Revert commit fe4abe62c7.
The goal was to make sure for health-checks, we would not get sockets in
TIME_WAIT. To do so, we would not call shutdown() if linger_risk is set.
However that is wrong, and that means shutw would never be forwarded to
the server, and thus we could get connection that are never properly closed.
Instead, to fix the original problem as described here :
https://www.mail-archive.com/haproxy@formilux.org/msg34080.html
Just make sure the checks code call cs_shutr() before calling cs_shutw().
If shutr has been called, conn_sock_shutw() will make no attempt to call
shutdown(), as it knows close() will be called.
We should really review and revamp the shutr/shutw code, as described in
github issue #142.
This should be backported to 1.9 and 2.0.
In commit 86eded6c6 ("CLEANUP: tasks: rename task_remove_from_tasklet_list()
to tasklet_remove_*") which consisted in removing the casts between tasks
and tasklet, I was a bit too fast to believe that we only saw tasklets in
this function since process_runnable_tasks() also uses it with tasks under
a cast. So removing the bookkeeping on task_list_size was not appropriate.
Bah, the joy of casts which hide the real thing...
This patch does two things at once to address this mess once for all:
- it restores the decrement of task_list_size when it's a real task,
but moves it to process_runnable_task() since it's the only place
where it's allowed to call it with a task
- it moves the increment there as well and renames
task_insert_into_tasklet_list() to tasklet_insert_into_tasklet_list()
of obvious consistency reasons.
This way the increment/decrement of task_list_size is made at the only
places where the cast is enforced, so it has less risks to be missed.
The comments on top of these functions were updated to reflect that they
are only supposed to be used with tasklets and that the caller is responsible
for keeping task_list_size up to date if it decides to enforce a task there.
Now we don't have to worry anymore about how these functions work outside
of the scheduler, which is better longterm-wise. Thanks to Christopher for
spotting this mistake.
No backport is needed.
In conn_sock_shutw(), avoid calling shutdown() if linger_risk is set. Not
doing so will result in getting sockets in TIME_WAIT for some time.
This is particularly observable with health checks.
This should be backported to 1.9.
The function really only operates on tasklets, its arguments are always
tasklets cast as tasks to match the function's type, to be cast back to
a struct tasklet. Let's rename it to tasklet_remove_from_tasklet_list(),
take a struct tasklet, and get rid of the undesired task casts.
It's really confusing to call it a task because it's a tasklet and used
in places where tasks and tasklets are used together. Let's rename it
to tasklet to remove this confusion.
In the HTX structure, the field <first> is used to know where to (re)start the
analysis. It may differ from the message's head. It is especially important to
update it to handle 1xx messages, to be sure to restart the analysis on the next
message (another 1xx message or the final one). It is also updated when some
data are forwarded (the headers or part of the body). But this update is an
error and must never be done at the analysis level. It is a bug, because some
sample fetches may be used after the data forwarding (but before the first send
of course). At this stage, if the first block position does not point on the
start-line, most of HTTP sample fetches fail.
So now, when something is forwarding by HTX analyzers, the first block position
is not update anymore.
This issue was reported on Github. See #119. No backport needed.
When channel_full() is called for an HTX stream, we fall back on the HTX
version. This function is called, among other, from tcp_inspect_request(). With
this patch, the inspect delay is respected again.
This patch must be backported to 1.9.
With both I/O and tasks in the same tasklet list, we now have a very
smooth and responsive scheduler, providing a good fairness between I/O
activities. With the lower layers relying on tasklet a lot (I/O wakeup,
subscribe, etc), there may often be a large number of totally autonomous
tasklets doing their business such as forwarding data between two muxes.
But the task scheduler historically refrained from picking tasks from the
priority-ordered run queue to put them into the tasklet list until this
later had less than max_runqueue_depth entries. This was to make sure that
low-latency, high-priority tasks would have an opportunity to be dequeued
before others even if they arrive late. But the counter used for this is
still the tasklet list size, which contains countless I/O events. This
causes an unfairness between unbounded I/Os and bounded tasks, resulting
for example in the CLI responding slower when forwarding 40 Gbps of HTTP
traffic spread over a thousand of connections.
A good solution consists in sticking to the initial intent of
max_runqueue_depth which is to limit the number of tasks in the list
(to maintain fairness between them) and not to limit the number of these
tasks among tasklets. It just turns out that the task_list_size initially
was this task counter and changed over time to be a tasklet list size.
Let's simply refrain from updating it for pure tasklets so that it takes
back its original role of counting real tasks as its name implies. With
this change the CLI becomes instantly responsive under load again.
This patch may possibly be backported to 1.9 though it requires some
careful checks.
Just like we have a synchronous recv() function for the stream interface,
let's have a synchronous send function that we'll be able to call from
different places. For now this only moves the code, nothing more.
We should not update the two directions at once, in fact we should update
the Rx path after recv() and the Tx path after send(). Let's start by
splitting the update function in two for this.
The purpose of making idle-conns switch to SI_ST_CON was to make the
transition detectable and the operation retryable in case of connection
error. Now we have the RDY state for this which is much more suitable
since it indicates a validated connection on which we didn't necessarily
send anything yet. This will still lead to a transition to EST while not
requiring unnatural write polling nor connect timeouts.
The main reason for all the trouble we're facing with stream interface
error or timeout reports during the connection phase is that we currently
can't make the difference between a connection attempt and a validated
connection attempt. It is problematic because we tend to switch early
to SI_ST_EST but can't always do what we want in this state since it's
supposed to be set when we don't need to visit sess_establish() again.
This patch introduces a new state betwen SI_ST_CON and SI_ST_EST, which
is SI_ST_RDY. It indicates that we've verified that the connection is
ready. It's a transient state, like SI_ST_DIS, that cannot persist when
leaving process_stream(). For now it is not set, only verified in various
tests where SI_ST_CON was used or SI_ST_EST depending on the cases.
The stream-int state diagram was minimally updated to reflect the new
state, though it is largely obsolete and would need to be seriously
updated.
The stream interface state checks involving ranges were replaced with
checks on a set of states, already revealing some issues. No issue was
fixed, all was replaced in a one-to-one mapping for easier control. Some
checks involving a strict difference were also replaced with fields to
be clearer. At this stage, the result must be strictly equivalent. A few
tests were also turned to their bit-field equivalent for better readability
or in preparation for upcoming changes.
The test performed in the SPOE filter was swapped so that the closed and
error states are evicted first and that the established vs conn state is
tested second.
At some places we do check for ranges of stream-int states but those
are confusing as states ordering is not well known (e.g. it's not obvious
that CER is between CON and EST). Let's create a bit field from states so
that we can match multiple states at once instead. The new enum si_state_bit
contains SI_SB_* which are state bits instead of state values. The function
si_state_in() indicates if the state in argument is one of those represented
by the bit mask in second argument.
Now that the various handshakes come with their own XPRT, there's no
need for the CONN_FL_SOCK* flags, and the conn_sock_want|stop functions,
so garbage-collect them.
Add a new XPRT that is used when using non-SSL handshakes, such as proxy
protocol or Netscaler, instead of taking care of it in conn_fd_handler().
This XPRT is installed when any of those is used, and it removes itself once
the handshake is done.
This should allow us to remove the distinction between CO_FL_SOCK* and
CO_FL_XPRT*.
In channel_htx_forward() and channel_htx_forward_forever(), if the HTX message
is empty, the underlying buffer may be really empty too. And we have no warranty
the caller will call htx_to_buf() later. And in practice, it is almost never
done. So the channel's buffer must not be altered. Otherwise, the buffer may be
considered as full (data == size) for an empty HTX message and no outgoing data.
This patch must be backported to 1.9.
Make usage of the APIs implemented for dictionaries (dict.c) and their LRU caches (struct dcache)
so that to send/receive server names used for the server by name stickiness. These
names are sent over the network as follows:
- in every case we send the encode length of the data (STD_T_DICT), then
- if the server names is not present in the cache used upon transmission (struct dcache_tx)
we cache it and we the ID of this TX cache entry followed the encode length of the
server name, and finally the sever name itseft (non NULL terminated string).
- if the server name is present, we repead these operations but we only send the TX cache
entry ID.
Upon receipt, the couple of (cache IDs, server name) are stored the LRU cache used
only upon receipt (struct dcache_rx). As the peers protocol is symetrical, the fact
that the server name is present in the received data (resp. or not) denotes if
the entry is absent (resp. or not).
This simple patch only adds definitions to create a new stick-table
data type ID and a new standard type to store information in relation
wich dictionary entries (STD_T_DICT).
This patch adds minimalistic definitions to implement dictionary new data structure
which is an ebtree of ebpt_node structs with strings as keys. Note that this has nothing
to see with real dictionary data structure (maps of keys in association with values).
Have "socks4" and "check-via-socks4" server keyword added.
Implement handshake with SOCKS4 proxy server for tcp stream connection.
See issue #82.
I have the "SOCKS: A protocol for TCP proxy across firewalls" doc found
at "https://www.openssh.com/txt/socks4.protocol". Please reference to it.
[wt: for now connecting to the SOCKS4 proxy over unix sockets is not
supported, and mixing IPv4/IPv6 is discouraged; indeed, the control
layer is unique for a connection and will be used both for connecting
and for target address manipulation. As such it may for example report
incorrect destination addresses in logs if the proxy is reached over
IPv6]
Remove the active_tasks_mask variable, we can deduce if we've work to do
by other means, and it is costly to maintain. Instead, introduce a new
function, thread_has_tasks(), that returns non-zero if there's tasks
scheduled for the thread, zero otherwise.
The functions channel_htx_fwd_payload() and channel_htx_fwd_all() should now be
used to forward, respectively, a part of the HTX payload or all of it. These
functions forward data and update the first block position.
We don't store the start-line position anymore in the HTX message. Instead we
store the first block position to analyze. For now, it is almost the same. But
once all changes will be made on this part, this position will have to be used
by HTX analyzers, and only in the analysis context, to know where the analyse
should start.
When new blocks are added in an HTX message, if the first block position is not
defined, it is set. When the block pointed by it is removed, it is set to the
block following it. -1 remains the value to unset the position. the first block
position is unset when the HTX message is empty. It may also be unset on a
non-empty message, meaning every blocks were already analyzed.
From HTX analyzers point of view, this position is always set during headers
analysis. When they are waiting for a request or a response, if it is unset, it
means the analysis should wait. But once the analysis is started, and as long as
headers are not forwarded, it points to the message start-line.
As mentionned, outside the HTX analysis, no code must rely on the first block
position. So multiplexers and applets must always use the head position to start
a loop on an HTX message.
The function channel_htx_fwd_headers() should now be used by HTX analyzers to
forward all headers of an HTX message, from the start-line to the corresponding
EOH. It takes care to update the star-line position.
When channel_recv_max() is called for an HTX stream, we fall back on the HTX
version. This function is called from si_cs_recv(). This will let us pass the
max amount of bytes to read to HTX multiplexers.
Now, we only return the start-line. If not found, NULL is returned. No lookup is
performed and the HTX message is no more updated. It is now the caller
responsibility to update the position of the start-line to the right value. So
when it is not found, i.e sl_pos is set to -1, it means the last start-line has
been already processed and the next one has not been inserted yet.
It is mandatory to rely on this kind of warranty to store 1xx informational
responses and final reponse in the same HTX message.
It's amazing that the value was still incremented under the date lock,
let's first use an atomic increment for the counter and move it out of
the date lock to reduce contention. These are just counters, we don't
need to take locks if we're not rotating, atomic ops are enough. This
patch does this, and leaves the lock for when the period is over. It's
important to note that some values might be added just before or just
after a rotation but this is not a problem since we don't care if a
value is counted in the previous or next period when it's exactly on
the edge. Great care was taken to ensure that the current counter is
always atomically updated.
Other minor cleanups were performed, such as avoiding to reload the
value from memory after a CAS, or using &~1 instead of two shifts to
remove the lowest bit.
This function dumps a lot of information about a stream into the provided
buffer. It is now used by stream_dump_and_crash() and will be used by the
debugger as well.
The struct mworker_proc is not uniformly freed everywhere, sometimes leading
to leaks of the `id` string (and possibly the other strings).
Introduce a mworker_free_child function instead of duplicating the freeing
logic everywhere to prevent this kind of issues.
This leak was reported in issue #96.
It looks like the leaks have been introduced in commit 9a1ee7ac31,
which is specific to 2.0-dev. Backporting `mworker_free_child` might be
helpful to ease backporting other fixes, though.
Now that we have the guarantee that init calls happen before any other
thread starts, we don't need anymore the workaround installed by commit
1605c7ae6 ("BUG/MEDIUM: threads/mworker: fix a race on startup") and we
can instead rely on a regular per-thread initcall for this function. It
will only be performed on worker thread #0, the other ones and the master
have nothing to do, just like in the original code that was only moved
to the function.
With the thread debugger it becomes visible that we can leave some
wandering pointers for a while in curr_task, which is inappropriate.
This patch addresses this by resetting curr_task to NULL before really
freeing the area. This way it becomes safe even regarding signals.
In conn_xprt_close(), after calling xprt->close(), don't forget to set
conn->xprt_ctx to NULL, or we may attempt to reuse the now-free'd
conn->xprt_ctx if the connection failed and we're retrying it.
It's always a pain to have to stuff lots of #ifdef USE_OPENSSL around
ssl headers, it even results in some of them appearing in a random order
and multiple times just to benefit form an existing ifdef block. Let's
make these headers safe for inclusion when USE_OPENSSL is not defined,
they now perform the test themselves and do nothing if USE_OPENSSL is
not defined. This allows to remove no less than 8 such ifdef blocks
and make include blocks more readable.
Since we're providing a compatibility layer for multiple OpenSSL
implementations and their derivatives, it is important that no C file
directly includes openssl headers but only passes via openssl-compat
instead. As a bonus this also gets rid of redundant complex rules for
inclusion of certain files (engines etc).
LibreSSL causes lots of build issues by pretending to be OpenSSL 2.0.0,
and it requires lots of care for each #if added to cover any specific
OpenSSL features.
This commit addresses the problem by making LibreSSL only advertise the
version it forked from (1.0.1g) and by starting to use tests based on
its real version to enable features instead of working by exclusion.
Most tests on OPENSSL_VERSION_NUMBER have become complex and break all
the time because this number is fake for some derivatives like LibreSSL.
This patch creates a new macro, HA_OPENSSL_VERSION_NUMBER, which will
carry the real openssl version defining the compatibility level, and
this version will be adjusted depending on the variants.
When a sample fetch is encoded, we use its context to set info about the
fragmentation. But if the sample is not found, the function sample_process()
returns NULL. So we me be sure the sample exists before setting its context.
This patch must be backported to 1.9 and 1.8.
This patch implements a new global parameter for the master-worker mode.
When setting the mworker-max-reloads value, a worker receive a SIGTERM
if its number of reloads is greater than this value.
Since previous commit it's not needed anymore to test a task pointer
before calling task_destory() so let's just remove these tests from
the various callers before they become confusing. The function's
arguments were also documented. The same should probably be done
with tasklet_free() which involves a test in roughly half of the
call places.
Commit 3f795f7 ("MEDIUM: tasks: Merge task_delete() and task_free() into
task_destroy().") replaced task_delete() and task_free() with a single
function named task_destroy().
This patch adds a check for struct task* argument in function
task_destroy() to prevent a possible segfault on NULL and also to make
the function safer for use in other cases.
With this patch we add a prefix to stick-table names declared in "peers" sections
concatenating the "peers" section name followed by a '/' character with
the stick-table name. Consequently, "peers" sections have their own
namespace for their stick-tables. Obviously, these stick-table names are not the
ones which should be sent over the network. So these configurations must be
compatible and should make A and B peers communicate with peers protocol:
# haproxy A config, old way stick-table declerations
peers mypeers
peer A ...
peer B ...
backend t1
stick-table type string size 10m store gpc0 peers mypeers
# haproxy B config, new way stick-table declerations
peers mypeers
peer A ...
peer B ...
table t1 type string size store gpc0 10m
This "network" name is stored in ->nid new field of stktable struct. The "local"
stktable-name is still stored in ->id.
This patch adds the support for the "table" line parsing in "peers" sections
to declare stick-table in such sections. This also prevents the user from having
to declare dummy backends sections with a unique stick-table inside.
Even if still supported, this usage will become deprecated.
To do so, the ->table member of proxy struct which is a stktable struct is replaced
by a pointer to a stktable struct allocated at parsing time in src/cfgparse-listen.c
for the dummy stick-table backends and in src/cfgparse.c for "peers" sections.
This has an impact on the code for stick-table sample converters and on the stickiness
rules parsers which first store the name of the dummy before resolving the rules.
This patch replaces proxy_tbl_by_name() calls by stktable_find_by_name() calls
to lookup for stick-tables stored in "stktable_by_name" ebtree at parsing time.
There is only one remaining place where proxy_tbl_by_name() is used: src/hlua.c.
At several places in the code we relied on the fact that ->size member of stick-table
was equal to zero to consider the stick-table was present by not configured,
this do not make sense anymore as ->table member of struct proxyis fow now on a pointer.
These tests are replaced by a test on ->table value itself.
In "peers" section we do not have to temporary store the name of the section the
stick-table are attached to because this name is obviously already known just after
having entered this "peers" section.
About the CLI stick-table I/O handler, the pointer to proxy struct is replaced by
a pointer to a stktable struct.
With this patch we move the code responsible of parsing "stick-table"
lines to implement parse_stick_table() function in src/stick-tabble.c
so that to be able to parse "stick-table" elsewhere than in proxy sections.
We have have also added a conf struct to stktable struct to store the filename
and the line in the file the stick-table has been parsed to help in
diagnosing and displaying any configuration issue.
This implements support for the new API which relies on a call to
setsockopt().
On systems that support it (currently, only Linux >= 4.11), this enables
using TCP fast open when connecting to server.
Please note that you should use the retry-on "conn-failure", "empty-response"
and "response-timeout" keywords, or the request won't be able to be retried
on failure.
Co-authored-by: Olivier Houchard <ohouchard@haproxy.com>
The connect() method had 2 arguments, "data", that tells if there's pending
data to be sent, and "delack" that tells if we have to use a delayed ack
inconditionally, or if the backend is configured with tcp-smart-connect.
Turn that into one argument, "flags".
That way it'll be easier to provide more informations to connect() without
adding extra arguments.
SSL_SESSION_get0_id_context is introduced in LibreSSL-2.7.0
async operations are not supported by LibreSSL
early data is not supported by LibreSSL
packet_length is removed from SSL struct in LibreSSL
When running in HTX mode, if we sent the request, but failed to get the
answer, either because the server just closed its socket, we hit a server
timeout, or we get a 404, 408, 425, 500, 501, 502, 503 or 504 error,
attempt to retry the request, exactly as if we just failed to connect to
the server.
To do so, add a new backend keyword, "retry-on".
It accepts a list of keywords, which can be "none" (never retry),
"conn-failure" (we failed to connect, or to do the SSL handshake),
"empty-response" (the server closed the connection without answering),
"response-timeout" (we timed out while waiting for the server response),
or "404", "408", "425", "500", "501", "502", "503" and "504".
The default is "conn-failure".
When for some reason the session is not the owner of the connection anymore,
make sure we remove CO_FL_SESS_IDLE, even if we're about to call
conn->mux->destroy(), as the destroy may not destroy the connection
immediately if it's still in use.
This should be backported to 1.9.
u
In channel_erase(), don't forget to set output to 0, otherwise the
channel won't seem empty, when it really is, and that could lead to
stream never closing properly.
This should be backported to 1.9.
Port range uses a ring buffer, and unfortunately, when making haproxy
multithreaded, it's been overlooked, and the ring buffer is not thread-safe.
When specifying a source range, 2 or more threads could pick the same
port, and of course only one of them could use the port, the others would
always fail the connection.
To fix this, make it a lock-free ring buffer. This is easier than usual
because we know the ring buffer can never be full.
This should be backported to 1.8 and 1.9.
This patch implements the sampling and load-balancing of log servers configured
with "sample" new keyword implemented by this commit:
'MINOR: log: Add "sample" new keyword to "log" lines'.
As the list of ranges used to sample the log to balance is ordered, we only
have to maintain ->curr_idx member of smp_info struct which is the index of
the sample and check if it belongs or not to the current range to decide if we
must send it to the log server or not.
This simplifies the API and hide the details in the sample. This way, only
string and binary are aware of these info, because other types cannot be
partially encoded.
This patch may be backported to 1.9 and 1.8.
Fragmented arg will do fetch at every encode time, each fetch may get
different result if SMP_F_MAY_CHANGE, for example res.payload, but
the length already encoded in first fragment of the frame, that will
cause SPOA decode failed and waste resources.
This patch must be backported to 1.9 and 1.8.
During 1.9 development (and even a bit after) we've started to face a
significant number of situations where streams were abusively spinning
due to an uncaught error flag or complex conditions that couldn't be
correctly identified. Sometimes streams wake appctx up and conversely
as well. More importantly when this happens the only fix is to restart.
This patch adds a new function to report a serious error, some relevant
info and to crash the process using abort() so that a core dump is
available. The purpose will be for this function to be called in various
situations where the process is unfixable. It will help detect these
issues much earlier during development and may even help fixing test
platforms which are able to automatically restart when such a condition
happens, though this is not the primary purpose.
This patch only provides the function and doesn't use it yet.
We have various functions like conn_get_ctrl_name() to retrieve
some information reported in "show sess" for debugging, which
assume that the connection is valid. This is really not convenient
in code aimed at debugging and is error-prone. Let's add a validity
test first.
Instead of having to manually turn task profiling on/off in the
configuration, by default it will work in "auto" mode, which
automatically turns on on any thread experiencing sustained loop
latencies over one millisecond averaged over the last 1024 samples.
This may happen with configs using lots of regex (thing map_reg for
example, which is the lazy way to convert Apache's rewrite rules but
must not be abused), and such high latencies affect all the process
and the problem is most often intermittent (e.g. hitting a map which
is only used for certain host names).
Thus now by default, with profiling set to "auto", it remains off all
the time until something bad happens. This also helps better focus on
the issues when looking at the logs as well as in "show sess" output.
It automatically turns off when the average loop latency over the last
1024 calls goes below 990 microseconds (which typically takes a while
when in idle).
This patch could be backported to stable versions after a bit more
exposure, as it definitely improves observability and the ability to
quickly spot the culprit. In this case, previous patch ("MINOR:
activity: make the profiling status per thread and not global") must
also be taken.
In order to later support automatic profiling turn on/off, we need to
have it per-thread. We're keeping the global option to know whether to
turn it or on off, but the profiling status is now set per thread. We're
updating the status in activity_count_runtime() which is called before
entering poll(). The reason is that we'll extend this with run time
measurement when deciding to automatically turn it on or off.
Very similarly to previous commit doing the same for streams, we now
measure and report an appctx's call rate. This will help catch applets
which do not consume all their data and/or which do not properly report
that they're waiting for something else. Some of them like peers might
theorically be able to exhibit some occasional peeks when teaching a
full table to a nearby peer (e.g. the new replacement process), but
nothing close to what a bogus service can do so there is no risk of
confusion.
Since the commit 1d2b586cd ("MAJOR: htx: Enable the HTX mode by default for all
proxies"), the HTX is enabled by default for all proxies, HTTP and TCP, but also
CLI and HEALTH proxies. But when the best mux is retrieved, only HTTP and TCP
modes are checked. If the TCP mode is not explicitly set, it is considered as an
HTTP proxy. It is an hidden bug introduced when the option "http-use-htx" was
added. It has no effect until the commit 1d2b586cd. But now, when a stats socket
is created for the master process, the mux h1 is installed on all incoming
connections to the CLI proxy, leading to segfaults because HTX operations are
performed on raw buffers.
So to fix the buf, when a mux is installed, all proxies are considered as TCP
proxies, except HTTP ones. This way, CLI and HEALTH proxies will be handled as
TCP proxies.
This patch must be backported to 1.9 although it has no effect. It is safer to
not keep hidden bugs.
The 'do-resolve' action is an http-request or tcp-request content action
which allows to run DNS resolution at run time in HAProxy.
The name to be resolved can be picked up in the request sent by the
client and the result of the resolution is stored in a variable.
The time the resolution is being performed, the request is on pause.
If the resolution can't provide a suitable result, then the variable
will be empty. It's up to the admin to take decisions based on this
statement (return 503 to prevent loops).
Read carefully the documentation concerning this feature, to ensure your
setup is secure and safe to be used in production.
This patch creates a global counter to track various errors reported by
the action 'do-resolve'.
As by default we add all keepalive connections to the idle pool, if we run
into a pathological case, where all client don't do keepalive, but the server
does, and haproxy is configured to only reuse "safe" connections, we will
soon find ourself having lots of idling, unusable for new sessions, connections,
while we won't have any file descriptors available to create new connections.
To fix this, add 2 new global settings, "pool_low_ratio" and "pool_high_ratio".
pool-low-fd-ratio is the % of fds we're allowed to use (against the maximum
number of fds available to haproxy) before we stop adding connections to the
idle pool, and destroy them instead. The default is 20. pool-high-fd-ratio is
the % of fds we're allowed to use (against the maximum number of fds available
to haproxy) before we start killing idling connection in the event we have to
create a new outgoing connection, and no reuse is possible. The default is 25.
For most of the xprt methods, provide a xprt_ctx. This will be useful later
when we'll want to be able to stack xprts.
The init() method now has to create and provide the said xprt_ctx if needed.
task_delete() was never used without calling task_free() just after, and
task_free() was only used on error pathes to destroy a just-created task,
so merge them into task_destroy(), that will remove the task from the
wait queue, and make sure the task is either destroyed immediately if it's
not in the run queue, or destroyed when it's supposed to run.
The wakeup condition in task_wakeup() is redundant as it is already
validated by the CAS. Better move the __task_wakeup() call there, it
also has the merit of being easier to audit this way. This also reduces
the code size by around 1.8 kB :
$ size haproxy-?
text data bss dec hex filename
2153806 100208 1307676 3561690 3658da haproxy-1
2152094 100208 1307676 3559978 36522a haproxy-2
Commit 0c7a4b6 ("MINOR: tasks: Don't set the TASK_RUNNING flag when
adding in the tasklet list.") revealed a hole in the way tasks may
be freed : they could be removed while in the run queue when the
TASK_QUEUED flag was present but not the TASK_RUNNING one. But it
seems the issue was emphasized by commit cde7902 ("MEDIUM: tasks:
improve fairness between the local and global queues") though the
code it replaces was already affected given how late the TASK_RUNNING
flag was set after removal from the global queue.
At the moment the task is picked from the global run queue, if it
is the last one, the global run queue lock is dropped, and then
the TASK_RUNNING flag was added. In the mean time another thread
might have performed a task_free(), and immediately after, the
TASK_RUNNING flag was re-added to the task, which was then added
to the tasklet list. The unprotected window was extremely faint
but does definitely exist and inconsistent task lists have been
observed a few times during very intensive tests over the last few
days. From this point various options are possible, the task might
have been re-allocated while running, and assigned state 0 and/or
state QUEUED while it was still running, resulting in the tast not
being put back into the tree.
This commit simply makes sure that tests on TASK_RUNNING before removing
the task also cover TASK_QUEUED.
It must be backported to 1.9 along with the previous ones touching
that area.
Now that we have the warranty that a task won't be added in the runqueue
while the TASK_QUEUED or the TASK_RUNNING flag is set, don't bother trying
to lock the task by setting leaf_p to 0x1 while inserting it in the runqueue
or having it in the tasklet_list, as nobody else will attempt to add it.
Make sure we set TASK_QUEUED in every case before adding the task to the
run queue. task_wakeup() now checks if either TASK_QUEUED or TASK_RUNNING
is set, and if neither is set, add TASK_QUEUED and effectively add the task
to the runqueue.
No longer use __task_wakeup() anywhere except in task_wakeup(), always use
task_wakeup() instead.
With the old code, process_runnable_task() may re-add a task in the runqueue
without setting the TASK_QUEUED flag, and there were race conditions that could
lead to a task having the TASK_QUEUED flag but not in the runqueue, thus
being unschedulable.
This should be backported to 1.9.
As for smp_prefetch_http(), there is now a way to successfully perform a
prefetch in HTX, even if the message forwarding already begun. It is used for
the sample fetches "req.proto_http" and "method".
This patch must be backported to 1.9.
All HTTP samples are buggy because the channel tested in the prefetch functions
(HTX and legacy HTTP) is chosen depending on the sample direction and not the
keyword really used. It means the request channel is used if the sample is
called during the request analysis and the response channel is used if it is
called during the response analysis, regardless the sample really called. For
instance, if you use the sample "req.ver" in an http-response rule, the response
channel will be prefeched because it is called during the response analysis,
while the request channel should have been used instead. So some assumptions on
the validity of the sample may be made on the wrong channel. It is the first
bug.
Then the same error is done in some samples themselves. So fetches are performed
on the wrong channel. For instance, the header extraction (req.fhdr, res.fhdr,
req.hdr, res.hdr...). If the sample "req.hdr" is used in an http-response rule,
then the matching is done on the response headers and not the request ones. It
is the second bug.
Finally, the last one but not the least, in some samples, the right channel is
used. But because the prefetch was done on the wrong one, this channel may be in
a undefined state. For instance, using the sample "req.ver" in an http-response
rule leads to a matching on a posibility released buffer.
To fix all these bugs, the right channel is now chosen in sample fetches, before
the prefetch. If the same function is used to fetch requests and responses
elements, then the keyword is used to choose the right one. This channel is then
used by the functions smp_prefetch_htx() and smp_prefetch_http(). Of course, it
is also used by the samples themselves to extract information.
This patch must be backported to all supported versions. For version 1.8 and
priors, it must be totally refactored. First because there is no HTX into these
versions. Then the buffers API has changed in HAProxy 1.9. The files
http_fetch.{ch} doesn't exist on old versions.
Maksim Kupriianov reported very strange crashes in fwrr_update_position()
which didn't make sense because of an apparent divide overflow except that
the value was not null in the core.
It happens that while the locking is correct in all the functions' call
graph, the uppermost one (fwrr_get_next_server()) incorrectly expected
that its target server was already locked when called. This stupid
assumption causd the server lock not to be held when calling the other
ones, explaining how it was possible to change the server's eweight by
calling srv_lb_commit_status() under the server lock yet collide with
its unprotected usage.
This commit makes sure that fwrr_get_server_from_group() retrieves a
locked server and that fwrr_get_next_server() is responsible for
unlocking the server before returning it. There is one subtlety in
this function which is that it builds a list of avoided servers that
were full while scanning the tree, and all of them are queued in a
full state so they must be unlocked upon return.
Many thanks to Maksim for providing detailed info allowing to narrow
down this bug.
This fix must be backported to 1.9. In 1.8 the lock seems much wider
and changes to the server's state are performed under the rendez-vous
point so this it doesn't seem possible that it happens there.
Older compilers don't like to see "inline" placed after the type in a
function declaration, it must be "static inline <type>" only. This
patch touches various areas. The warnings were seen with gcc-3.4.
This one hasn't been used anymore since the scheduler changes after 1.8
but it kept being exported and maintained up to date while it's always
reset when scanning the trees. Let's stop exporting it and updating it.
This function will handle mux upgrades, for frontend connections only. It will
retrieve the best mux in the same way than conn_install_mux_fe except that the
mode and optionnally the proto are forced.
The new multiplexer is initialized using a new context and a specific input
buffer. Then, the old one is destroyed. If an error occurred, everything is
rolled back.
It is mandatory to handle mux upgrades, because during a mux upgrade, the
connection will be reassigned to another multiplexer. So when the old one is
destroyed, it does not own the connection anymore. Or in other words, conn->ctx
does not point to the old mux's context when its destroy() callback is
called. So we now rely on the multiplexer context do destroy it instead of the
connection.
In addition, h1_release() and h2_release() have also been updated in the same
way.
The mux's callback init() now take a pointer to a buffer as extra argument. It
must be used by the multiplexer as its input buffer. This buffer is always NULL
when a multiplexer is initialized with a fresh connection. But if a mux upgrade
is performed, it may be filled with existing data. Note that, for now, mux
upgrades are not supported. But this commit is mandatory to do so.
Because the option http-tunnel is now ignored in HTX, there is no longer any
need to adjust the transaction mode in HTX analyzers. A channel can still be
switch to the tunnel mode for legitimate cases (HTTP CONNECT or switching
protocols). So the function htx_adjust_conn_mode() is now useless.
This patch must be backported to 1.9. It is not strictly speaking required but
it will ease futur backports.
This patch implements the external binary support in the master worker.
To configure an external process, you need to use the program section,
for example:
program dataplane-api
command ./dataplane_api
Those processes are launched at the same time as the workers.
During a reload of HAProxy, those processes are dealing with the same
sequence as a worker:
- the master is re-executed
- the master sends a USR1 signal to the program
- the master launches a new instance of the program
During a stop, or restart, a SIGTERM is sent to the program.
The children variable is still used in haproxy, it is not required
anymore since we have the information about the current workers in the
mworker_proc linked list.
The oldpids array is also replaced by this linked list when we
generated the arguments for the master reexec.
I found on an (old) AIX 5.1 machine that stdint.h didn't exist while
inttypes.h which is expected to include it does exist and provides the
desired functionalities.
As explained here, stdint being just a subset of inttypes for use in
freestanding environments, it's probably always OK to switch to inttypes
instead:
https://pubs.opengroup.org/onlinepubs/009696799/basedefs/stdint.h.html
Also it's even clearer here in the autoconf doc :
https://www.gnu.org/software/autoconf/manual/autoconf-2.61/html_node/Header-Portability.html
"The C99 standard says that inttypes.h includes stdint.h, so there's
no need to include stdint.h separately in a standard environment.
Some implementations have inttypes.h but not stdint.h (e.g., Solaris
7), but we don't know of any implementation that has stdint.h but not
inttypes.h"
Previous commit 3ea351368 ("BUG/MEDIUM: h2: Remove the tasklet from the
task list if unsubscribing.") uncovered an issue which needs to be
addressed in the scheduler's API. The function task_remove_from_task_list()
was initially designed to remove a task from the running tasklet list from
within the scheduler, and had to be used in h2 to abort pending I/O events.
However this function was not designed to be idempotent, occasionally
causing a double removal from the tasklet list, with the second doing
nothing but affecting the apparent tasks count and making haproxy use
100% CPU on some tests consisting in stopping the client during some
transfers. The h2_unsubscribe() function can sometimes be called upon
stream exit after an error where the tasklet was possibly already
removed, so it.
This patch does 2 things :
- it renames task_remove_from_task_list() to
__task_remove_from_tasklet_list() to discourage users from calling
it. Also note the fix in the naming since it's a tasklet list and
not a task list. This function is still uesd from the scheduler.
- it adds a new, idempotent, task_remove_from_tasklet_list() function
which does nothing if the task is already not in the tasklet list.
This patch will need to be backported where the commit above is backported.
It's never easy to guess what services are built in. We currently have
the prometheus exporter in contrib/ which is the only extension for now.
Let's enumerate all available ones just like we do for filterr and pollers.
The status codes definition (STAT_STATUS_*) and their string representation
stat_status_codes) have been moved in stats files. There is no reason to keep
them in proto_http files.
In task_unlink_rq, to decide if we should logk the global runqueue lock,
use the TASK_GLOBAL flag instead of relying on t->thread_mask being tid_bit,
as it could be so while still being in the global runqueue if another thread
woke that task for us.
This should be backported to 1.9.
In this function we end up with successive locked operations then a
store barrier, and in addition the compiler has to emit less efficient
code due to a longer jump. There's no need for absolutely updating the
tasks_run_queue counter before clearing the task's leaf pointer, so
let's swap the two operations and benefit from a single barrier as much
as possible. This code is on the hot path and shows about half a percent
of improvement with 8 threads.
By using LIST_DEL_INIT() instead of LIST_DEL()+LIST_INIT() we manage
to bump the peak connection rate by no less than 3% on 8 threads.
The perf top profile shows much less contention in this area which
suffered from the second reload.
We move the code responsible of parsing protocol buffers messages
inside gRPC messages from sample.c to include/proto/protocol_buffers.h
so that to reuse it to cascade "ungrpc" converter.
For now on, "ungrpc" may take a second optional argument to provide
the protocol buffers types used to encode the field value to be extracted.
When absent the field value is extracted as a binary sample which may then
followed by others converters like "hex" which takes binary as input sample.
When this second argument is a type which does not match the one found by "ungrpc",
this field is considered as not found even if present.
With this patch we also remove the useless "varint" and "svarint" converters.
Update the documentation about "ungrpc" converters.
Parsing protocol buffer fields always consists in skip the field
if the field is not found or store the field value if found.
So, with this patch we factorize a little bit the code for "ungrpc" converter.
Historically the default frontend's maxconn used to be quite low (2000),
which was sufficient two decades ago but often proved to be a problem
when users had purposely set the global maxconn value but forgot to set
the frontend's.
There is no point in keeping this arbitrary limit for frontends : when
the global maxconn is lower, it's already too high and when the global
maxconn is much higher, it becomes a limiting factor which causes trouble
in production.
This commit allows the value to be set to zero, which becomes the new
default value, to mean it's not directly limited, or in fact it's set
to the global maxconn. Since this operation used to be performed before
computing a possibly automatic global maxconn based on memory limits,
the calculation of the maxconn value and its propagation to the backends'
fullconn has now moved to a dedicated function, proxy_adjust_all_maxconn(),
which is called once the global maxconn is stabilized.
This comes with two benefits :
1) a configuration missing "maxconn" in the defaults section will not
limit itself to a magically hardcoded value but will scale up to the
global maxconn ;
2) when the global maxconn is not set and memory limits are used instead,
the frontends' maxconn automatically adapts, and the backends' fullconn
as well.
In an attempt to try to provide automatic maxconn settings, we need to
decorrelate a listner's backlog and maxconn so that these values can be
independent. This introduces a listener_backlog() function which retrieves
the backlog value from the listener's backlog, the frontend's, the
listener's maxconn, the frontend's or falls back to 1024. This
corresponds to what was done in cfgparse.c to force a value there except
the last fallback which was not set since the frontend's maxconn is always
known.
Since LIST_DEL_LOCKED() and LIST_POP_LOCKED() now automatically reinitialize
the removed element, there's no need for keeping this LIST_INIT() call in the
idle connection code.
There is one point where we can migrate a connection to another thread
without taking risk, it's when we accept it : the new FD is not yet in
the fd cache and no task was created yet. It's still possible to assign
it a different thread than the one which accepted the connection. The
only requirement for this is to have one accept queue per thread and
their respective processing tasks that have to be woken up each time
an entry is added to the queue.
This is a multiple-producer, single-consumer model. Entries are added
at the queue's tail and the processing task is woken up. The consumer
picks entries at the head and processes them in order. The accept queue
contains the fd, the source address, and the listener. Each entry of
the accept queue was rounded up to 64 bytes (one cache line) to avoid
cache aliasing because tests have shown that otherwise performance
suffers a lot (5%). A test has shown that it's important to have at
least 256 entries for the rings, as at 128 it's still possible to fill
them often at high loads on small thread counts.
The processing task does almost nothing except calling the listener's
accept() function and updating the global session and SSL rate counters
just like listener_accept() does on synchronous calls.
At this point the accept queue is implemented but not used.
In order to quickly pick a thread ID when accepting a connection, we'll
need to know certain pre-computed values derived from the thread mask,
which are counts of bits per position multiples of 1, 2, 4, 8, 16 and
32. In practice it is sufficient to compute only the 4 first ones and
store them in the bind_conf. We update the count every time the
bind_thread value is adjusted.
The fields in the bind_conf struct have been moved around a little bit
to make it easier to group all thread bit values into the same cache
line.
The function used to return a thread number is bind_map_thread_id(),
and it maps a number between 0 and 31/63 to a thread ID between 0 and
31/63, starting from the left.
Instead of having one task per thread and per server that does clean the
idling connections, have only one global task for every servers.
That tasks parses all the servers that currently have idling connections,
and remove half of them, to put them in a per-thread list of connections
to kill. For each thread that does have connections to kill, wake a task
to do so, so that the cleaning will be done in the context of said thread.
Use the locked macros when manipulating idle_orphan_conns, so that other
threads can remove elements from it.
It will be useful later to avoid having a task per server and per thread to
cleanup the orphan list.
This patch implements "req.ungrpc" sample fetch method to decode and
parse a gRPC request. It takes only one argument: a protocol buffers
field number to identify the protocol buffers message number to be looked up.
This argument is a sort of path in dotted notation to the terminal field number
to be retrieved.
ex:
req.ungrpc(1.2.3.4)
This sample fetch catch the data in raw mode, without interpreting them.
Some protocol buffers specific converters may be used to convert the data
to the correct type.
It is the HTX version of co_skip(). Internally, It uses the function htx_drain().
It will be used by other commits to fix bugs, so it must be backported to 1.9.
in co_skip(), the flag CF_WRITE_PARTIAL is set on the channel. The flag
CF_WROTE_DATA must also be set to notify the channel some data were sent.
This patch must be backported to 1.9.
Calculate if the fd or task should be locked once, before locking, and
reuse the calculation when determing when to unlock.
Fixes a race condition added in 87d54a9a for fds, and b20aa9ee for tasks,
released in 1.9-dev4. When one thread modifies thread_mask to be a single
thread for a task or fd while a second thread has locked or is waiting on a
lock for that task or fd, the second thread will not unlock it. For FDs,
this is observable when a listener is polled by multiple threads, and is
closed while those threads have events pending. For tasks, this seems
possible, where task_set_affinity is called, but I did not observe it.
This must be backported to 1.9.
This is a naive implementation of closefrom() which closes all FDs
starting from the one passed in argument. closefrom() is not provided
on all operating systems, and other versions will follow.
Add a per-thread counter of idling connections, and use it to determine
how many connections we should kill after the timeout, instead of using
the global counter, or we're likely to just kill most of the connections.
This should be backported to 1.9.
Use atomic operations when dealing with srv->curr_idle_conns, as it's shared
between threads, otherwise we could get inconsistencies.
This should be backported to 1.9.
Initialize ->srv peer field for all the peers, the local peer included.
Indeed, a haproxy process needs to connect to the local peer of a remote
process. Furthermore, when a "peer" or "server" line is parsed by parse_server()
the address must be copied to ->addr field of the peer object only if this address
has been also parsed by parse_server(). This is not the case if this address belongs
to the local peer and is provided on a "server" line.
After having parsed the "peer" or "server" lines of a peer
sections, the ->srv part of all the peer must be initialized for SSL, if
enabled. Same thing for the binding part.
Revert 1417f0b commit which is no more required.
No backport is needed, this is purely 2.0.
Commit 1055e687a ("MINOR: peers: Make outgoing connection to SSL/TLS
peers work.") introduced an "srv" field in the peers, which points to
the equivalent server to hold SSL settings. This one is not set when
the peer is local so we must always test it before testing p->srv->use_ssl
otherwise haproxy dies during reloads.
No backport is needed, this is purely 2.0.
This also depends on the nbthread count, so it must only be performed after
parsing the whole config file. As a side effect, this removes some code
duplication between servers and server-templates.
This must be backported to 1.9.
With variable connection limits, it's not possible to accurately determine
whether the mux is still in use by comparing usage and max to be equal due
to the fact that one determines the capacity and the other one takes care
of the context. This can cause some connections to be dropped before they
reach their stream ID limit.
It seems it could also cause some connections to be terminated with
streams still alive if the limit was reduced to match the newly computed
avail_streams() value, though this cannot yet happen with existing muxes.
Instead let's switch to usage reports and simply check whether connections
are both unused and available before adding them to the idle list.
This should be backported to 1.9.
The new flag SI_FL_KILL_CONN is now set by the rare actions which
deliberately want the whole connection (and not just the stream) to be
killed. This is only used for "tcp-request content reject",
"tcp-response content reject", "tcp-response content close" and
"http-request reject". The purpose is to desambiguate the close from
a regular shutdown. This will be used by the next patches.
If we're adding a connection to the server orphan idle list, don't forget
to remove the CO_FL_SESS_IDLE flag, or we will assume later it's still
attached to a session.
This should be backported to 1.9.
There's a very small but existing uncertainty window when waking another
thread up where it is possible for task_wakeup() not to wake the other
task up because it's still running while this once is in the process of
finishing and loses its TASK_RUNNING flag. In this case the wakeup will
be missed.
The problem is that we have a single flag to store 3 states, since the
transition from running to sleeping isn't atomic. Thus we need to have
another flag to cover this part. This patch introduces TASK_QUEUED to
mention that the task is already in the run queue, running or not. This
bit will be removed while TASK_RUNNING is kept once dequeued, and will
be used when removing TASK_RUNNING to check if the task has been requeued.
It might be possible to slightly improve this but the occurrence rate
is quite low and we don't really need to complexify the scheduler to
optimize for a rare case.
The impact with the current code is very low since we have few inter-
thread wakeups. Most of them are caused by checks killing sessions.
This must be backported to 1.9.
Before the first send() attempt, we should be in SI_ST_CON, not
SI_ST_EST, since we have not yet attempted to send and we are
allowed to retry. This is particularly important with complex
outgoing muxes which can fail during the first send attempt (e.g.
failed stream ID allocation).
It only requires that sess_update_st_con_tcp() knows about this
possibility, as we must not forcefully close a reused connection
when facing an error in this case, this will be handled later.
This may be backported to 1.9 with care after some observation period.
Make "bind" keywork be supported in "peers" sections.
All "bind" settings are supported on this line.
Add "default-bind" option to parse the binding options excepted the bind address.
Do not parse anymore the bind address for local peers on "server" lines.
Do not use anymore list_for_each_entry() to set the "peers" section
listener parameters because there is only one listener by "peers" section.
May be backported to 1.5 and newer.
This patch adds pointer to a struct server to peer structure which
is initialized after having parsed a remote "peer" line.
After having parsed all peers section we run ->prepare_srv to initialize
all SSL/TLS stuff of remote perr (or server).
Remaining thing to do to completely support peer protocol over SSL/TLS:
make "bind" keyword be supported in "peers" sections to make SSL/TLS
incoming connections to local peers work.
May be backported to 1.5 and newer.
When using the peers feature a race condition could prevent
a connection from being properly counted. When this connection
exits it is being "uncounted" nonetheless, leading to a possible
underflow (-1) of the conn_curr stick table entry in the following
scenario :
- Connect to peer A (A=1, B=0)
- Peer A sends 1 to B (A=1, B=1)
- Kill connection to A (A=0, B=1)
- Connect to peer B (A=0, B=2)
- Peer A sends 0 to B (A=0, B=0)
- Peer B sends 0/2 to A (A=?, B=0)
- Kill connection to B (A=?, B=-1)
- Peer B sends -1 to A (A=-1, B=-1)
This fix may be backported to all supported branches.
Openssl switched from aes128 to aes256 since may 2016 to compute
tls ticket secrets used by default. But Haproxy still handled only
128 bits keys for both tls key file and CLI.
This patch permit the user to set aes256 keys throught CLI or
the key file (80 bytes encoded in base64) in the same way that
aes128 keys were handled (48 bytes encoded in base64):
- first 16 bytes for the key name
- next 16/32 bytes for aes 128/256 key bits key
- last 16/32 bytes for hmac 128/256 bits
Both sizes are now supported (but keys from same file must be
of the same size and can but updated via CLI only using a key of
the same size).
Note: This feature need the fix "dec func ignores padding for output
size checking."
