When queuing a timer, it's very likely that an expiration date is
equal to that of the previously queued timer, due to time rounding
to the millisecond. Optimizing for this case provides a noticeable
1% performance boost.
The run queue scheduler now considers task->nice to queue a task and
to pick a task out of the queue. This makes it possible to boost the
access to statistics (both via HTTP and UNIX socket). The UNIX socket
receives twice as much a boost as the HTTP socket because it is more
sensible.
We now insert tasks in a certain sequence in the run queue.
The sorting key currently is the arrival order. It will now
be possible to apply a "nice" value to any task so that it
goes forwards or backwards in the run queue.
The calls to wake_expired_tasks() and maintain_proxies()
have been moved to the main run_poll_loop(), because they
had nothing to do in process_runnable_tasks().
The task_wakeup() function is not inlined anymore, as it was
only used at one place.
The qlist member of the task structure has been removed now.
The run_queue list has been replaced for an integer indicating
the number of tasks in the run queue.
The ultree code has been removed in favor of a simpler and
cleaner ebtree implementation. The eternity queue does not
need to exist anymore, and the pool_tree64 has been removed.
The ebtree node is stored in the task itself. The qlist list
header is still used by the run-queue, but will be able to
disappear once the run-queue uses ebtree too.
The dequeuing logic was completely wrong. First, a task was assigned
to all servers to process the queue, but this task was never scheduled
and was only woken up on session free. Second, there was no reservation
of server entries when a task was assigned a server. This means that
as long as the task was not connected to the server, its presence was
not accounted for. This was causing trouble when detecting whether or
not a server had reached maxconn. Third, during a redispatch, a session
could lose its place at the server's and get blocked because another
session at the same moment would have stolen the entry. Fourth, the
redispatch option did not work when maxqueue was reached for a server,
and it was not possible to do so without indefinitely hanging a session.
The root cause of all those problems was the lack of pre-reservation of
connections at the server's, and the lack of tracking of servers during
a redispatch. Everything relied on combinations of flags which could
appear similarly in quite distinct situations.
This patch is a major rework but there was no other solution, as the
internal logic was deeply flawed. The resulting code is cleaner, more
understandable, uses less magics and is overall more robust.
As an added bonus, "option redispatch" now works when maxqueue has
been reached on a server.
New functions implemented:
- deinit_pollers: called at the end of deinit())
- prune_acl: called via list_for_each_entry_safe
Add missing pool_destroy2 calls:
- p->hdr_idx_pool
- pool2_tree64
Implement all task stopping:
- health-check: needs new "struct task" in the struct server
- queue processing: queue_mgt
- appsess_refresh: appsession_refresh
before (idle system):
==6079== LEAK SUMMARY:
==6079== definitely lost: 1,112 bytes in 75 blocks.
==6079== indirectly lost: 53,356 bytes in 2,090 blocks.
==6079== possibly lost: 52 bytes in 1 blocks.
==6079== still reachable: 150,996 bytes in 504 blocks.
==6079== suppressed: 0 bytes in 0 blocks.
after (idle system):
==6945== LEAK SUMMARY:
==6945== definitely lost: 7,644 bytes in 137 blocks.
==6945== indirectly lost: 9,913 bytes in 587 blocks.
==6945== possibly lost: 0 bytes in 0 blocks.
==6945== still reachable: 0 bytes in 0 blocks.
==6945== suppressed: 0 bytes in 0 blocks.
before (running system for ~2m):
==9343== LEAK SUMMARY:
==9343== definitely lost: 1,112 bytes in 75 blocks.
==9343== indirectly lost: 54,199 bytes in 2,122 blocks.
==9343== possibly lost: 52 bytes in 1 blocks.
==9343== still reachable: 151,128 bytes in 509 blocks.
==9343== suppressed: 0 bytes in 0 blocks.
after (running system for ~2m):
==11616== LEAK SUMMARY:
==11616== definitely lost: 7,644 bytes in 137 blocks.
==11616== indirectly lost: 9,981 bytes in 591 blocks.
==11616== possibly lost: 0 bytes in 0 blocks.
==11616== still reachable: 4 bytes in 1 blocks.
==11616== suppressed: 0 bytes in 0 blocks.
Still not perfect but significant improvement.
This patch adds two optional arguments "len" and "depth" to
"balance uri". They are used to limit the length in characters
of the analysis, as well as the number of directory components
it applies to.
This patch extends the "url_param" load balancing method by introducing
the "check_post" option. Using this option enables analysis of the beginning
of POST requests to search for the specified URL parameter.
The patch also fixes a few minor typos in comments that were discovered
during code review.
Due to the way the stats socket work, it was not possible to
maintain the information related to the command entered, so
after filling a whole buffer, the request was lost and it was
considered that there was nothing to write anymore.
The major reason was that some flags were passed directly
during the first call to stats_dump_raw() instead of being
stored persistently in the session.
To definitely fix this problem, flags were added to the stats
member of the session structure.
A second problem appeared. When the stats were produced, a first
call to client_retnclose() was performed, then one or multiple
subsequent calls to buffer_write_chunks() were done. But once the
stats buffer was full and a reschedule operated, the buffer was
flushed, the write flag cleared from the buffer and nothing was
done to re-arm it.
For this reason, a check was added in the proto_uxst_stats()
function in order to re-call the client FSM when data were added
by stats_dump_raw(). Finally, the whole unix stats dump FSM was
rewritten to avoid all the magics it depended on. It is now
simpler and looks more like the HTTP one.
The new "leastconn" LB algorithm selects the server which has the
least established or pending connections. The weights are considered,
so that a server with a weight of 20 will get twice as many connections
as the server with a weight of 10.
The algorithm respects the minconn/maxconn settings, as well as the
slowstart since it is a dynamic algorithm. It also correctly supports
backup servers (one and all).
It is generally suited for protocols with long sessions (such as remote
terminals and databases), as it will ensure that upon restart, a server
with no connection will take all new ones until its load is balanced
with others.
A test configuration has been added in order to ease regression testing.
Currently there is a ~16KB limit for a data size passed via unix socket.
It is caused by a trivial bug ttat is going to fixed soon, however
in most cases there is no need to dump a full stats.
This patch makes possible to select a scope of dumped data by extending
current "show stat" to "show stat [<iid> <type> <sid>]":
- iid is a proxy id, -1 to dump all proxies
- type selects type of dumpable objects: 1 for frontend, 2 for backend, 4 for
server, -1 for all types. Values can be ORed, for example:
1+2=3 -> frontend+backend.
1+2+4=7 -> frontend+backend+server.
- sid is a service id, -1 to dump everything from the selected proxy.
To do this I implemented a new session flag (SN_STAT_BOUND), added three
variables in data_ctx.stats (iid, type, sid), modified dumpstats.c and
completely revorked the process_uxst_stats: now it waits for a "\n"
terminated string, splits args and uses them. BTW: It should be quite easy
to add new commands, for example to enable/disable servers, the only problem
I can see is a not very lucky config name (*stats* socket). :|
During the work I also fixed two bug:
- s->flags were not initialized for proto_uxst
- missing comma if throttling not enabled (caused by a stupid change in
"Implement persistent id for proxies and servers")
Other changes:
- No more magic type valuse, use STATS_TYPE_FE/STATS_TYPE_BE/STATS_TYPE_SV
- Don't memset full s->data_ctx (it was clearing s->data_ctx.stats.{iid/type/sid},
instead initialize stats.sv & stats.sv_st (stats.px and stats.px_st were already
initialized)
With all that changes it was extremely easy to write a short perl plugin
for a perl-enabled net-snmp (also included in this patch).
29385 is my PEN (Private Enterprise Number) and I'm willing to donate
the SNMPv2-SMI::enterprises.29385.106.* OIDs for HAProxy if there
is nothing assigned already.
When haproxy decides that session needs to be redispatched it chose a server,
but there is no guarantee for it to be a different one. So, it often
happens that selected server is exactly the same that it was previously, so
a client ends up with a 503 error anyway, especially when one sever has
much bigger weight than others.
Changes from the previous version:
- drop stupid and unnecessary SN_DIRECT changes
- assign_server(): use srvtoavoid to keep the old server and clear s->srv
so SRV_STATUS_NOSRV guarantees that t->srv == NULL (again)
and get_server_rr_with_conns has chances to work (previously
we were passing a NULL here)
- srv_redispatch_connect(): remove t->srv->cum_sess and t->srv->failed_conns
incrementing as t->srv was guaranteed to be NULL
- add avoididx to get_server_rr_with_conns. I hope I correctly understand this code.
- fix http_flush_cookie_flags() and move it to assign_server_and_queue()
directly. The code here was supposed to set CK_DOWN and clear CK_VALID,
but: (TX_CK_VALID | TX_CK_DOWN) == TX_CK_VALID == TX_CK_MASK so:
if ((txn->flags & TX_CK_MASK) == TX_CK_VALID)
txn->flags ^= (TX_CK_VALID | TX_CK_DOWN);
was really a:
if ((txn->flags & TX_CK_MASK) == TX_CK_VALID)
txn->flags &= TX_CK_VALID
Now haproxy logs "--DI" after redispatching connection.
- defer srv->redispatches++ and s->be->redispatches++ so there
are called only if a conenction was redispatched, not only
supposed to.
- don't increment lbconn if redispatcher selected the same sarver
- don't count unsuccessfully redispatched connections as redispatched
connections
- don't count redispatched connections as errors, so:
- the number of connections effectively served by a server is:
srv->cum_sess - srv->failed_conns - srv->retries - srv->redispatches
and
SUM(servers->failed_conns) == be->failed_conns
- requires the "Don't increment server connections too much + fix retries" patch
- needs little more testing and probably some discussion so reverting to the RFC state
Tests #1:
retries 4
redispatch
i) 1 server(s): b (wght=1, down)
b) sessions=5, lbtot=1, err_conn=1, retr=4, redis=0
-> request failed
ii) server(s): b (wght=1, down), u (wght=1, down)
b) sessions=4, lbtot=1, err_conn=0, retr=3, redis=1
u) sessions=1, lbtot=1, err_conn=1, retr=0, redis=0
-> request FAILED
iii) 2 server(s): b (wght=1, down), u (wght=1, up)
b) sessions=4, lbtot=1, err_conn=0, retr=3, redis=1
u) sessions=1, lbtot=1, err_conn=0, retr=0, redis=0
-> request OK
iv) 2 server(s): b (wght=100, down), u (wght=1, up)
b) sessions=4, lbtot=1, err_conn=0, retr=3, redis=1
u) sessions=1, lbtot=1, err_conn=0, retr=0, redis=0
-> request OK
v) 1 server(s): b (down for first 4 SYNS)
b) sessions=5, lbtot=1, err_conn=0, retr=4, redis=0
-> request OK
Tests #2:
retries 4
i) 1 server(s): b (down)
b) sessions=5, lbtot=1, err_conn=1, retr=4, redis=0
-> request FAILED
This patch implements ability to set the current state of one server
by tracking another one. It:
- adds two variables: *tracknext, *tracked to struct server
- implements findserver(), similar to findproxy()
- adds "track" keyword accepting both "proxy/server" and "server" (assuming current proxy)
- verifies if both checks and tracking is not enabled at the same time
- changes set_server_down() to notify tracking server
- creates set_server_up(), set_server_disabled(), set_server_enabled() by
moving the code from process_chk() and adding notifications
- changes stats to show a name of tracked server instead of Chk/Dwn/Dwntime(html)
or by adding new variable (csv)
Changes from the previuos version:
- it is possibile to track independently of the declaration order
- one extra comma bug is fixed
- new condition to check if there is no disable-on-404 inconsistency
GCC4 is stupid (unbelievable news!).
When some code uses __builtin_expect(x != 0, 1), it really performs
the check of x != 0 then tests that the result is not zero! This is
a double check when only one was expected. Some performance drops
of 10% in the HTTP parser code have been observed due to this bug.
GCC 3.4 is fine though.
A solution consists in expecting that the tested value is 1. In
this case, it emits the correct code, but it's still not optimal
it seems. Finally the best solution is to ignore likely() and to
pray for the compiler to emit correct code. However, we still have
to fix unlikely() to remove the test there too, and to fix all
code which passed pointers overthere to pass integers instead.
This patch adds two new variables: fastinter and downinter.
When server state is:
- non-transitionally UP -> inter (no change)
- transitionally UP (going down), unchecked or transitionally DOWN (going up) -> fastinter
- down -> downinter
It allows to set something like:
server sr6 127.0.51.61:80 cookie s6 check inter 10000 downinter 20000 fastinter 500 fall 3 weight 40
In the above example haproxy uses 10000ms between checks but as soon as
one check fails fastinter (500ms) is used. If server is down
downinter (20000) is used or fastinter (500ms) if one check pass.
Fastinter is also used when haproxy starts.
New "timeout.check" variable was added, if set haproxy uses it as an additional
read timeout, but only after a connection has been already established. I was
thinking about using "timeout.server" here but most people set this
with an addition reserve but still want checks to kick out laggy servers.
Please also note that in most cases check request is much simpler
and faster to handle than normal requests so this timeout should be smaller.
I also changed the timeout used for check connections establishing.
Changes from the previous version:
- use tv_isset() to check if the timeout is set,
- use min("timeout connect", "inter") but only if "timeout check" is set
as this min alone may be to short for full (connect + read) check,
- debug code (fprintf) commented/removed
- documentation
Compile tested only (sorry!) as I'm currently traveling but changes
are rather small and trivial.
Due to the way Linux delivers EPOLLIN and EPOLLHUP, a closed connection
received after some server data sometimes results in truncated responses
if the client disconnects before server starts to respond. The reason
is that the EPOLLHUP flag is processed as an indication of end of
transfer while some data may remain in the system's socket buffers.
This problem could only be triggered with sepoll, although nothing should
prevent it from happening with normal epoll. In fact, the work factoring
performed by sepoll increases the risk that this bug appears.
The fix consists in making FD_POLL_HUP and FD_POLL_ERR sticky and that
they are only checked if FD_POLL_IN is not set, meaning that we have
read all pending data.
That way, the problem is definitely fixed and sepoll still remains about
17% faster than epoll since it can take into account all information
returned by the kernel.
The source address selection for health checks did not consider
the new transparent proxy method. Rely on the same unified function
as the other connect() calls.
This patch also fixes a bug by which the proxy's source address was
ignored if cttproxy was used.
It is sometimes required to know some informations such as the
process uptime when consulting statistics. This patch adds the
"show info" command to query those informations on the UNIX
socket.
A new "timeout" keyword replaces old "{con|cli|srv}timeout", and
provides the ability to independantly set the following timeouts :
- client
- tarpit
- queue
- connect
- server
- appsession
Additionally, the "clitimeout", "contimeout" and "srvtimeout" values
are supported but deprecated. No warning is emitted yet when they are
used since the option is very new.
Other timeouts should follow soon now.
This round robin algorithm was written from trees, so that we
do not have to recompute any table when changing server weights.
This solution allows on-the-fly weight adjustments with immediate
effect on the load distribution.
There is still a limitation due to 32-bit computations, to about
2000 servers at full scale (weight 255), or more servers with
lower weights. Basically, sum(srv.weight)*4096 must be below 2^31.
Test configurations and an example program used to develop the
tree will be added next.
Many changes have been brought to the weights computations and
variables in order to accomodate for the possiblity of a server to
be running but disabled from load balancing due to a null weight.
Under some circumstances, it will be useful to be able to have
a server's effective weight bigger than the user weight, and this
is particularly true for dynamic weight-based algorithms. In order
to support this, we add a "wdiv" member to the lbprm structure
which will always be used to divide the weights before reporting
them.
Since the introduction of server weights, all load balancing algorithms
relied on a pre-computed map. Incidently, quite a bunch of map-specific
parameters were used at random places in order to get the number of
servers or their total weight. It was not architecturally acceptable
that optimizations for the map computation had impact on external parts.
For instance, during this cleanup it was found that a backend weight was
seen as 1 when only the first backup server is used, whatever its weight.
This cleanup consists in differentiating between LB-generic parameters,
such as total weights, number of servers, etc... and map-specific ones.
The struct proxy has been enhanced in order to make it easier to later
support other algorithms. The recount_servers() function now also
updates generic values such as total weights so that it's not needed
anymore to call recalc_server_map() when weights are needed. This
permitted to simplify some code which does not need to know about map
internals anymore.
By default, counters used for statistics calculation are incremented
only when a session finishes. It works quite well when serving small
objects, but with big ones (for example large images or archives) or
with A/V streaming, a graph generated from haproxy counters looks like
a hedgehog.
This patch implements a contstats (continous statistics) option.
When set counters get incremented continuously, during a whole session.
Recounting touches a hotpath directly so it is not enabled by default,
as it has small performance impact (~0.5%).
Proxy listeners were very special and not very easy to manipulate.
A proto_tcp file has been created with all that is required to
manage TCPv4/TCPv6 as raw protocols, and provide generic listeners.
The code of start_proxies() and maintain_proxies() now looks less
like spaghetti. Also, event_accept will need a serious lifting in
order to use more of the information provided by the listener.
There was a missing state for listeners, when they are not listening
but still attached to the protocol. The LI_ASSIGNED state was added
for this purpose. This permitted to clean up the assignment/release
workflow quite a bit. Generic enable/enable_all/disable/disable_all
primitives were added, and a disable_all entry was added to the
struct protocol.
It's not easy to report useful information to help the user quickly
fix a configuration. This patch :
- removes the word "listener" in favor of "proxy" as it has been
used since the beginning ;
- ensures that the same function (hence the same words) will be
used to report capabilities of a proxy being declared and an
existing proxy ;
- avoid the term "conflicting capabilities" in favor of "overlapping
capabilities" which is more exact.
- just report that the same name is reused in case of warnings
This patch:
- adds proxy_mode_str() similar to proxy_type_str()
- adds a generic findproxy function used with default_backend/setbe/use_backed
- rewrite default_backend/senbe/use_backed to use introduced findproxy()
- relaxes duplicated proxy check
- changes capabilities displaying from "%X" to "%s" with a call to proxy_type_str()
A new function "backend_parse_balance" has been created in backend.c,
which is dedicated to the parsing of the "balance" keyword. It will
provide easier methods for adding new algorithms.
Hello,
This patch implements new statistics for SLA calculation by adding new
field 'Dwntime' with total down time since restart (both HTTP/CSV) and
extending status field (HTTP) or inserting a new one (CSV) with time
showing how long each server/backend is in a current state. Additionaly,
down transations are also calculated and displayed for backends, so it is
possible to know how many times selected backend was down, generating "No
server is available to handle this request." error.
New information are presentetd in two different ways:
- for HTTP: a "human redable form", one of "100000d 23h", "23h 59m" or
"59m 59s"
- for CSV: seconds
I believe that seconds resolution is enough.
As there are more columns in the status page I decided to shrink some
names to make more space:
- Weight -> Wght
- Check -> Chk
- Down -> Dwn
Making described changes I also made some improvements and fixed some
small bugs:
- don't increment s->health above 's->rise + s->fall - 1'. Previously it
was incremented an then (re)set to 's->rise + s->fall - 1'.
- do not set server down if it is down already
- do not set server up if it is up already
- fix colspan in multiple places (mostly introduced by my previous patch)
- add missing "status" header to CSV
- fix order of retries/redispatches in server (CSV)
- s/Tthen/Then/
- s/server/backend/ in DATA_ST_PX_BE (dumpstats.c)
Changes from previous version:
- deal with negative time intervales
- don't relay on s->state (SRV_RUNNING)
- little reworked human_time + compacted format (no spaces). If needed it
can be used in the future for other purposes by optionally making "cnt"
as an argument
- leave set_server_down mostly unchanged
- only little reworked "process_chk: 9"
- additional fields in CSV are appended to the rigth
- fix "SEC" macro
- named arguments (human_time, be_downtime, srv_downtime)
Hope it is OK. If there are only cosmetic changes needed please fill free
to correct it, however if there are some bigger changes required I would
like to discuss it first or at last to know what exactly was changed
especially since I already put this patch into my production server. :)
Thank you,
Best regards,
Krzysztof Oledzki
Removed old unused MODE_LOG and MODE_STATS, and replaced the "stats"
keyword in the global section. The new "stats" keyword in the global
section is used to create a UNIX socket on which the statistics will
be accessed. The client must issue a "show stat\n" command in order
to get a CSV-formated output similar to the output on the HTTP socket
in CSV mode.
A unix socket can now access the statistics. It currently only
recognizes the "show stat\n" command at the beginning of the
input, then returns the statistics in CSV format.
It is now possible to get CSV ouput from the statistics by
simply appending ";csv" to the HTTP request sent to get the
stats. The fields keep the same ordering as in the HTML page,
and a field "pxname" has been prepended at the beginning of
the line.
A new file, proto_uxst.c, implements support of PF_UNIX sockets
of type SOCK_STREAM. It relies on generic stream_sock_read/write
and uses its own accept primitive which also tries to be generic.
Right now it only implements an echo service in sight of a general
support for start dumping via unix socket. The echo code is more
of a proof of concept than useful code.
A new generic protocol mechanism has been added. It provides
an easy method to implement new protocols with different
listeners (eg: unix sockets).
The listeners are automatically started at the right moment
and enabled after the possible fork().
When one server appears at the same position in multiple backends, it
receives all the checks from all the backends exactly at the same time
because the health-checks are only spread within a backend but not
globally.
Attached patch implements per-server start delay in a different way.
Checks are now spread globally - not locally to one backend. It also makes
them start faster - IMHO there is no need to add a 'server->inter' when
calculating first execution. Calculation were moved from cfgparse.c to
checks.c. There is a new function start_checks() and now it is not called
when haproxy is started in MODE_CHECK.
With this patch it is also possible to set a global 'spread-checks'
parameter. It takes a percentage value (1..50, probably something near
5..10 is a good idea) so haproxy adds or removes that many percent to the
original interval after each check. My test shows that with 18 backends,
54 servers total and 10000ms/5% it takes about 45m to mix them completely.
I decided to use rand/srand pseudo-random number generator. I am aware it
is not recommend for a good randomness but a) we do not need a good random
generator here b) it is probably the most portable one.
When a very large number of servers is configured (thousands),
shutting down many of them at once could lead to large number
of calls to recalc_server_map() which already takes some time.
This would result in an O(N^3) computation time, leading to
noticeable pauses on slow embedded CPUs on test platforms.
Instead, mark the map as dirty and recalc it only when needed.
Those ACLs are sometimes useful for troubleshooting. Two ACL subjects
"always_true" and "always_false" have been added too. They return what
their subject says for every pattern. Also, acl_match_pst() has been
removed.
Some fetches such as 'line' or 'hdr' need to know the direction of
the test (request or response). A new 'dir' parameter is now
propagated from the caller to achieve this.
ACLs now support operators such as 'eq', 'le', 'lt', 'ge' and 'gt'
in order to give more flexibility to the language. Because of this
change, the 'dst_limit' keyword changed to 'dst_conn' and now requires
either a range or a test such as 'dst_conn lt 1000' which is more
understandable.
A second occurrence of read-timeout rearming was present in stream_sock.c.
To fix the problem, it was necessary to put the shutdown information in
the buffer (already planned).
