call to cfg_parse_listen() will automatically allocate a dummy
defaults section which corresponds to the previous static one, since
the code requires to have one at a few places.
The first immediately visible benefit is that it allows to reuse
alloc_new_proxy() to allocate a defaults section instead of doing it by
hand. And the secret goal is to allow to keep multiple named defaults
section in memory to reuse them from various proxies.
Now we'll have a tree of named defaults sections. The regular insertion
and lookup functions take care of the capability in order to select the
appropriate tree. A new function proxy_destroy_defaults() removes a
proxy from this tree and frees it entirely.
We don't want to expose this one anymore as we'll soon keep multiple
default proxies. Let's move it inside the parser which is the only
place which still uses it, and initialize it on the fly once needed
instead of doing it at boot time.
The default proxy was passed as a variable, which in addition to being
a PITA to deal with in the config parser, doesn't feel safe to use when
it ought to be const.
This will only affect new code so no backport is needed.
The default proxy was passed as a variable, which in addition to being
a PITA to deal with in the config parser, doesn't feel safe to use when
it ought to be const.
This will only affect new code so no backport is needed.
The default proxy was passed as a variable, which in addition to being
a PITA to deal with in the config parser, doesn't feel safe to use when
it ought to be const.
This will only affect new code so no backport is needed.
This used to be open-coded in cfgparse-listen.c when facing a "defaults"
keyword. Let's move this into proxy_free_defaults(). This code is ugly and
doesn't even reset the just freed pointers. Let's not change this yet.
This code should probably be merged with a generic proxy deinit function
called from deinit(). However there's a catch on uri_auth which cannot be
freed because it might be used by one or several proxies. We definitely
need refcounts there!
This new function takes over the old open-coding that used to be done
for too long in cfg_parse_listen() and it now does everything at once
in a proxy-centric function. The function does all the job of allocating
the structure, initializing it, presetting its defaults from the default
proxy and checking for errors. The code was almost unchanged except for
defproxy being passed as a pointer, and the error message being passed
using memprintf().
This change will be needed to ease reuse of multiple default proxies,
or to create dynamic backends in a distant future.
init_default_instance() was still left in cfgparse.c which is not the
best place to pre-initialize a proxy. Let's place it in proxy.c just
after init_new_proxy(), take this opportunity for renaming it to
proxy_preset_defaults() and taking out init_new_proxy() from it, and
let's pass it the pointer to the default proxy to be initialized instead
of implicitly assuming defproxy. We'll soon be able to exploit this.
Only two call places had to be updated.
struct comp is used in struct proxy but never declared prior to this
so depending on where proxy.h is included, touching the <comp> field
can break the build.
This is just an API bug but it's annoying when trying to tidy the code.
The source list passed in argument must be a const and not a variable,
as it's typically the list head from a default proxy and must obviously
not be modified by the function. No backport is needed as it only impacts
new code.
This is just an API bug but it's annoying when trying to tidy the code.
The default proxy passed in argument must be a const and not a variable.
No backport is needed as it only impacts new code.
In 2.1, commit ee4f5f83d ("MINOR: stats: get rid of the ST_CONVDONE flag")
introduced a subtle bug. By testing curproxy against defproxy in
check_config_validity(), it tried to eliminate the need for a flag
to indicate that stats authentication rules were already compiled,
but by doing so it left the issue opened for the case where a new
defaults section appears after the two proxies sharing the first
one:
defaults
mode http
stats auth foo:bar
listen l1
bind :8080
listen l2
bind :8181
defaults
# just to break above
This config results in:
[ALERT] 042/113725 (3121) : proxy 'f2': stats 'auth'/'realm' and 'http-request' can't be used at the same time.
[ALERT] 042/113725 (3121) : Fatal errors found in configuration.
Removing the last defaults remains OK. It turns out that the cleanups
that followed that patch render it useless, so the best fix is to revert
the change (with the up-to-date flags instead). The flag was marked as
belonging to the config. It's not exact but it's the closest to the
reality, as it's not there to configure the behavior but ti mention
that the config parser did its job.
This could be backported as far as 2.1, but in practice it looks like
nobody ever hit it.
logical followup from cli commands addition, so that the state server
file stays compatible with the changes made at runtime; use previously
added helper to load server attributes.
also alloc a specific chunk to avoid mixing with other called functions
using it
Signed-off-by: William Dauchy <wdauchy@gmail.com>
Even if it is possibly too much work for the current usage, it makes
sure we don't break states file from v2.3 to v2.4; indeed, since v2.3,
we introduced two new fields, so we put them aside to guarantee we can
easily reload from a version 1.
The diff seems huge but there is no specific change apart from:
- introduce v2 where it is needed (parsing, update)
- move away from switch/case in update to be able to reuse code
- move srv lock to the whole function to make it easier
this patch confirm how painful it is to maintain this functionality.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
Use the proxy protocol frame if proxy protocol is activated on the
server line. Do not add anymore these connections in the private list.
If some requests are made with the same proxy fields, they can reuse
the idle connection.
The reg-tests proxy_protocol_send_unique_id must be adapted has it
relied on the side effect behavior that every requests from a same
connection reused a private server connection. Now, a new connection is
created as expected if the proxy protocol fields differ.
The source address is used as an input to the the server connection hash. The
address and port are used as separate hash inputs. Do not add anymore these
connections in the private list.
This parameter is set only if used in the transparent-proxy mode.
The destination address is used as an input to the server connection hash. The
address and port are used as separated hash inputs. Note that they are not used
when statically specified on the server line. This is only useful for dynamic
destination address.
This is typically used when the server address is dynamically set via the
set-dst action. The address and port are separated hash parameters.
Most notably, it should fixed set-dst use case (cf github issue #947).
The sni parameter is an input to the server connection hash. Do not add
anymore connections with dynamic sni in the private list. Thus, it is
now possible to reuse a server connection if they use the same sni.
Compare the connection hash when reusing a connection from the session.
This ensures that a private connection is reused only if it shares the
same set of parameters.
The pointer of the target server is used as a first parameter for the
server connection hash calcul. This prevents the hash to be null when no
specific parameters are present, and can serve as a simple defense
against an attacker trying to reuse a non-conform connection.
This is a preliminary work for the calcul of the backend connection
hash. A structure conn_hash_params is the input for the operation,
containing the various specific parameters of a connection.
The high bits of the hash will reflect the parameters present as input.
A set of macros is written to manipulate the connection hash and extract
the parameters/payload.
The server idle/safe/available connection lists are replaced with ebmb-
trees. This is used to store backend connections, with the new field
connection hash as the key. The hash is a 8-bytes size field, used to
reflect specific connection parameters.
This is a preliminary work to be able to reuse connection with SNI,
explicit src/dst address or PROXY protocol.
This is a preparation work for connection reuse with sni/proxy
protocol/specific src-dst addresses.
Protect every access to idle conn lists with a lock. This is currently
strictly not needed because the access to the list are made with atomic
operations. However, to be able to reuse connection with specific
parameters, the list storage will be converted to eb-trees. As this
structure does not have atomic operation, it is mandatory to protect it
with a lock.
For this, the takeover lock is reused. Its role was to protect during
connection takeover. As it is now extended to general idle conns usage,
it is renamed to idle_conns_lock. A new lock section is also
instantiated named IDLE_CONNS_LOCK to isolate its impact on performance.
Since commit 3169471964 ("MINOR: Add
server port field to server state file.") max_fields was not increased
on version number 1. So this patch aims to fix it. This should be
backported as far as v1.8, but the numbering should be adpated depending
on the version: simply increase the field by 1.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
Amaury reported that the commit 3ce6eed ("MEDIUM: ssl: add a rwlock for
SSL server session cache") introduced some warning during compilation:
include/haproxy/thread.h|411 col 2| warning: enumeration value 'SSL_SERVER_LOCK' not handled in switch [-Wswitch]
This patch fix the issue by adding the right entry in the switch block.
Must be backported where 3ce6eed is backported. (2.4 only for now)
Historically we've been counting lots of client-triggered events in stick
tables to help detect misbehaving ones, but we've been missing the same on
the server side, and there's been repeated requests for being able to count
the server errors per URL in order to precisely monitor the quality of
service or even to avoid routing requests to certain dead services, which
is also called "circuit breaking" nowadays.
This commit introduces http_fail_cnt and http_fail_rate, which work like
http_err_cnt and http_err_rate in that they respectively count events and
their frequency, but they only consider server-side issues such as network
errors, unparsable and truncated responses, and 5xx status codes other
than 501 and 505 (since these ones are usually triggered by the client).
Note that retryable errors are purposely not accounted for, so that only
what the client really sees is considered.
With this it becomes very simple to put some protective measures in place
to perform a redirect or return an excuse page when the error rate goes
beyond a certain threshold for a given URL, and give more chances to the
server to recover from this condition. Typically it could look like this
to bypass a URL causing more than 10 requests per second:
stick-table type string len 80 size 4k expire 1m store http_fail_rate(1m)
http-request track-sc0 base # track host+path, ignore query string
http-request return status 503 content-type text/html \
lf-file excuse.html if { sc0_http_fail_rate gt 10 }
A more advanced mechanism using gpt0 could even implement high/low rates
to disable/enable the service.
Reg-test converteers_ref_cnt_never_dec.vtc was updated to test it.
mul32hi() multiples a constant a with a variable b from 0 to 0xffffffff
and shifts the result by 32 bits. It's visible that it's always impossible
to reach the constant a this way because the product always misses exactly
one unit of a to be preserved. And this cannot be corrected by the caller
either as adding one to the output will only shift the output range, and
it's not possible to pass 2^32 on the ratio <b>. The right approach is to
add "a" after the multiplication so that the input range is always
preserved for all ratio values from 0 to 0xffffffff:
(a=0x00000000 * b=0x00000000 + a=0x00000000) >> 32 = 0x00000000
(a=0x00000000 * b=0x00000001 + a=0x00000000) >> 32 = 0x00000000
(a=0x00000000 * b=0xffffffff + a=0x00000000) >> 32 = 0x00000000
(a=0x00000001 * b=0x00000000 + a=0x00000001) >> 32 = 0x00000000
(a=0x00000001 * b=0x00000001 + a=0x00000001) >> 32 = 0x00000000
(a=0x00000001 * b=0xffffffff + a=0x00000001) >> 32 = 0x00000001
(a=0xffffffff * b=0x00000000 + a=0xffffffff) >> 32 = 0x00000000
(a=0xffffffff * b=0x00000001 + a=0xffffffff) >> 32 = 0x00000001
(a=0xffffffff * b=0xffffffff + a=0xffffffff) >> 32 = 0xffffffff
This is only used in freq_ctr calculations and the slightly lower value
is unlikely to have ever been noticed by anyone. This may be backported
though it is not important.
When adding the server side support for certificate update over the CLI
we encountered a design problem with the SSL session cache which was not
locked.
Indeed, once a certificate is updated we need to flush the cache, but we
also need to ensure that the cache is not used during the update.
To prevent the use of the cache during an update, this patch introduce a
rwlock for the SSL server session cache.
In the SSL session part this patch only lock in read, even if it writes.
The reason behind this, is that in the session part, there is one cache
storage per thread so it is not a problem to write in the cache from
several threads. The problem is only when trying to write in the cache
from the CLI (which could be on any thread) when a session is trying to
access the cache. So there is a write lock in the CLI part to prevent
simultaneous access by a session and the CLI.
This patch also remove the thread_isolate attempt which is eating too
much CPU time and was not protecting from the use of a free ptr in the
session.
HAVE_SSL_CTX_ADD_SERVER_CUSTOM_EXT was introduced in ec60909871
however it was defined as HAVE_SL_CTX_ADD_SERVER_CUSTOM_EXT (missing "S")
let us fix typo
There was a special case made to allow ARMv6 to use unaligned accesses
via a cast in xxHash when __ARM_FEATURE_UNALIGNED is defined. But while
ARMv6 (and v7) does support unaligned accesses, it's only for 32-bit
pointers, not 64-bit ones, leading to bus errors when the compiler emits
an ldrd instruction and the input (e.g. a pattern) is not aligned, as in
issue #1035.
Note that v7 was properly using the packed approach here and was safe,
however haproxy versions 2.3 and older use the old r39 xxhash code which
has the same issue for armv7. A slightly different fix is required there,
by using a different definition of packed for 32 and 64 bits.
The problem is really visible when running v7 code on a v8 kernel because
such kernels do not implement alignment trap emulation, and the process
dies when this happens. This is why in the issue above it was only detected
under lxc. The emulation could have been disabled on v7 as well by writing
zero to /proc/cpu/alignment though.
This commit is a backport of xxhash commit a470f2ef ("update default memory
access for armv6").
Thanks to @srkunze for the report and tests, @stgraber for his help on
setting up an easy reproducer outside of lxc, and @Cyan4973 for the
discussion around the best way to fix this. Details and alternate patches
available on https://github.com/Cyan4973/xxHash/issues/490.
in the same manner of agentaddr, we now:
- permit to set agentport through `port` keyword, like it is the case
for agentaddr through `addr`
- set the priority on `agent-port` keyword when used
- add a flag to be able to test when the value is set like for agentaddr
it makes the behaviour between `addr` and `port` more consistent.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
small consistency problem with `addr` and `agent-addr` options:
for the both options, the last one parsed is always used to set the
agent-check addr. Thus these two lines don't have the same behavior:
server ... addr <addr1> agent-addr <addr2>
server ... agent-addr <addr2> addr <addr1>
After this patch `agent-addr` will always be the priority option over
`addr`. It means we test the flag before setting agentaddr.
We also fix all the places where we did not set the flag to be coherent
everywhere.
I was not really able to determine where this issue is coming from. So
it is probable we may backport it to all stable version where the agent
is supported.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
We can currently change the check-port using the cli command `set server
check-port` but there is a consistency issue when using server state.
This patch aims to fix this problem but will be also a good preparation
work to get rid of checkport flag, so we are able to know when checkport
was set by config.
I am fully aware this is not making github #953 moving forward, I
however think this might be acceptable while waiting for a proper
solution and resolve consistency problem faced with port settings.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
While trying to fix some consistency problem with the config file/cli
(e.g. check-port cli command does not set the flag), we realised
checkport flag was not necessarily needed. Indeed tcpcheck uses service
port as the last choice if check.port is zero. So we can assume if
check.port is zero, it means it was never set by the user, regardless if
it is by the cli or config file. In the longterm this will avoid to
introduce a new consistency issue if we forget to set the flag.
in the same manner of checkport flag, we don't really need checkaddr
flag. We can assume if checkaddr is not set, it means it was never set
by the user or config.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
In order to unify prometheus and stats description, we need to clarify
the description for pending connections.
- remove the BE reference in counters struct, as it is also used in
servers
- remove reference of `qcur` field in description as it is specific to
stats implemention
- try to reword cur and max pending connections description
Signed-off-by: William Dauchy <wdauchy@gmail.com>
The function get_check_status_result() can now be used to get the result
code (CHK_RES_*) corresponding to a check status (HCHK_STATUS_*). It will be
used by the Prometheus exporter when reporting the check status of a server.
The new sched_activity structure will be used to collect task-level
activity based on the target function. The principle is to declare a
large enough array to make collisions rare (256 entries), and hash
the function pointer using a reduced XXH to decide where to store the
stats. On first computation an entry is definitely assigned to the
array and it's done atomically. A special entry (0) is used to store
collisions ("others"). The goal is to make it easy and inexpensive for
the scheduler code to use these to store #calls, cpu_time and lat_time
for each task.
In 2.0, commit d2d3348ac ("MINOR: activity: enable automatic profiling
turn on/off") introduced an automatic mode to enable/disable profiling.
The problem is that the automatic mode automatically changes to on/off,
which implied that the forced on/off modes aren't sticky anymore. It's
annoying when debugging because as soon as the load decreases, profiling
stops.
This makes a small change which ought to have been done first, which
consists in having two states for "auto" (auto-on, auto-off) to
distinguish them from the forced states. Setting to "auto" in the config
defaults to "auto-off" as before, and setting it on the CLI switches to
auto but keeps the current operating state.
This is simple enough to be backported to older releases if needed.
When reporting some values in debugging output we often need to have
some condensed, stable-length values. This function prints a duration
from nanosecond to years with at least 4 digits of accuracy using the
most suitable unit, always on 7 chars.
The allowed chunk size was historically limited to 2GB to avoid risk of
overflow. This restriction is no longer necessary because the chunk size is
immediately stored into a 64bits integer after the parsing. Thus, it is now
possible to raise this limit. However to never fed possibly bogus values
from languages that use floats for their integers, we don't get more than 13
hexa-digit (2^52 - 1). 4 petabytes is probably enough !
This patch should fix the issue #1065. It may be backported as far as
2.1. For the 2.0, the legacy HTTP part must be reviewed. But there is
honestely no reason to do so.
In order to announce support for the Extended CONNECT h2 method by
haproxy, always send the ENABLE_CONNECT_PROTOCOL h2 settings. This new
setting has been described in the rfc 8441.
After receiving ENABLE_CONNECT_PROTOCOL, the client is free to use the
Extended CONNECT h2 method. This can notably be useful for the support
of websocket handshake on http/2.
Support for the rfc 8441 Bootstraping WebSockets with HTTP/2
Convert an Extended CONNECT HTTP/2 request into a htx representation.
The htx message uses the GET method with an Upgrade header field to be
fully compatible with the equivalent HTTP/1.1 Upgrade mechanism.
The Extended CONNECT is of the following form :
:method = CONNECT
:protocol = websocket
:scheme = https
:path = /chat
:authority = server.example.com
The new pseudo-header :protocol has been defined and is used to identify
an Extended CONNECT method. Contrary to standard CONNECT, Extended
CONNECT must have :scheme, :path and :authority defined.
Add the header Sec-Websocket-Key when generating a h1 handshake websocket
without this header. This is the case when doing h2-h1 conversion.
The key is randomly generated and base64 encoded. It is stored on the session
side to be able to verify response key and reject it if not valid.
Support for the rfc 8441 Bootstraping WebSockets with HTTP/2
Convert a 200 status reply from an Extended CONNECT request into a htx
representation. The htx message is set to 101 status code to be fully
compatible with the equivalent HTTP/1.1 Upgrade mechanism.
This conversion is only done if the stream flags H2_SF_EXT_CONNECT_SENT
has been set. This is true if an Extended CONNECT request has already
been seen on the stream.
Besides the 101 status, the additional headers Connection/Upgrade are
added to the htx message. The protocol is set from the value stored in
h2s. Typically it will be extracted from the client request. This is
only used if the client is using h1 as only the HTTP/1.1 101 Response
contains the Upgrade header.
Add the Sec-Websocket-Accept header on a websocket handshake response.
This header may be missing if a h2 server is used with a h1 client.
The response key is calculated following the rfc6455. For this, the
handshake request key must be stored in the h1 session, as a new field
name ws_key. Note that this is only done if the message has been
prealably identified as a Websocket handshake request.
If a request is identified as a WebSocket handshake, it must contains a
websocket key header or else it can be reject, following the rfc6455.
A new flag H1_MF_UPG_WEBSOCKET is set on such messages. For the request
te be identified as a WebSocket handshake, it must contains the headers:
Connection: upgrade
Upgrade: websocket
This commit is a compagnon of
"MEDIUM: h1: generate WebSocket key on response if needed" and
"MEDIUM: h1: add a WebSocket key on handshake if needed".
Indeed, it ensures that a WebSocket key is added only from a http/2 side
and not for a http/1 bogus peer.
The H2 message flag H2_MSGF_BODYLESS_RSP is now used during the request or
the response parsing to notify the mux that, considering the parsed message,
the response is known to have no body. This happens during HEAD requests
parsing and during 204/304 responses parsing.
On the H2 multiplexer, the equivalent flag is set on H2 streams. Thus the
H2_SF_BODYLESS_RESP flag is set on a H2 stream if the H2_MSGF_BODYLESS_RSP
is found after a HEADERS frame parsing. Conversely, this flag is also set
when a HEADERS frame is emitted for HEAD requests and for 204/304 responses.
The H2_SF_BODYLESS_RESP flag will be used to ignore data payload from the
response but not the trailers.
The EOM block may be removed. The HTX_FL_EOM flags is enough. Most of time,
to know if the end of the message is reached, we just need to have an empty
HTX message with HTX_FL_EOM flag set. It may also be detected when the last
block of a message with HTX_FL_EOM flag is manipulated.
Removing EOM blocks simplifies the HTX message filling. Indeed, there is no
more edge problems when the message ends but there is no more space to write
the EOM block. However, some part are more tricky. Especially the
compression filter or the FCGI mux. The compression filter must finish the
compression on the last DATA block. Before it was performed on the EOM
block, an extra DATA block with the checksum was added. Now, we must detect
the last DATA block to be sure to finish the compression. The FCGI mux on
its part must be sure to reserve the space for the empty STDIN record on the
last DATA block while this record was inserted on the EOM block.
The H2 multiplexer is probably the part that benefits the most from this
change. Indeed, it is now fairly easier to known when to set the ES flag.
The HTX documentaion has been updated accordingly.
The htx_is_unique_blk() function may now be used to know if a block is the
only one in an HTX message, excluding all unused blocks. Note the purpose of
this function is not to know if a block is the last one of an HTTP message.
This means no more data part from the message are expected, except tunneled
data. It only says if a block is alone in an HTX message.
Add an HTX start-line flag and its counterpart into the HTTP message to
track the presence of the Upgrade option into the Connection header. This
way, without parsing the Connection header again, it will be easy to know if
a client asks for a protocol upgrade and if the server agrees to do so. It
will also be easy to perform some conformance checks when a
101-switching-protocols is received.
TCP to H1 upgrades are buggy for now. When such upgrade is performed, a
crash is experienced. The bug is the result of the recent H1 mux
refactoring, and more specifically because of the commit c4bfa59f1 ("MAJOR:
mux-h1: Create the client stream as later as possible"). Indeed, now the H1
mux is responsible to create the frontend conn-stream once the request
headers are fully received. Thus the TCP to H1 upgrade is a problem because
the frontend conn-stream already exists.
To fix the bug, we must keep this conn-stream and the associate stream and
use it in the H1 mux. To do so, the upgrade will be performed in two
steps. First, the mux is upgraded from mux-pt to mux-h1. Then, the mux-h1
performs the stream upgrade, once the request headers are fully received and
parsed. To do so, stream_upgrade_from_cs() must be used. This function set
the SF_HTX flags to switch the stream to HTX mode, it removes the SF_IGNORE
flags and eventually it fills the request channel with some input data.
This patch is required to fix the TCP to H1 upgrades and is intimately
linked with the next commits.
A bug was introduced by the early insertion of idle connections at the
end of connect_server. It is possible to reuse a connection not yet
ready waiting for an handshake (for example with proxy protocol or ssl).
A wrong duplicate xprt_handshake_io_cb tasklet is thus registered as a
side-effect.
This triggers the BUG_ON statement of xprt_handshake_subscribe :
BUG_ON(ctx->subs && ctx->subs != es);
To counter this, a check is now present in session_get_conn to only
return a connection without the flag CO_FL_WAIT_XPRT. This might cause
sometimes the creation of dedicated server connections when in theory
reuse could have been used, but probably only occurs rarely in real
condition.
This behavior is present since commit :
MEDIUM: connection: Add private connections synchronously in session server list
It could also be further exagerated by :
MEDIUM: backend: add reused conn to sess if mux marked as HOL blocking
It can be backported up to 2.3.
NOTE : This bug seems to be only reproducible with mode tcp, for an
unknown reason. However, reuse should never happen when not in http
mode. This improper behavior will be the subject of a dedicated patch.
This bug can easily be reproducible with the following config (a
webserver is required to accept proxy protocol on port 31080) :
global
defaults
mode tcp
timeout connect 1s
timeout server 1s
timeout client 1s
listen li
bind 0.0.0.0:4444
server bla1 127.0.0.1:31080 check send-proxy-v2
with the inject client :
$ inject -u 10000 -d 10 -G 127.0.0.1:4444
This should fix the github issue #1058.
Building with `"DEBUG=-DDEBUG_STRICT=1 -DDEBUG_USE_ABORT=1"` previously emitted the warning:
In file included from include/haproxy/api.h:35:0,
from src/mux_pt.c:13:
include/haproxy/buf.h: In function ‘br_init’:
include/haproxy/bug.h:42:90: warning: implicit declaration of function ‘abort’ [-Wimplicit-function-declaration]
#define ABORT_NOW() do { extern void ha_backtrace_to_stderr(); ha_backtrace_to_stderr(); abort(); } while (0)
^
include/haproxy/bug.h:56:21: note: in expansion of macro ‘ABORT_NOW’
#define CRASH_NOW() ABORT_NOW()
^
include/haproxy/bug.h:68:4: note: in expansion of macro ‘CRASH_NOW’
CRASH_NOW(); \
^
include/haproxy/bug.h:62:35: note: in expansion of macro ‘__BUG_ON’
#define _BUG_ON(cond, file, line) __BUG_ON(cond, file, line)
^
include/haproxy/bug.h:61:22: note: in expansion of macro ‘_BUG_ON’
#define BUG_ON(cond) _BUG_ON(cond, __FILE__, __LINE__)
^
include/haproxy/buf.h:875:2: note: in expansion of macro ‘BUG_ON’
BUG_ON(size < 2);
^
This patch fixes that issue. The `DEBUG_USE_ABORT` option exists for use with
static analysis tools. No backport needed.
Since the server SSL_CTX is now stored in the ckch_inst, it is not
needed anymore to pass an SSL_CTX to ckch_inst_new_load_srv_store() and
ssl_sock_load_srv_ckchs().
The client_crt member is not used anymore since the server's ssl context
initialization now behaves the same way as the bind lines one (using
ckch stores and instances).
When trying to update a backend certificate, we should find a
server-side ckch instance thanks to which we can rebuild a new ssl
context and a new ckch instance that replace the previous ones in the
server structure. This way any new ssl session will be built out of the
new ssl context and the newly updated certificate.
This resolves a subpart of GitHub issue #427 (the certificate part)
In order for the backend server's certificate to be hot-updatable, it
needs to fit into the implementation used for the "bind" certificates.
This patch follows the architecture implemented for the frontend
implementation and reuses its structures and general function calls
(adapted for the server side).
The ckch store logic is kept and a dedicated ckch instance is used (one
per server). The whole sni_ctx logic was not kept though because it is
not needed.
All the new functions added in this patch are basically server-side
copies of functions that already exist on the frontend side with all the
sni and bind_cond references removed.
The ckch_inst structure has a new 'is_server_instance' flag which is
used to distinguish regular instances from the server-side ones, and a
new pointer to the server's structure in case of backend instance.
Since the new server ckch instances are linked to a standard ckch_store,
a lookup in the ckch store table will succeed so the cli code used to
update bind certificates needs to be covered to manage those new server
side ckch instances.
Split the server's ssl context initialization into the general ssl
related initializations and the actual initialization of a single
SSL_CTX structure. This way the context's initialization will be
usable by itself from elsewhere.
It is only a problem on the response path because the request payload length
it always known. But when a filter is registered to analyze the response
payload, the filtering may hang if the server closes just after the headers.
The root cause of the bug comes from an attempt to allow the filters to not
immediately forward the headers if necessary. A filter may choose to hold
the headers by not forwarding any bytes of the payload. For a message with
no payload but a known payload length, there is always a EOM block to
forward. Thus holding the EOM block for bodyless messages is a good way to
also hold the headers. However, messages with an unknown payload length,
there is no EOM block finishing the message, but only a SHUTR flag on the
channel to mark the end of the stream. If there is no payload when it
happens, there is no payload at all to forward. In the filters API, it is
wrongly detected as a condition to not forward the headers.
Because it is not the most used feature and not the obvious one, this patch
introduces another way to hold the message headers at the begining of the
forwarding. A filter flag is added to explicitly says the headers should be
hold. A filter may choose to set the STRM_FLT_FL_HOLD_HTTP_HDRS flag and not
forwad anything to hold the headers. This flag is removed at each call, thus
it must always be explicitly set by filters. This flag is only evaluated if
no byte has ever been forwarded because the headers are forwarded with the
first byte of the payload.
reg-tests/filters/random-forwarding.vtc reg-test is updated to also test
responses with unknown payload length (with and without payload).
This patch must be backported as far as 2.0.
prometheus approach requires to output all values for a given metric
name; meaning we iterate through all metrics, and then iterate in the
inner loop on all objects for this metric.
In order to allow more code reuse, adapt the stats API to be able to
select one field or fill them all otherwise.
This patch follows what has already been done on frontend and backend
side.
From this patch it should be possible to remove most of the duplicate
code on prometheuse side for the server.
A few things to note though:
- state require prior calculation, so I moved that to a sort of helper
`stats_fill_be_stats_computestate`.
- all ST_F*TIME fields requires some minor compute, so I moved it at te
beginning of the function under a condition.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
prometheus approach requires to output all values for a given metric
name; meaning we iterate through all metrics, and then iterate in the
inner loop on all objects for this metric.
In order to allow more code reuse, adapt the stats API to be able to
select one field or fill them all otherwise.
This patch follows what has already been done on frontend side.
From this patch it should be possible to remove most of the duplicate
code on prometheuse side for the backend
A few things to note though:
- status and uweight field requires prior compute, so I moved that to a
sort of helper `stats_fill_be_stats_computesrv`.
- all ST_F*TIME fields requires some minor compute, so I moved it at te
beginning of the function under a condition.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
while working on backend/servers I realised I could have written that in
a better way and avoid one extra break. This is slightly improving
readiness.
also while being here, fix function declaration which was not 100%
accurate.
this patch does not change the behaviour of the code.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
The dump state is now passed to the function so that the caller can adjust
the behavior. A new series of 4 values allow to stop *after* dumping main
instead of before it or any of the usual loops. This allows to also report
BUG_ON() that could happen very high in the call graph (e.g. startup, or
the scheduler itself) while still understanding what the call path was.
The purpose is to enable the dumping of a backtrace on BUG_ON(). While
it's very useful to know that a condition was met, very often some
caller context is missing to figure how the condition could happen.
From now on, on systems featuring backtrace, a backtrace of the calling
thread will also be dumped to stderr in addition to the unexpected
condition. This will help users of DEBUG_STRICT as they'll most often
find this backtrace in their logs even if they can't find their core
file.
A new "debug dev bug" expert-mode CLI command was added to test the
feature.
This function calls the ha_dump_backtrace() function with a locally
allocated buffer and sends the output slightly indented to fd #2. It's
meant to be used as an emergency backtrace dump.
The backtrace dumping code was located into the thread dump function
but it looks particularly convenient to be able to call it to produce
a dump in other situations, so let's move it to its own function and
make sure it's called last in the function so that we can benefit from
tail merging to save one entry.
In order to simplify the code and remove annoying ifdefs everywhere,
let's always export my_backtrace() and make it adapt to the situation
and return zero if not supported. A small update in the thread dump
function was needed to make sure we don't use its results if it fails
now.
use `stats_fill_fe_stats` when possible to avoid duplicating code; make
use of field selector to get the needed field only.
this should not introduce any difference of output.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
prometheus approach requires to output all values for a given metric
name; meaning we iterate through all metrics, and then iterate in the
inner loop on all objects for this metric.
In order to allow more code reuse, adapt the stats API to be able to
select one field or fill them all otherwise.
From this patch it should be possible to remove most of the duplicate
code on prometheuse side for the frontend.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
Another patch in order to try to reconciliate haproxy stats and
prometheus. Here I'm adding a proper start time field in order to make
proper use of uptime field.
That being done we can move the calculation in `fill_info`
Signed-off-by: William Dauchy <wdauchy@gmail.com>
in order to prepare a possible merge of fields between haproxy stats and
prometheus, duplicate 3 fields:
INF_MEMMAX
INF_POOL_ALLOC
INF_POOL_USED
Those were specifically named in MB unit which is not what prometheus
recommends. We therefore used them but changed the unit while doing the
calculation. It created a specific case for that, up to the description.
This patch:
- removes some possible confusion, i.e. using MB field for bytes
- will permit an easier merge of fields such as description
First consequence for now, is that we can remove the calculation on
prometheus side and move it on `fill_info`.
Signed-off-by: William Dauchy <wdauchy@gmail.com>
The MUX_ES_NOTIMPL_ERR exit status is added to allow the multiplexers to
report errors about not implemented features. This will be used by the H1
mux to return 501-not-implemented errors.
Add the support for the 501-not-implemented status code with the
corresponding default message. The documentation is updated accordingly
because it is now part of status codes HAProxy may emit via an errorfile or
a deny/return HTTP action.
When a tcpcheck ruleset uses multiple connections, the existing one must be
closed and destroyed before openning the new one. This part is handled in
the tcpcheck_main() function, when called from the wake callback function
(wake_srv_chk). But it is indeed a problem, because this function may be
called from the mux layer. This means a mux may call the wake callback
function of the data layer, which may release the connection and the mux. It
is easy to see how it is hazardous. And actually, depending on the
scheduling, it leads to crashes.
Thus, we must avoid to release the connection in the wake callback context,
and move this part in the check's process function instead. To do so, we
rely on the CHK_ST_CLOSE_CONN flags. When a connection must be replaced by a
new one, this flag is set on the check, in tcpcheck_main() function, and the
check's task is woken up. Then, the connection is really closed in
process_chk_conn() function.
This patch must be backported as far as 2.2, with some adaptations however
because the code is not exactly the same.
Now the show_fd helpers at the transport and mux levels return an integer
which indicates whether or not the inspected entry looks suspicious. When
an entry is reported as suspicious, "show fd" will suffix it with an
exclamation mark ('!') in the dump, that is supposed to help detecting
them.
For now, helpers were adjusted to adapt to the new API but none of them
reports any suspicious entry yet.
Just like we did for the muxes, now the transport layers will have the
ability to provide helpers to report more detailed information about their
internal context. When the helper is not known, the pointer continues to
be dumped as-is if it's not NULL. This way a transport with no context nor
dump function will not add a useless "xprt_ctx=(nil)" but the pointer will
be emitted if valid or if a helper is defined.
When 0c439d895 ("BUILD: tools: make resolve_sym_name() return a const")
was written, the pointer argument ought to have been turned to const for
more flexibility. Let's do it now.
It's about the third time I get confused by these functions, half of
which manipulate the reference as a whole and those manipulating only
an entry. For me "pat_ref_commit" means committing the pattern reference,
not just an element, so let's rename it. A number of other ones should
really be renamed before 2.4 gets released :-/
This function must be used to emit an alert if a proxy does not have at
least one of the requested capabilities. An additional message may be
appended to the alert.
This is from the output of codespell. It's done at once over a bunch
of files and only affects comments, so there is nothing user-visible.
No backport needed.
commit c55a626217 ("MINOR: contrib/prometheus-exporter: Add
missing global and per-server metrics") is renaming two metrics between
v2.2 and v2.3:
server_idle_connections_current
server_idle_connections_limit
It is breaking some tools which are making use of those metrics while
supporting several haproxy versions. This build_info will permit tools
which make use of metrics to be able to match the haproxy version and
change the list of expected metrics. This was possible using the haproxy
stats socket but not with prometheus export.
This patch follows prometheus best pratices to export specific software
informations. It is adding a new field `build_info` so we can extend it
to other parameters if needed in the future.
example output:
# HELP haproxy_process_build_info HAProxy build info.
# TYPE haproxy_process_build_info gauge
haproxy_process_build_info{version="2.4-dev5-2e1a3f-5"} 1
Even though it is not a bugfix, this patch will make more sense when
backported up to >= 2.0
Signed-off-by: William Dauchy <wdauchy@gmail.com>
This allows using the address of the server rather than the name of the
server for keeping track of servers in a backend for stickiness.
The peers code was also extended to support feeding the dictionary using
this key instead of the name.
Fixes#814
The accept-encoding normalizer now explicitely manages a subset of
encodings which will all have their own bit in the encoding bitmap
stored in the cache entry. This way two requests with the same primary
key will be served the same cache entry if they both explicitely accept
the stored response's encoding, even if their respective secondary keys
are not the same and do not match the stored response's one.
The actual hash of the accept-encoding will still be used if the
response's encoding is unmanaged.
The encoding matching and the encoding weight parsing are done for every
subpart of the accept-encoding values, and a bitmap of accepted
encodings is built for every request. It is then tested upon any stored
response that has the same primary key until one with an accepted
encoding is found.
The specific "identity" and "*" accept-encoding values are managed too.
When storing a response in the key, we also parse the content-encoding
header in order to only set the response's corresponding encoding's bit
in its cache_entry encoding bitmap.
This patch fixes GitHub issue #988.
It does not need to be backported.
If any of the secondary hash normalizing functions raises an error, the
secondary hash will be unusable. In this case, the response will not be
stored anymore.
Add traces to have an idea why this function may fail. In fact
in never fails when the passed parameters are correct, especially the
lengths. This is not the case when a packet is not correctly built
before being encrypted.
Even if the size of frames built by qc_build_frm() are computed so that
not to overflow a buffer, do not rely on this and always makes a packet
build fails if we could not build a frame.
Also add traces to have an idea where qc_build_frm() fails.
Fixes a memory leak in qc_build_phdshk_apkt().
Remove ->ifcdata which was there to control the CRYPTO data sent to the
peer so that not to saturate its reception buffer. This was a sort
of flow control.
Add ->prep_in_flight counter to the QUIC path struct to control the
number of bytes prepared to be sent so that not to saturare the
congestion control window. This counter is increased each time a
packet was built. This has nothing to see with ->in_flight which
is the real in flight number of bytes which have really been sent.
We are olbiged to maintain two such counters to know how many bytes
of data we can prepared before sending them.
Modify traces consequently which were useful to diagnose issues about
the congestion control window usage.
As there is a lot of information in this protocol, this is not
easy to make the traces readable. We remove here a few of them and
shorten some line shortening the variable names.
This patch adds QUIC structs to server struct so that to make the QUIC code
compile. Also initializes the ebtree to store the connections by connection
IDs.
This patch adds a quic_transport_params struct to bind_conf struct
used for the listeners. This is to store the QUIC transport parameters
for the listeners. Also initializes them when calling str2listener().
Before str2sa_range() it's too early to figure we're going to speak QUIC,
and after it's too late as listeners are already created. So it seems that
doing it in str2listener() when the protocol is discovered is the best
place.
Also adds two ebtrees to the underlying receivers to store the connection
by connections IDs (one for the original connection IDs, and another
one for the definitive connection IDs which really identify the connections.
However it doesn't seem normal that it is stored in the receiver nor the
listener. There should be a private context in the listener so that
protocols can store internal information. This element should in
fact be the listener handle.
Something still feels wrong, and probably we'll have to make QUIC and
SSL co-exist: a proof of this is that there's some explicit code in
bind_parse_ssl() to prevent the "ssl" keyword from replacing the xprt.
This patch imports all the definitions for QUIC protocol with few modifications
from 20200720-quic branch of quic-dev repository found at
https://github.com/haproxytech/quic-dev.
QUIC needs to initialize its BIO and SSL session the same way as for SSL over TCP
connections. It needs also to use the same ClientHello callback.
This patch only exports functions and variables shared between QUIC and SSL/TCP
connections.
We add src/quic_sock.c QUIC specific socket management functions as callbacks
for the control layer: ->accept_conn, ->default_iocb and ->rx_listening.
accept_conn() will have to be defined. The default I/O handler only recvfrom()
the datagrams received. Furthermore, ->rx_listening callback always returns 1 at
this time but should returns 0 when reloading the processus.
This is a simple patch to prepare the integration of QUIC support to come.
quic_conn struct is supposed to embed any QUIC specific information for a QUIC
connection.
As QUIC is a connection oriented protocol, this file is almost a copy of
proto_tcp without TCP specific features. To suspend/resume a QUIC receiver
we proceed the same way as for proto_udp receivers.
With the recent updates to the listeners, we don't need a specific set of
quic*_add_listener() functions, the default ones are sufficient. The fields
declaration were reordered to make the various layers more visible like in
other protocols.
udp_suspend_receiver/udp_resume_receiver are up-to-date (the check for INHERITED
is present) and the code being UDP-specific, it's normal to use UDP here.
Note that in the future we might more reasily reference stacked layers so that
there's no more need for specifying the pointer here.
This way we make all xxhash functions inline, with implementations being
directly included within xxhash.h.
Makefile is updated as well, since we don't need to compile and link
xxhash.o anymore.
Inlining should improve performance on small data inputs.
A new XXH3 variant of hash functions shows a noticeable improvement in
performance (especially on small data), and also brings 128-bit support,
better inlining and streaming capabilities.
Performance comparison is available here:
https://github.com/Cyan4973/xxHash/wiki/Performance-comparison
The assembler on MacOS aarch64 interprets ; as the beginning of comments,
so it is not suitable for separating instructions in inline asm. Use \n
instead.
This should be backported to 2.3, 2.2, 2.1, 2.0 and 1.9.
The MAX_DELAY_MS which is set an upper limit to the poll wait time and
force a wakeup this often used to be set to 1 second in order to easily
spot and correct time drifts. This was added 12 years ago at an era
where virtual machines were starting to become common in server
environments while not working particularly well. Nowadays, such issues
are not as common anymore, however forcing 64 threads to wake up every
single second starts to make the process visible on otherwise idle
systems. Let's increase this wakeup interval to one minute. In the worst
case it will make idle threads wake every second, which remains low.
If this is not sufficient anymore on some systems, another approach
would consist in implementing a deep-sleep mode which only triggers
after a while and which is always disabled if any time drift is
observed.
This fixes building hpack from contrib, which failed because of the
undeclared VAR_ARRAY:
make -C contrib/hpack
...
cc -O2 -Wall -g -I../../include -fwrapv -fno-strict-aliasing -c -o gen-enc.o gen-enc.c
In file included from gen-enc.c:18:
../../include/haproxy/hpack-tbl-t.h:105:23: error: 'VAR_ARRAY' undeclared here (not in a function)
105 | struct hpack_dte dte[VAR_ARRAY]; /* dynamic table entries */
...
As discussed in the thread below, let's redefine VAR_ARRAY in this file
so that it remains self-sustaining:
https://www.mail-archive.com/haproxy@formilux.org/msg39212.html
SSL_CTX_get0_privatekey is openssl/boringssl specific function present
since openssl-1.0.2, let us define readable guard for it, not depending
on HA_OPENSSL_VERSION
As Ilya reported in issue #998, gcc 11 complains about misleading code
indentation which is in fact caused by dead assignments to zero after
a loop which stops on zero. Let's clean both of these.
Due to the addition of the OpenTracing filter it is necessary to define
ARGC_OT enum. This value is used in the functions fmt_directive() and
smp_resolve_args().
The OpenTracing filter uses several internal HAProxy functions to work
with variables and therefore requires two static local HAProxy functions,
var_accounting_diff() and var_clear(), to be declared global.
In fact, the var_clear() function was not originally defined as static,
but it lacked a declaration.
SSL_CTX_add_server_custom_ext is openssl specific function present
since openssl-1.0.2, let us define readable guard for it, not depending
on HA_OPENSSL_VERSION
Right now the connection subscribe/unsubscribe code needs to manipulate
FDs, which is not compatible with QUIC. In practice what we need there
is to be able to either subscribe or wake up depending on readiness at
the moment of subscription.
This commit introduces two new functions at the control layer, which are
provided by the socket code, to check for FD readiness or subscribe to it
at the control layer. For now it's not used.
Now we don't touch the fd anymore there, instead we rely on the ->drain()
provided by the control layer. As such the function was renamed to
conn_ctrl_drain().
This is what we need to drain pending incoming data from an connection.
The code was taken from conn_sock_drain() without the connection-specific
stuff. It still takes a connection for now for API simplicity.
conn_fd_handler() is 100% specific to socket code. It's about time
it moves to sock.c which manipulates socket FDs. With it comes
conn_fd_check() which tests for the socket's readiness. The ugly
connection status check at the end of the iocb was moved to an inlined
function in connection.h so that if we need it for other socket layers
it's not too hard to reuse.
The code was really only moved and not changed at all.
The send() loop present in this function and the error handling is already
present in raw_sock_from_buf(). Let's rely on it instead and stop touching
the FD from this place. The send flag was changed to use a more agnostic
CO_SFL_*. The name was changed to "conn_ctrl_send()" to remind that it's
meant to be used to send at the lowest level.
These are two other areas where this fd_stop_recv()/fd_stop_send() makes no
sense anymore. Both happen by definition while the FD is *not* subscribed,
since nowadays it's subscribed after failing recv()/send(), in which case
we cannot close.
These functions used to disable polling for writes when shutting down
but this is no longer used as it still happens later when closing if the
connection was subscribed to FD events. Let's just remove this fake and
undesired dependency on the FD layer.
Add a new http-request action 'set-timeout [server/tunnel]'. This action
can be used to update the server or tunnel timeout of a stream. It takes
two parameters, the timeout name to update and the new timeout value.
This rule is only valid for a proxy with backend capabilities. The
timeout value cannot be null. A sample expression can also be used
instead of a plain value.
Allow the modification of the tunnel timeout on the stream side.
Use a new field in the stream for the tunnel timeout. It is initialized
by the tunnel timeout from backend unless it has already been set by a
set-timeout tunnel rule.
These functions are not used anymore and were quite confusing given that
their names reflected their original role and not the current ones. Let's
kill them before they inspire anyone.
We had cs_close() which forces a CS_SHR_RESET mode on the read side,
and due to this there are a few call places in the checks which
perform a manual call to conn_sock_drain() before calling cs_close().
This is absurd by principle, and it can be counter-productive in the
case of a mux where this could even cause the opposite of the desired
effect by deleting pending frames on the socket before closing.
Let's add cs_drain_and_close() which uses the CS_SHR_DRAIN mode to
prepare this.
QUIC will rely on UDP at the receiver level, and will need these functions
to suspend/resume the receivers. In the future, protocol chaining may
simplify this.
In conn_ctrl_init() and conn_ctrl_close() we now use the control layer's
functions instead of manipulating the FD directly. This is safe since the
control layer is always present when done. Note that now we also adjust
the flag before calling the function to make things cleaner in case such
a layer would need to call the same functions again for any reason.
Currnetly conn_ctrl_init() does an fd_insert() and conn_ctrl_close() does an
fd_delete(). These are the two only short-term obstacles against using a
non-fd handle to set up a connection. Let's have pur these into the protocol
layer, along with the other connection-level stuff so that the generic
connection code uses them instead. This will allow to define new ones for
other protocols (e.g. QUIC).
Since we only support regular sockets at the moment, the code was placed
into sock.c and shared with proto_tcp, proto_uxst and proto_sockpair.
For the sake of an improved readability, let's group the protocol
field members according to where they're supposed to be defined:
- connection layer (note: for now even UDP needs one)
- binding layer
- address family
- socket layer
Nothing else was changed.
The various protocols were made static since there was no point in
exporting them in the past. Nowadays with QUIC relying on UDP we'll
significantly benefit from UDP being exported and more generally from
being able to declare some functions as being the same as other
protocols'.
In an ideal world it should not be these protocols which should be
exported, but the intermediary levels:
- socket layer (sock.c only right now), already exported as functions
but nothing structured at the moment ;
- family layer (sock_inet, sock_unix, sockpair etc): already structured
and exported
- binding layer (the part that relies on the receiver): currently fused
within the protocol
- connectiong layer (the part that manipulates connections): currently
fused within the protocol
- protocol (connection's control): shouldn't need to be exposed
ultimately once the elements above are in an easily sharable way.
This field used to be needed before commit 2b5e0d8b6 ("MEDIUM: proto_udp:
replace last AF_CUST_UDP* with AF_INET*") as it was used as a protocol
entry selector. Since this commit it's always equal to the socket family's
value so it's entirely redundant. Let's remove it now to simplify the
protocol definition a little bit.
The input buffer passed as argument to create a new stream must not be
transferred when the request channel is initialized because the channel
flags are not set at this stage. In addition, the API is a bit confusing
regarding the buffer owner when an error occurred. The caller remains the
owner, but reading the code it is not obvious.
So, first of all, to avoid any ambiguities, comments are added on the
calling chain to make it clear. The buffer owner is the caller if any error
occurred. And the ownership is transferred to the stream on success.
Then, to make things simple, the ownership is transferred at the end of
stream_new(), in case of success. And the input buffer is updated to point
on BUF_NULL. Thus, in all cases, if the caller try to release it calling
b_free() on it, it is not a problem. Of course, it remains the caller
responsibility to release it on error.
The patch fixes a bug introduced by the commit 26256f86e ("MINOR: stream:
Pass an optional input buffer when a stream is created"). No backport is
needed.
With the removal of the family-specific port setting, all protocol had
exactly the same implementation of ->add(). A generic one was created
with the name "default_add_listener" so that all other ones can now be
removed. The API was slightly adjusted so that the protocol and the
listener are passed instead of the listener and the port.
Note that all protocols continue to provide this ->add() method instead
of routinely calling default_add_listener() from create_listeners(). This
makes sure that any non-standard protocol will still be able to intercept
the listener addition if needed.
This could be backported to 2.3 along with the few previous patches on
listners as a pure code cleanup.
At various places we need to set a port on an IPv4 or IPv6 address, and
it requires casts that are easy to get wrong. Let's add a new set_port()
helper to the address family to assist in this. It will be directly
accessible from the protocol and will make the operation seamless.
Right now this is only implemented for sock_inet as other families do
not need a port.
This flags is now unused. It was used in REQ_WAIT_HTTP analyser, when a
stream was waiting for a request, to set the keep-alive timeout or to avoid
to send HTTP errors to client.
The ctl param MUX_EXIT_STATUS can be request to get the exit status of a
multiplexer. For instance, it may be an HTTP status code or an H2 error. For
now, 0 is always returned. When the mux h1 will be able to return HTTP
errors itself, this ctl param will be used to get the HTTP status code from
the logs.
the mux_exit_status enum has been created to map internal mux exist status
to generic one. Thus there is 5 possible status for now: success, invalid
error, timeout error, internal error and unknown.
cumulative numbers of http request and http errors of counters tracked at
the session level and their rates can now be updated at the session level
thanks to two new functions. These functions are not used for now, but it
will be called to keep tracked counters up-to-date if an error occurs before
the stream creation.
The cumulative numbers of http requests, http errors, bytes received and
sent and their respective rates for a tracked counters are now updated using
specific stream independent functions. These functions are used by the
stream but the aim is to allow the session to do so too. For now, there is
no reason to perform these updates from the session, except from the mux-h2
maybe. But, the mux-h1, on the frontend side, will be able to return some
errors to the client, before the stream creation. In this case, it will be
mandatory to update counters tracked at the session level.
It is now possible to set the buffer used by the channel request buffer when
a stream is created. It may be useful if input data are already received,
instead of waiting the first call to the mux rcv_buf() callback. This change
is mandatory to support H1 connection with no stream attached.
For now, the multiplexers don't pass any buffer. BUF_NULL is thus used to
call stream_create_from_cs().
This callback function was only defined by the mux-h1. But it has been
removed in the previous commit because it is unused now. So, we can do a
step forward removing the callback function from the mux definition and the
cs_info structure.
The idle duration between two streams is added to the session structure. It
is not necessarily pertinent on all protocols. In fact, it is only defined
for H1 connections. It is the duration between two H1 transactions. But the
.get_cs_info() callback function on the multiplexers only exists because
this duration is missing at the session level. So it is a simplification
opportunity for a really low cost.
To reduce the cost, a hole in the session structure is filled by moving
.srv_list field at the end of the structure.
The goal is to allow execution of one main lua state per thread.
The array introduces storage of one reference per thread, because each
lua state can have different reference id for a same function. A function
returns the preferred state id according to configuration and current
thread id.
The goal is to allow execution of one main lua state per thread.
"state_from" is a pointer to the parent lua state. "state_id"
is the index of the parent state id in the reference lua states
array. "state_id" is better because the lock is a "== 0" test
which is quick than pointer comparison. In other way, the state_id
index could index other things the the Lua state concerned. I
think to the function references.
The goal is to allow execution of one main lua state per thread.
This function will initialize the struct with other things than 0.
With this function helper, the initialization is centralized and
it prevents mistakes. This patch also keeps a reference to each
declared function in a list. It will be useful in next patches to
control consistency of declared references.
The goal is to allow execution of one main lua state per thread.
The function hlua_ctx_init() now gets the original lua state from
its caller. This allows the initialisation of lua_thread (coroutines)
from any master lua state.
The parent lua state is stored in the hlua struct.
This patch is a temporary transition, it will be modified later.
The goal is to allow execution of one main lua state per thread.
Because this struct will be filled after the configuration parser, we
cannot copy the content. The actual state of the Haproxy code doesn't
justify this change, it is an update preparing next steps.
This function will be useful to check if the keyword is already registered.
Also add a define for the max number of args.
This will be needed by a next patch to fix a bug and will have to be
backported.
This function simply calls action_lookup() on the private service_keywords,
to look up a service name. This will be used to detect double registration
of a same service from Lua.
This will be needed by a next patch to fix a bug and will have to be
backported.
These functions will be useful to check if a keyword is already registered.
This will be needed by a next patch to fix a bug, and will need to be
backported.
Sometimes it would be nice to be able to only trace abnormal events such
as protocol errors. Let's add a new "error" level below the "user" level
for this. This will allow to add TRACE_ERROR() at various error points
and only see them.
This patch adds a new logging variable '%HPO' for logging HTTP path only
(without query string) from relative or absolute URI.
For example:
log-format "hpo=%HPO hp=%HP hu=%HU hq=%HQ"
GET /r/1 HTTP/1.1
=>
hpo=/r/1 hp=/r/1 hu=/r/1 hq=
GET /r/2?q=2 HTTP/1.1
=>
hpo=/r/2 hp=/r/2 hu=/r/2?q=2 hq=?q=2
GET http://host/r/3 HTTP/1.1
=>
hpo=/r/3 hp=http://host/r/3 hu=http://host/r/3 hq=
GET http://host/r/4?q=4 HTTP/1.1
=>
hpo=/r/4 hp=http://host/r/4 hu=http://host/r/4?q=4 hq=?q=4
Since 2.3 default local log format always adds hostame field.
This behavior change was due to log/sink re-work, because according
to rfc3164 the hostname field is mandatory.
This patch re-introduce a legacy "local" format which is analog
to rfc3164 but with hostname stripped. This is the new
default if logs are generated by haproxy.
To stay compliant with previous configurations, the option
"log-send-hostname" acts as if the default format is switched
to rfc3164.
This patch addresses the github issue #963
This patch should be backported in branches >= 2.3.
In issue #958 Ashley Penney reported intermittent crashes on AWS's ARM
nodes which would not happen on x86 nodes. After investigation it turned
out that the Neoverse N1 CPU cores used in the Graviton2 CPU are much
more aggressive than the usual Cortex A53/A72/A55 or any x86 regarding
memory ordering.
The issue that was triggered there is that if a tasklet_wakeup() call
is made on a tasklet scheduled to run on a foreign thread and that
tasklet is just being dequeued to be processed, there can be a race at
two places:
- if MT_LIST_TRY_ADDQ() happens between MT_LIST_BEHEAD() and
LIST_SPLICE_END_DETACHED() if the tasklet is alone in the list,
because the emptiness tests matches ;
- if MT_LIST_TRY_ADDQ() happens during LIST_DEL_INIT() in
run_tasks_from_lists(), then depending on how LIST_DEL_INIT() ends
up being implemented, it may even corrupt the adjacent nodes while
they're being reused for the in-tree storage.
This issue was introduced in 2.2 when support for waking up remote
tasklets was added. Initially the attachment of a tasklet to a list
was enough to know its status and this used to be stable information.
Now it's not sufficient to rely on this anymore, thus we need to use
a different information.
This patch solves this by adding a new task flag, TASK_IN_LIST, which
is atomically set before attaching a tasklet to a list, and is only
removed after the tasklet is detached from a list. It is checked
by tasklet_wakeup_on() so that it may only be done while the tasklet
is out of any list, and is cleared during the state switch when calling
the tasklet. Note that the flag is not set for pure tasks as it's not
needed.
However this introduces a new special case: the function
tasklet_remove_from_tasklet_list() needs to keep both states in sync
and cannot check both the state and the attachment to a list at the
same time. This function is already limited to being used by the thread
owning the tasklet, so in this case the test remains reliable. However,
just like its predecessors, this function is wrong by design and it
should probably be replaced with a stricter one, a lazy one, or be
totally removed (it's only used in checks to avoid calling a possibly
scheduled event, and when freeing a tasklet). Regardless, for now the
function exists so the flag is removed only if the deletion could be
done, which covers all cases we're interested in regarding the insertion.
This removal is safe against a concurrent tasklet_wakeup_on() since
MT_LIST_DEL() guarantees the atomic test, and will ultimately clear
the flag only if the task could be deleted, so the flag will always
reflect the last state.
This should be carefully be backported as far as 2.2 after some
observation period. This patch depends on previous patch
"MINOR: task: remove __tasklet_remove_from_tasklet_list()".
This function is only used at a single place directly within the
scheduler in run_tasks_from_lists() and it really ought not be called
by anything else, regardless of what its comment says. Let's delete
it, move the two lines directly into the call place, and take this
opportunity to factor the atomic decrement on tasks_run_queue. A comment
was added on the remaining one tasklet_remove_from_tasklet_list() to
mention the risks in using it.
This function is only called at a single place and adds more confusion
than it removes. It also makes one think it could be used outside of
the scheduler while it must absolutely not. Let's just move its two
lines to the call place, making the code more readable there. In
addition this clearly shows that the preliminary LIST_INIT() is
useless since the entry is immediately overwritten.
In MT_LIST_TRY_ADDQ() and MT_LIST_TRY_ADD() we can't just check if the
element is already in a list, because there's a small race condition, it
could be added between the time we checked, and the time we actually set
its next and prev, so we have to lock it first.
This is required to address issue #958.
This should be backported to 2.3, 2.2 and 2.1.
As suggested by @AGSaidi in issue #958, on ARMv8 its convenient to use
an "isb" instruction in pl_cpu_relax() to improve fairness. Without it
I've met a few watchdog conditions on valid locks with 16 threads,
indicating that some threads couldn't manage to get it in 2 seconds. I
never happened again with it. In addition, the performance increased
by slightly more than 5% thanks to the reduced contention.
This should be backported as far as 2.2, possibly even 2.0.
The parsing of the check options based on tcp-check rules (redis, spop,
smtp, http...) are moved aways from check.c. Now, these functions are placed
in tcpcheck.c. These functions are only related to the tcpcheck ruleset
configured on a proxy and not to the health-check attached to a server.
This option is now ignored because I/O check buffers are now allocated using the
buffer pool. Thus, it is marked as deprecated in the documentation and ignored
during the configuration parsing. The field is also removed from the global
structure.
Because this option is ignored since a recent fix, backported as fare as 2.2,
this patch should be backported too. Especially because it updates the
documentation.
Historically, the input and output buffers of a check are allocated by hand
during the startup, with a specific size (not necessarily the same than
other buffers). But since the recent refactoring of the checks to rely
exclusively on the tcp-checks and to use the underlying mux layer, this part
is totally buggy. Indeed, because these buffers are now passed to a mux,
they maybe be swapped if a zero-copy is possible. In fact, for now it is
only possible in h2_rcv_buf(). Thus the bug concretely only exists if a h2
health-check is performed. But, it is a latent bug for other muxes.
Another problem is the size of these buffers. because it may differ for the
other buffer size, it might be source of bugs.
Finally, for configurations with hundreds of thousands of servers, having 2
buffers per check always allocated may be an issue.
To fix the bug, we now allocate these buffers when required using the buffer
pool. Thus not-running checks don't waste memory and muxes may swap them if
possible. The only drawback is the check buffers have now always the same
size than buffers used by the streams. This deprecates indirectly the
"tune.chksize" global option.
In addition, the http-check regtest have been update to perform some h2
health-checks.
Many thanks to @VigneshSP94 for its help on this bug.
This patch should solve the issue #936. It relies on the commit "MINOR:
tcpcheck: Don't handle anymore in-progress send rules in tcpcheck_main".
Both must be backport as far as 2.2.
bla
The Vary functionality is based on a secondary key that needs to be
calculated for every request to which a server answers with a Vary
header. The Vary header, which can only be found in server responses,
determines which headers of the request need to be taken into account in
the secondary key. Since we do not want to have to store all the headers
of the request until we have the response, we will pre-calculate as many
sub-hashes as there are headers that we want to manage in a Vary
context. We will only focus on a subset of headers which are likely to
be mentioned in a Vary response (accept-encoding and referer for now).
Every managed header will have its own normalization function which is
in charge of transforming the header value into a core representation,
more robust to insignificant changes that could exist between multiple
clients. For instance, two accept-encoding values mentioning the same
encodings but in different orders should give the same hash.
This patch adds a function that parses a Vary header value and checks if
all the values belong to our supported subset. It also adds the
normalization functions for our two headers, as well as utility
functions that can prebuild a secondary key for a given request and
transform it into an actual secondary key after the vary signature is
determined from the response.
When at least one data filter is registered on a channel, the offsets of all
filters must be kept up to date. For data filters but also for others. It is
safer to do it in that way. Indirectly, this patch fixes 2 hidden bugs
revealed by the commit 22fca1f2c ("BUG/MEDIUM: filters: Forward all filtered
data at the end of http filtering").
The first one, the worst of both, happens at the end of http filtering when
at least one data filtered is registered on the channel. We call the
http_end() callback function on the filters, when defined, to finish the
http filtering. But it is performed for all filters. Before the commit
22fca1f2c, the only risk was to call the http_end() callback function
unexpectedly on a filter. Now, we may have an overflow on the offset
variable, used at the end to forward all filtered data. Of course, from the
moment we forward an arbitrary huge amount of data, all kinds of bad things
may happen. So offset computation is performed for all filters and
http_end() callback function is called only for data filters.
The other one happens when a data filter alter the data of a channel, it
must update the offsets of all previous filters. But the offset of non-data
filters must be up to date, otherwise, here too we may have an integer
overflow.
Another way to fix these bugs is to always ignore non-data filters from the
offsets computation. But this patch is safer and probably easier to
maintain.
This patch must be backported in all versions where the above commit is. So
as far as 2.0.
Level-7 retries are only possible with a restricted number of HTTP
return codes. While it is usually not safe to retry on 401 and 403, I
came up with an authentication backend which was not synchronizing
authentication of users. While not perfect, being allowed to also retry
on those return codes is really helpful and acts as a hotfix until we
can fix the backend.
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
This patch adds -m flag which allows to specify header name
matching method when deleting headers from http request/response.
Currently beg, end, sub, str and reg are supported.
This is related to GitHub issue #909
Baptiste reported a new crash affecting 2.3 which can be triggered
when using H2 on the backend, with http-reuse always and with a tens
of clients doing close only. There are a few combined cases which cause
this to happen, but each time the issue is the same, an already freed
session is dereferenced in session_unown_conn().
Two cases were identified to cause this:
- a connection referencing a session as its owner, which is detached
from the session's list and is destroyed after this session ends.
The test on conn->owner before calling session_unown_conn() is not
sufficent as the pointer is not null but is not valid anymore.
- a connection that never goes idle and that gets killed form the
mux, where session_free() is called first, then conn_free() calls
session_unown_conn() which scans the just freed session for older
connections. This one is only triggered with DEBUG_UAF
The reason for this session to be present here is that it's needed during
the connection setup, to be passed to conn_install_mux_be() to mux->init()
as the owning session, but it's never deleted aftrewards. Furthermore, even
conn_session_free() doesn't delete this pointer after freeing the session
that lies there. Both do definitely result in a use-after-free that's more
easily triggered under DEBUG_UAF.
This patch makes sure that the owner is always deleted after detaching
or killing the session. However it is currently not possible to clear
the owner right after a synchronous init because the proxy protocol
apparently needs it (a reg test checks this), and if we leave it past
the connection setup with the session not attached anywhere, it's hard
to catch the right moment to detach it. This means that the session may
remain in conn->owner as long as the connection has never been added to
nor removed from the session's idle list. Given that this patch needs to
remain simple enough to be backported, instead it adds a workaround in
session_unown_conn() to detect that the element is already not attached
anywhere.
This fix absolutely requires previous patch "CLEANUP: connection: do not
use conn->owner when the session is known" otherwise the situation will
be even worse, as some places used to rely on conn->owner instead of the
session.
The fix could theorically be backported as far as 1.8. However, the code
in this area has significantly changed along versions and there are more
risks of breaking working stuff than fixing real issues there. The issue
was really woken up in two steps during 2.3-dev when slightly reworking
the idle conns with commit 08016ab82 ("MEDIUM: connection: Add private
connections synchronously in session server list") and when adding
support for storing used H2 connections in the session and adding the
necessary call to session_unown_conn() in the muxes. But the same test
managed to crash 2.2 when built in DEBUG_UAF and patched like this,
proving that we used to already leave dangling pointers behind us:
| diff --git a/include/haproxy/connection.h b/include/haproxy/connection.h
| index f8f235c1a..dd30b5f80 100644
| --- a/include/haproxy/connection.h
| +++ b/include/haproxy/connection.h
| @@ -458,6 +458,10 @@ static inline void conn_free(struct connection *conn)
| sess->idle_conns--;
| session_unown_conn(sess, conn);
| }
| + else {
| + struct session *sess = conn->owner;
| + BUG_ON(sess && sess->origin != &conn->obj_type);
| + }
|
| sockaddr_free(&conn->src);
| sockaddr_free(&conn->dst);
It's uncertain whether an existing code path there can lead to dereferencing
conn->owner when it's bad, though certain suspicious memory corruption bugs
make one think it's a likely candidate. The patch should not be hard to
adapt there.
Backports to 2.1 and older are left to the appreciation of the person
doing the backport.
A reproducer consists in this:
global
nbthread 1
listen l
bind :9000
mode http
http-reuse always
server s 127.0.0.1:8999 proto h2
frontend f
bind :8999 proto h2
mode http
http-request return status 200
Then this will make it crash within 2-3 seconds:
$ h1load -e -r 1 -c 10 http://0:9000/
If it does not, it might be that DEBUG_UAF was not used (it's harder then)
and it might be useful to restart.
At a few places we used to rely on conn->owner to retrieve the session
while the session is already known. This is not correct because at some
of these points the reason the connection's owner was still the session
(instead of NULL) is a mistake. At one place a comparison is even made
between the session and conn->owner assuming it's valid without checking
if it's NULL. Let's clean this up to use the session all the time.
Note that this will be needed for a forthcoming fix and will have to be
backported.
