Implement the scheme-based uri normalization as described in rfc3986
6.3.2. Its purpose is to remove the port of an uri if the default one is
used according to the uri scheme : 80/http and 443/https. All other
ports are not touched.
This method uses an htx message as an input. It requires that the target
URI is in absolute-form with a http/https scheme. This represents most
of h2 requests except CONNECT. On the contrary, most of h1 requests
won't be elligible as origin-form is the standard case.
The normalization is first applied on the target URL of the start line.
Then, it is conducted on every Host headers present, assuming that they
are equivalent to the target URL.
This change will be notably useful to not confuse users who are
accustomed to use the host for routing without specifying default ports.
This problem was recently encountered with Firefox which specify the 443
default port for http2 websocket Extended CONNECT.
gcc 8.3.0 spews a bunch of:
src/stick_table.c: In function 'action_inc_gpc0':
include/haproxy/freq_ctr.h:66:12: warning: 'period' may be used uninitialized in this function [-Wmaybe-uninitialized]
curr_tick += period;
^~
src/stick_table.c:2241:15: note: 'period' was declared here
unsigned int period;
^~~~~~
but they're incorrect because all accesses are guarded by the exact same
condition (ptr1 not being null), it's just the compiler being overzealous
about the uninitialized detection that seems to be stronger than its
ability to follow its own optimizations. This code path is not critical,
let's just pre-initialize the period to zero.
No backport is needed.
After reloading HAProxy, the old process may still hold active sessions.
Currently there is no way to gather information, how many sessions such
a process still holds. This patch will not exclude disabled proxies from
stats output when they hold at least one active session. This will allow
sending `!@<PID> show stat` through a master socket to the disabled
process and have it returning its stats data.
This reverts commit 19bbbe0562.
For now, set-src/set-src-port actions are directly performed on the client
connection. Using these actions at the stream level is really a problem with
HTTP connection (See #90) because all requests are affected by this change
and not only the current request. And it is worse with the H2, because
several requests can set their source address into the same connection at
the same time.
It is already an issue when these actions are called from "http-request"
rules. It is safer to wait a bit before adding the support to "tcp-request
content" rules. The solution is to be able to set src/dst address on the
stream and not on the connection when the action if performed from the L7
level..
Reverting the above commit means the issue #1303 is no longer fixed.
This patch must be backported in all branches containing the above commit
(as far as 2.0 for now).
A server name was displayed as <srv>/<proxy> instead of the reverse.
It only confuses diagnostics. This was introduced by commit 7a4a0ac71
("MINOR: cli: add a new "show fd" command") so this fix can be backport
down to 1.8.
As shown in issue #1251, it is possible for a connect() to report an
error directly via the poller without ever reporting send readiness,
but currentlt sock_conn_check() manages to ignore that situation,
leading to high CPU usage as poll() wakes up on these FDs.
The bug was apparently introduced in 1.5-dev22 with commit fd803bb4d
("MEDIUM: connection: add check for readiness in I/O handlers"), but
was likely only woken up by recent changes to conn_fd_handler() that
made use of wakeups instead of direct calls between 1.8 and 1.9,
voiding any chance to catch such errors in the early recv() callback.
The exact sequence that leads to this situation remains obscure though
because the poller does not report send readiness nor does it report an
error. Only HUP and IN are reported on the FD. It is also possible that
some recent kernel updates made this condition appear while it never
used to previously.
This needs to be backported to all stable branches, at least as far
as 2.0. Before 2.2 the code was in tcp_connect_probe() in proto_tcp.c.
This patch makes the use of 'gpc' excluding the use of the legacy
types 'gpc0' and 'gpc1" on the same table.
It also makes the use of 'gpc_rate' excluding the use of the legacy
types 'gpc0_rate' and 'gpc1_rate" on the same table.
The 'gpc0' and 'gpc1' related fetches and actions will apply
to the first two elements of the 'gpc' array if stored in table.
The 'gpc0_rate' and 'gpc1_rate' related fetches and actions will apply
to the first two elements of the 'gpc_rate' array if stored in table.
This patch adds the definition of two new array data_types:
'gpc': This is an array of 32bits General Purpose Counters.
'gpc_rate': This is an array on increment rates of General Purpose Counters.
Like for all arrays, they are limited to 100 elements.
This patch also adds actions and fetches to handle
elements of those arrays.
Note: As documented, those new actions and fetches won't
apply to the legacy 'gpc0', 'gpc1', 'gpc0_rate' nor 'gpc1_rate'.
This patch makes the use of 'gpt' excluding the use of the legacy
type 'gpt0' on the same table.
It also makes the 'gpt0' related fetches and actions applying
to the first element of the 'gpt' array if stored in table.
This patch adds the definition of a new array data_type
'gpt'. This is an array of 32bits General Purpose Tags.
Like for all arrays, it is limited to 100 elements.
This patch also adds actions and fetches to handle
elements of this array.
Note: As documented, those new actions and fetches won't
apply to the legacy 'gpt0' data type.
The store type 'gpt0' was present in code but was not documented.
The patch fix this and should be backported since 'gpt0' is supported.
[wt: ~1.6-dev4 hence all stable]
This patch adds support of array data_types on the peer protocol.
The table definition message will provide an additionnal parameter
for array data-types: the number of elements of the array.
In case of array of frqp it also provides a second parameter:
the period used to compute freq counter.
The array elements are std_type values linearly encoded in
the update message.
Note: if a remote peer announces an array data_type without
parameters into the table definition message, all updates
on this table will be ignored because we can not
parse update messages consistently.
This patch provides the code to handle arrays of some
standard types (SINT, UINT, ULL and FRQP) in stick table.
This way we could define new "array" data types.
Note: the number of elements of an array was limited
to 100 to put a limit and to ensure that an encoded
update message will continue to fit into a buffer
when the peer protocol will handle such data types.
This patch replaces all advanced data type aliases on
stktable_data_cast calls by standard types.
This way we could call the same stktable_data_cast
regardless of the used advanced data type as long they
are using the same std type.
It also removes all the advanced data type aliases.
This patch fixes the computation of the bit of the current data_type
in some part of code of peer protocol where the computation is limited
to 32bits whereas the bitfield of data_types can support 64bits.
Without this patch it could result in bugs when we will define more
than 32 data_types.
Backport is useless because there is currently less than 32 data_types
This patch fixes several errors printing integers
of stick table entry values and args during dump on cli.
This patch should be backported since the dump of entries
is supported. [wt: roughly 1.5-dev1 hence all stable branches]
CREATE USER has been the standard way of creating users since
MySQL-5.0 (2005).
The current syntax of INSERT INTO mysql.user won't actually work
on MariaDB-10.4+.
Because haproxy doesn't use any resources the MySQL executable comment
syntax provides resource contraints to make it more palatable
to risk adverse users.
/*!50701 is a syntax recognised by MySQL and MariaDB 5.7.1+ when
resource contraints where added.
/*M!100201 is a MariaDB executable comment syntax recognised for MariaDB
for the 10.2.1 where the MAX_STATEMENT_TIME was added.
This patch may be backported as far as 2.0.
Released version 2.5-dev1 with the following main changes :
- CLEANUP: ssl: Move ssl_store related code to ssl_ckch.c
- MINOR: ssl: Allow duplicated entries in the cafile_tree
- MEDIUM: ssl: Chain ckch instances in ca-file entries
- MINOR: ssl: Add reference to default ckch instance in bind_conf
- MINOR: ssl: Add helper functions to create/delete cafile entries
- MEDIUM: ssl: Add a way to load a ca-file content from memory
- MINOR: ssl: Add helper function to add cafile entries
- MINOR: ssl: Ckch instance rebuild and cleanup factorization in CLI handler
- MEDIUM: ssl: Add "set+commit ssl ca-file" CLI commands
- REGTESTS: ssl: Add new ca-file update tests
- MINOR: ssl: Add "abort ssl ca-file" CLI command
- MINOR: ssl: Add a cafile_entry type field
- MINOR: ssl: Refactorize the "show certificate details" code
- MEDIUM: ssl: Add "show ssl ca-file" CLI command
- MEDIUM: ssl: Add "new ssl ca-file" CLI command
- MINOR: ssl: Add "del ssl ca-file" CLI command
- REGTESTS: ssl: Add "new/del ssl ca-file" tests
- DOC: ssl: Add documentation about CA file hot update commands
- DOC: internals: update the SSL architecture schema
- MINOR: ssl: Chain instances in ca-file entries
- MEDIUM: ssl: Add "set+commit ssl crl-file" CLI commands
- MEDIUM: ssl: Add "new+del crl-file" CLI commands
- MINOR: ssl: Add "abort ssl crl-file" CLI command
- MEDIUM: ssl: Add "show ssl crl-file" CLI command
- REGTESTS: ssl: Add "new/del ssl crl-file" tests
- REGTESTS: ssl: Add "set/commit ssl crl-file" test
- DOC: ssl: Add documentation about CRL file hot update commands
- BUILD/MINOR: ssl: Fix compilation with SSL enabled
- BUILD/MINOR: ssl: Fix compilation with OpenSSL 1.0.2
- CI: introduce scripts/build-vtest.sh for installing VTest
- CLEANUP: ssl: Fix coverity issues found in CA file hot update code
- CI: github actions: add OpenTracing builds
- BUG/MEDIUM: ebtree: Invalid read when looking for dup entry
- BUG/MAJOR: server: prevent deadlock when using 'set maxconn server'
- BUILD/MINOR: opentracing: fixed build when using clang
- BUG/MEDIUM: filters: Exec pre/post analysers only one time per filter
- BUG/MINOR: http-comp: Preserve HTTP_MSGF_COMPRESSIONG flag on the response
- MINOR: map/acl: print the count of all the map/acl entries in "show map/acl"
- CLEANUP: pattern: remove export of non-existent function pattern_delete()
- MINOR: h1-htx: Update h1 parsing functions to return result as a size_t
- MEDIUM: h1-htx: Adapt H1 data parsing to copy wrapping data in one call
- MINOR: mux-h1/mux-fcgi: Don't needlessly loop on data parsing
- MINOR: h1-htx: Move HTTP chunks parsing into a dedicated function
- MEDIUM: h1-htx: Split function to parse a chunk and the loop on the buffer
- MEDIUM: h1-htx: Add a function to parse contiguous small chunks
- MINOR: h1-htx: Use a correlation table to speed-up small chunks parsing
- MINOR: buf: Add function to realign a buffer with a specific head position
- MINOR: muxes/h1-htx: Realign input buffer using b_slow_realign_ofs()
- CLEANUP: mux-h1: Rename functions parsing input buf and filling output buf
- Revert "MEDIUM: http-ana: Deal with L7 retries in HTTP analysers"
- BUG/MINOR: http-ana: Send the right error if max retries is reached on L7 retry
- BUG/MINOR: http-ana: Handle L7 retries on refused early data before K/A aborts
- MINOR: http-ana: Perform L7 retries because of status codes in response analyser
- MINOR: cfgparse: Fail when encountering extra arguments in macro
- DOC: intro: Fix typo in starter guide
- BUG/MINOR: server: Missing calloc return value check in srv_parse_source
- BUG/MINOR: peers: Missing calloc return value check in peers_register_table
- BUG/MINOR: ssl: Missing calloc return value check in ssl_init_single_engine
- BUG/MINOR: http: Missing calloc return value check in parse_http_req_capture
- BUG/MINOR: proxy: Missing calloc return value check in proxy_parse_declare
- BUG/MINOR: proxy: Missing calloc return value check in proxy_defproxy_cpy
- BUG/MINOR: http: Missing calloc return value check while parsing tcp-request/tcp-response
- BUG/MINOR: http: Missing calloc return value check while parsing tcp-request rule
- BUG/MINOR: compression: Missing calloc return value check in comp_append_type/algo
- BUG/MINOR: worker: Missing calloc return value check in mworker_env_to_proc_list
- BUG/MINOR: http: Missing calloc return value check while parsing redirect rule
- BUG/MINOR: http: Missing calloc return value check in make_arg_list
- BUG/MINOR: proxy: Missing calloc return value check in chash_init_server_tree
- CLEANUP: http-ana: Remove useless if statement about L7 retries
- BUG/MAJOR: stream-int: Release SI endpoint on server side ASAP on retry
- MINOR: backend: Don't release SI endpoint anymore in connect_server()
- BUG/MINOR: vars: Be sure to have a session to get checks variables
- DOC/MINOR: move uuid in the configuration to the right alphabetical order
- CLEANUP: mux-fcgi: Don't needlessly store result of data/trailers parsing
- BUILD: fix compilation for OpenSSL-3.0.0-alpha17
- MINOR: http-ana: Use -1 status for client aborts during queuing and connect
- REGTESTS: Fix http_abortonclose.vtc to support -1 status for some client aborts
- CLEANUP: backend: fix incorrect comments on locking conditions for lb functions
- CLEANUP: reg-tests: Remove obsolete no-htx parameter for reg-tests
- CI: github actions: add OpenSSL-3.0.0 builds
- CI: github actions: -Wno-deprecated-declarations with OpenSSL 3.0.0
- MINOR: errors: allow empty va_args for diag variadic macro
- REORG: errors: split errors reporting function from log.c
- CLEANUP: server: fix cosmetic of error message on sni parsing
- MEDIUM: errors: implement user messages buffer
- MINOR: log: do not discard stderr when starting is over
- MEDIUM: errors: implement parsing context type
- MINOR: errors: use user messages context in print_message
- MINOR: log: display exec path on first warning
- MINOR: errors: specify prefix "config" for parsing output
- MINOR: log: define server user message format
- REORG: server: use parsing ctx for server parsing
- REORG: config: use parsing ctx for server config check
- MINOR: server: use parsing ctx for server init addr
- MINOR: server: use ha_alert in server parsing functions
- DOC: use the req.ssl_sni in examples
- CLEANUP: cfgparse: Remove duplication of `MAX_LINE_ARGS + 1`
- CLEANUP: tools: Make errptr const in `parse_line()`
- MINOR: haproxy: Add `-cc` argument
- BUG: errors: remove printf positional args for user messages context
- CI: Make matrix.py executable and add shebang
- BUILD: make tune.ssl.keylog available again
- BUG/MINOR: ssl: OCSP stapling does not work if expire too far in the future
- Revert "BUG/MINOR: opentracing: initialization after establishing daemon mode"
- BUG/MEDIUM: opentracing: initialization before establishing daemon and/or chroot mode
- SCRIPTS: opentracing: enable parallel builds in build-ot.sh
- BUG/MEDIUM: compression: Fix loop skipping unused blocks to get the next block
- BUG/MEDIUM: compression: Properly get the next block to iterate on payload
- BUG/MEDIUM: compression: Add a flag to know the filter is still processing data
- MINOR: ssl: Keep the actual key length in the certificate_ocsp structure
- MINOR: ssl: Add new "show ssl ocsp-response" CLI command
- MINOR: ssl: Add the OCSP entry key when displaying the details of a certificate
- MINOR: ssl: Add the "show ssl cert foo.pem.ocsp" CLI command
- REGTESTS: ssl: Add "show ssl ocsp-response" test
- BUG/MINOR: server: explicitly set "none" init-addr for dynamic servers
- BUG/MINOR: pools: fix a possible memory leak in the lockless pool_flush()
- BUG/MINOR: pools: make DEBUG_UAF always write to the to-be-freed location
- MINOR: pools: do not maintain the lock during pool_flush()
- MINOR: pools: call malloc_trim() under thread isolation
- MEDIUM: pools: use a single pool_gc() function for locked and lockless
- BUG/MAJOR: pools: fix possible race with free() in the lockless variant
- CLEANUP: pools: remove now unused seq and pool_free_list
- MEDIUM: pools: remove the locked pools implementation
- BUILD: ssl: Fix compilation with BoringSSL
- BUG/MEDIUM: errors: include missing obj_type file
- REGTESTS: ssl: show_ssl_ocspresponce.vtc is broken with BoringSSL
- BUG/MAJOR: htx: Fix htx_defrag() when an HTX block is expanded
- BUG/MINOR: mux-fcgi: Expose SERVER_SOFTWARE parameter by default
- BUG/MINOR: h1-htx: Fix a signess bug with char data type when parsing chunk size
- CLEANUP: l7-retries: do not test the buffer before calling b_alloc()
- BUG/MINOR: resolvers: answser item list was randomly purged or errors
- MEDIUM: resolvers: add a ref on server to the used A/AAAA answer item
- MEDIUM: resolvers: add a ref between servers and srv request or used SRV record
- BUG/MINOR: server-state: load SRV resolution only if params match the config
- MINOR: config: remove support for deprecated option "tune.chksize"
- MINOR: config: completely remove support for "no option http-use-htx"
- MINOR: log: remove the long-deprecated early log-format tags
- MINOR: http: remove the long deprecated "set-cookie()" sample fetch function
- MINOR: config: reject long-deprecated "option forceclose"
- MINOR: config: remove deprecated option "http-tunnel"
- MEDIUM: proxy: remove the deprecated "grace" keyword
- MAJOR: config: remove parsing of the global "nbproc" directive
- BUILD: init: remove initialization of multi-process thread mappings
- BUILD: log: remove unused fmt_directive()
- REGTESTS: Remove REQUIRE_VERSION=1.6 from all tests
- REGTESTS: Remove REQUIRE_VERSION=1.7 from all tests
- CI: github actions: enable alpine/musl builds
- BUG/MAJOR: resolvers: segfault using server template without SRV RECORDs
- DOC: lua: Add a warning about buffers modification in HTTP
- MINOR: ssl: Use OpenSSL's ASN1_TIME convertor when available
- BUG/MINOR: stick-table: insert srv in used_name tree even with fixed id
- BUG/MEDIUM: server: extend thread-isolate over much of CLI 'add server'
- BUG/MEDIUM: server: clear dynamic srv on delete from proxy id/name trees
- BUG/MEDIUM: server: do not forget to generate the dynamic servers ids
- BUG/MINOR: server: do not keep an invalid dynamic server in px ids tree
- BUG/MEDIUM: server: do not auto insert a dynamic server in px addr_node
- BUG/MEDIUM: shctx: use at least thread-based locking on USE_PRIVATE_CACHE
- BUG/MINOR: ssl: use atomic ops to update global shctx stats
- BUG/MINOR: mworker: fix typo in chroot error message
- CLEANUP: global: remove unused definition of stopping_task[]
- MEDIUM: init: remove the loop over processes during init
- MINOR: mworker: remove the initialization loop over processes
- CLEANUP: global: remove the nbproc field from the global structure
- CLEANUP: global: remove pid_bit and all_proc_mask
- MEDIUM: global: remove dead code from nbproc/bind_proc removal
- MEDIUM: config: simplify cpu-map handling
- MEDIUM: cpu-set: make the proc a single bit field and not an array
- CLEANUP: global: remove unused definition of MAX_PROCS
- MEDIUM: global: remove the relative_pid from global and mworker
- DOC: update references to process numbers in cpu-map and bind-process
- MEDIUM: config: warn about "bind-process" deprecation
- CLEANUP: shctx: remove the different inter-process locking techniques
- BUG/MAJOR: queue: set SF_ASSIGNED when setting strm->target on dequeue
- MINOR: backend: only skip LB when there are actual connections
- BUG/MINOR: mux-h1: do not skip the error response on bad requests
- MINOR: connection: add helper conn_append_debug_info()
- MINOR: mux-h2/trace: report a few connection-level info during h2_init()
- CLEANUP: mux-h2/traces: better align user messages
- BUG/MINOR: stats: make "show stat typed desc" work again
- MINOR: mux-h2: obey http-ignore-probes during the preface
- BUG/MINOR: mux-h2/traces: bring back the lost "rcvd H2 REQ" trace
- BUG/MINOR: mux-h2/traces: bring back the lost "sent H2 REQ/RES" traces
- CLEANUP: assorted typo fixes in the code and comments
- CI: Replace the requirement for 'sudo' with a call to 'ulimit -n'
- REGTESTS: Replace REQUIRE_VERSION=2.5 with 'haproxy -cc'
- REGTESTS: Replace REQUIRE_OPTIONS with 'haproxy -cc' for 2.5+ tests
- REGTESTS: Replace REQUIRE_BINARIES with 'command -v'
- REGTESTS: Remove support for REQUIRE_BINARIES
- CI: ssl: enable parallel builds for OpenSSL on Linux
- CI: ssl: do not needlessly build the OpenSSL docs
- CI: ssl: keep the old method for ancient OpenSSL versions
- CLEANUP: server: a separate function for initializing the per_thr field
- BUG/MINOR: server: Forbid to set fqdn on the CLI if SRV resolution is enabled
- BUG/MEDIUM: server/cli: Fix ABBA deadlock when fqdn is set from the CLI
- MINOR: resolvers: Clean server in a dedicated function when removing a SRV item
- MINOR: resolvers: Remove server from named_servers tree when removing a SRV item
- BUG/MEDIUM: resolvers: Add a task on servers to check SRV resolution status
- BUG/MINOR: backend: restore the SF_SRV_REUSED flag original purpose
- BUG/MINOR: backend: do not set sni on connection reuse
- BUG/MINOR: resolvers: Use resolver's lock in resolv_srvrq_expire_task()
- BUG/MINOR: server/cli: Fix locking in function processing "set server" command
- BUG/MINOR: cache: Correctly handle existing-but-empty 'accept-encoding' header
- MINOR: ssl: fix typo in usage for 'new ssl ca-file'
- MINOR: ssl: always initialize random generator
- MINOR: ssl: check allocation in ssl_sock_init_srv
- MINOR: ssl: check allocation in parse ciphers/ciphersuites/verifyhost
- MINOR: ssl: check allocation in parse npn/sni
- MINOR: server: disable CLI 'set server ssl' for dynamic servers
- MINOR: ssl: render file-access optional on server crt loading
- MINOR: ssl: split parse functions for alpn/check-alpn
- MINOR: ssl: support ca-file arg for dynamic servers
- MINOR: ssl: support crt arg for dynamic servers
- MINOR: ssl: support crl arg for dynamic servers
- MINOR: ssl: enable a series of ssl keywords for dynamic servers
- MINOR: ssl: support ssl keyword for dynamic servers
- REGTESTS: server: test ssl support for dynamic servers
- MINOR: queue: update the stream's pend_pos before queuing it
- CLEANUP: Prevent channel-t.h from being detected as C++ by GitHub
- BUG/MAJOR: server: fix deadlock when changing maxconn via agent-check
- REGTESTS: fix maxconn update with agent-check
- MEDIUM: queue: make pendconn_process_next_strm() only return the pendconn
- MINOR: queue: update proxy->served once out of the loop
- MEDIUM: queue: refine the locking in process_srv_queue()
- MINOR: lb/api: remove the locked argument from take_conn/drop_conn
- MINOR: queue: create a new structure type "queue"
- MINOR: proxy: replace the pendconns-related stuff with a struct queue
- MINOR: server: replace the pendconns-related stuff with a struct queue
- MEDIUM: queue: use a dedicated lock for the queues
- MEDIUM: queue: simplify again the process_srv_queue() API
- MINOR: queue: factor out the proxy/server queuing code
- MINOR: queue: use atomic-ops to update the queue's index
- MEDIUM: queue: determine in process_srv_queue() if the proxy is usable
- MEDIUM: queue: move the queue lock manipulation to pendconn_process_next_strm()
- MEDIUM: queue: unlock as soon as possible
- MINOR: queue: make pendconn_first() take the lock by itself
- CLEANUP: backend: remove impossible case of round-robin + consistent hash
- MINOR: tcp-act: Add set-src/set-src-port for "tcp-request content" rules
- DOC: config: Add missing actions in "tcp-request session" documentation
- CLEANUP: dns: Remove a forgotten debug message
- DOC: Replace issue templates by issue forms
- Revert "MINOR: queue: make pendconn_first() take the lock by itself"
- Revert "MEDIUM: queue: unlock as soon as possible"
- Revert "MEDIUM: queue: move the queue lock manipulation to pendconn_process_next_strm()"
- Revert "MEDIUM: queue: determine in process_srv_queue() if the proxy is usable"
- Revert "MINOR: queue: use atomic-ops to update the queue's index"
- Revert "MINOR: queue: factor out the proxy/server queuing code"
- Revert "MEDIUM: queue: simplify again the process_srv_queue() API"
- Revert "MEDIUM: queue: use a dedicated lock for the queues"
- Revert "MEDIUM: queue: refine the locking in process_srv_queue()"
- Revert "MINOR: queue: update proxy->served once out of the loop"
- Revert "MEDIUM: queue: make pendconn_process_next_strm() only return the pendconn"
- MEDIUM: queue: update px->served and lb's take_conn once per loop
- MEDIUM: queue: use a dedicated lock for the queues (v2)
- MEDIUM: queue: simplify again the process_srv_queue() API (v2)
- MEDIUM: queue: determine in process_srv_queue() if the proxy is usable (v2)
- MINOR: queue: factor out the proxy/server queuing code (v2)
- MINOR: queue: use atomic-ops to update the queue's index (v2)
- MEDIUM: queue: take the proxy lock only during the px queue accesses
- MEDIUM: queue: use a trylock on the server's queue
- MINOR: queue: add queue_init() to initialize a queue
- MINOR: queue: add a pointer to the server and the proxy in the queue
- MINOR: queue: store a pointer to the queue into the pendconn
- MINOR: queue: remove the px/srv fields from pendconn
- MINOR: queue: simplify pendconn_unlink() regarding srv vs px
- BUG: backend: stop looking for queued connections once there's no more
- BUG/MINOR: queue/debug: use the correct lock labels on the queue lock
- BUG/MINOR: resolvers: Always attach server on matching record on resolution
- BUG/MINOR: resolvers: Reset server IP when no ip is found in the response
- MINOR: resolvers: Reset server IP on error in resolv_get_ip_from_response()
- BUG/MINOR: checks: return correct error code for srv_parse_agent_check
- BUILD: Makefile: fix linkage for Haiku.
- BUG/MINOR: tcpcheck: Fix numbering of implicit HTTP send/expect rules
- MINOR: http-act/tcp-act: Add "set-log-level" for tcp content rules
- MINOR: http-act/tcp-act: Add "set-nice" for tcp content rules
- MINOR: http-act/tcp-act: Add "set-mark" and "set-tos" for tcp content rules
- CLEANUP: tcp-act: Sort action lists
- BUILD/MEDIUM: tcp: set-mark setting support for FreeBSD.
- BUILD: tcp-act: avoid warning when set-mark / set-tos are not supported
- BUG/MINOR: mqtt: Fix parser for string with more than 127 characters
- BUG/MINOR: mqtt: Support empty client ID in CONNECT message
- BUG/MEDIUM: resolvers: Make 1st server of a template take part to SRV resolution
- CLEANUP: peers: re-write intdecode function comment.
The commit 3406766d5 ("MEDIUM: resolvers: add a ref between servers and srv
request or used SRV record") introduced a regression. The first server of a
template based on SRV record is no longer resolved. The same bug exists for
a normal server based on a SRV record.
In fact, the server used during parsing (used as reference when a
server-template line is parsed) is never attached to the corresponding srvrq
object. Thus with following lines, no resolution is performed because
"srvrq->attached_servers" is empty:
server-template test 1 _http.domain.tld resolvers dns ...
server test1 _http.domain.tld resolvers dns ...
This patch should fix the issue #1295 (but not confirmed yet it is the same
bug). It must be backported everywhere the above commit is.
As specified by the MQTT specification (MQTT-3.1.3-6), the client ID may be
empty. That means the length of the client ID string may be 0. However, The
MQTT parser does not support empty strings.
So, to fix the bug, the mqtt_read_string() function may now parse empty
string. 2 bytes must be found to decode the string length, but the length
may be 0 now. It is the caller responsibility to test the string emptiness
if necessary. In addition, in mqtt_parse_connect(), the client ID may be
empty now.
This patch should partely fix the issue #1310. It must be backported to 2.4.
Parsing of too long strings (> 127 characters) was buggy because of a wrong
cast on the length bytes. To fix the bug, we rely on mqtt_read_2byte_int()
function. This way, the string length is properly decoded.
This patch should partely fix the issue #1310. It must be backported to 2.4.
Since recent commit 469c06c30 ("MINOR: http-act/tcp-act: Add "set-mark"
and "set-tos" for tcp content rules") there's a build warning (or error)
on Windows due to static function tcp_action_set_mark() not being used
because the set-mark functionality is not supported there. It's caused
by the fact that only the parsing function uses it so if the code is
ifdefed out the function remains unused.
Let's surround it with ifdefs as well, and do the same for
tcp_action_set_tos() which could suffer the same fate on operating systems
not defining IP_TOS.
This may need to be backported if the patch above is backported. Also
be careful, the condition was adjusted to cover FreeBSD after commit
f7f53afcf ("BUILD/MEDIUM: tcp: set-mark setting support for FreeBSD.").
It is now possible to set the Netfilter MARK and the TOS field value in all
packets sent to the client from any tcp-request rulesets or the "tcp-response
content" one. To do so, the parsing of "set-mark" and "set-tos" actions are
moved in tcp_act.c and the actions evaluation is handled in dedicated functions.
This patch may be backported as far as 2.2 if necessary.
It is now possible to set the "nice" factor of the current stream from a
"tcp-request content" or "tcp-response content" ruleset. To do so, the
action parsing is moved in stream.c and the action evaluation is handled in
a dedicated function.
This patch may be backported as far as 2.2 if necessary.
It is now possible to set the stream log level from a "tcp-request content"
or "tcp-response content" ruleset. To do so, the action parsing is moved in
stream.c and the action evaluation is handled in a dedicated function.
This patch should fix issue #1306. It may be backported as far as 2.2 if
necessary.
The index of the failing rule is reported in the health-check log message. The
rules index is also used in the check traces. But for implicit HTTP send/expect
rules, the index is wrong. It must be incremented by one compared to the
preceding rule.
This patch may be backported as far as 2.2.
In srv_parse_agent_check the error code is not returned in case
something goes wrong. The value 0 is always return.
Additionally, there's a small cleanup of unreachable returns that in
most checks are not present either and removed in two places they were
present. This makes the code consistent across the different checks.
If resolv_get_ip_from_response() returns an error (or an unexpected return
value), the server is set to RMAINT status. However, its address must also
be reset. Otherwise, it is still reported by the cli on "show servers state"
commands. This may be confusing. Note that it is a theorical patch because
this code path does not exist. Thus it is not tagged as a BUG.
This patch may be backported as far as 2.0.
For A/AAAA resolution, if no ip is found for a server in the response, the
server is set to RMAINT status. However, its address must also be
reset. Otherwise, it is still reported by the cli on "show servers state"
commands. This may be confusing.
This patch may be backported as far as 2.0.
On A/AAAA resolution, for a given server, if a record is matching, we must
always attach the server to this record. Before it was only done if the
server IP was not the same than the record one. However, it is a problem if
the server IP was not set for a previous resolution. From the libc during
startup for instance. In this case, the server IP is not updated and the
server is not attached to any record. It remains in this state while a
matching record is found in the DNS response. It is especially a problem
when the resolution is used for server-templates.
This bug was introduced by the commit bd78c912f ("MEDIUM: resolvers: add a
ref on server to the used A/AAAA answer item").
This patch should solve the issue #1305. It must be backported to all
versions containing the above commit.
A dedicated queue lock was added by commit 16fbdda3c ("MEDIUM: queue:
use a dedicated lock for the queues (v2)") but during its rebase, some
labels were lost and left to SERVER_LOCK / PROXY_LOCK instead of
QUEUE_LOCK. It's harmless but can confuse the lock debugger, so better
fix it.
No backport is needed.
Commit ae0b12ee0 ("MEDIUM: queue: use a trylock on the server's queue")
introduced a hard to trigger bug that's more visible with a single thread:
if a server dequeues a connection and finds another free slot with no
connection to place there, process_srv_queue() will never break out of
the loop. In multi-thread it almost does not happen because other threads
bring new connections.
No backport is needed as it's only in -dev.
Since the code paths became exactly the same except for what log field
to update, let's simplify the code and move further code out of the
lock. The queue position update and the test for server vs proxy do not
need to be inside the lock.
Now we directly use p->queue to get to the queue, which is much more
straightforward. The performance on 100 servers and 16 threads
increased from 560k to 574k RPS, or 2.5%.
A lot more simplifications are possible, but the minimum was done at
this point.
A queue is specific to a server or a proxy, so we don't need to place
this distinction inside all pendconns, it can be in the queue itself.
This commit adds the relevant fields "px" and "sv" into the struct
queue, and initializes them accordingly.
Doing so makes sure that threads attempting to wake up new connections
for a server will give up early if another thread is already in charge
of this. The goal is to avoid unneeded contention on low server counts.
Now with a single server with 16 threads in roundrobin we get the same
performance as with multiple servers, i.e. ~575kreq/s instead of ~496k
before. Leastconn is seeing a similar jump, from ~460 to ~560k (the
difference being the calls to fwlc_srv_reposition).
The overhead of process_srv_queue() is now around 2% instead of ~20%
previously.
There's no point keeping the proxy lock held for a long time, it's
only needed when checking the proxy's queue, and keeping it prevents
multiple servers from dequeuing in parallel. Let's move it into
pendconn_process_next_strm() and release it ASAP. The pendconn
remains under the server queue lock's protection, guaranteeing that
no stream will release it while it's being touched.
For roundrobin, the performance increases by 76% (327k to 575k) on
16 threads. Even with a single server and maxconn=100, the performance
increases from 398 to 496 kreq/s. For leastconn, almost no change is
visible (less than one percent) but this is expected since most of the
time there is spent in fwlc_reposition() and fwlc_get_next_server().
Doing so allows to retrieve and update the pendconn's queue index outside
of the queue's lock and to save one more percent CPU on a highly-contented
backend.
The code only differed by the nbpend_max counter. Let's have a pointer
to it and merge the two variants to always use a generic queue. It was
initially considered to put the max inside the queue structure itself,
but the stats support clearing values and maxes and this would have been
the only counter having to be handled separately there. Given that we
don't need this max anywhere outside stats, let's keep it where it is
and have a pointer to it instead.
The CAS loop to update the max remains. It was naively thought that it
would have been faster without atomic ops inside the lock, but this is
not the case for the simple reason that it is a max, it converges very
quickly and never has to perform the check anymore. Thus this code is
better out of the lock.
The queue_idx is still updated inside the lock since that's where the
idx is updated, though it could be performed using atomic ops given
that it's only used to roughly count places for logging.
This basically undoes the API changes that were performed by commit
0274286dd ("BUG/MAJOR: server: fix deadlock when changing maxconn via
agent-check") to address the deadlock issue: since process_srv_queue()
doesn't use the server lock anymore, it doesn't need the "server_locked"
argument, so let's get rid of it before it gets used again.
Till now whenever a server or proxy's queue was touched, this server
or proxy's lock was taken. Not only this requires distinct code paths,
but it also causes unnecessary contention with other uses of these locks.
This patch adds a lock inside the "queue" structure that will be used
the same way by the server and the proxy queuing code. The server used
to use a spinlock and the proxy an rwlock, though the queue only used
it for locked writes. This new version uses a spinlock since we don't
need the read lock part here. Tests have not shown any benefit nor cost
in using this one versus the rwlock so we could change later if needed.
The lower contention on the locks increases the performance from 362k
to 374k req/s on 16 threads with 20 servers and leastconn. The gain
with roundrobin even increases by 9%.
This is tagged medium because the lock is changed, but no other part of
the code touches the queues, with nor without locking, so this should
remain invisible.
