summaryrefslogtreecommitdiffhomepage
path: root/src (follow)
AgeCommit message (Collapse)AuthorFilesLines
2011-08-22SIGWINCH/NOACCEPT signal is disabled now in non-daemon mode.Igor Sysoev1-2/+7
Non-daemon mode is currently used by supervisord, daemontools and so on or during debugging. The NOACCEPT signal is only used for online upgrade which is not supported when nginx is run under supervisord, etc., so this change should not break existant setups.
2011-08-22The change in adaptive loader behaviour introduced in r3975:Igor Sysoev1-40/+26
now cache loader processes either as many files as specified by loader_files or works no more than time specified by loader_threshold during each iteration. loader_threshold was previously used to decrease loader_files or to increase loader_timeout and this might eventually result in downgrading loader_files to 1 and increasing loader_timeout to large values causing loading cache for forever.
2011-08-22Fix buffer overrun under Windows.Maxim Dounin1-1/+2
2011-08-21Move SO_ACCEPTFILTER and TCP_DEFER_ACCEPT checks into configure.Maxim Dounin3-11/+1
NetBSD 5.0+ has SO_ACCEPTFILTER support merged from FreeBSD, and having accept filter check in FreeBSD-specific ngx_freebsd_config.h prevents it from being used on NetBSD. Therefore move the check into configure (and do the same for Linux-specific TCP_DEFER_ACCEPT, just to be in line).
2011-08-19Fix ignored headers handling in fastcgi/scgi/uwsgi.Maxim Dounin3-3/+27
The bug had appeared in r3561 (fastcgi), r3638 (scgi), r3567 (uwsgi).
2011-08-18Complain on invalid log levels.Maxim Dounin1-3/+7
Previously only first log level was required to be correct, while error_log directive in fact accepts list of levels (e.g. one may specify "error_log ... debug_core debug_http;"). This resulted in (avoidable) wierd behaviour on missing semicolon after error_log directive, e.g. error_log /path/to/log info index index.php; silently skipped index directive and it's arguments (trying to interpret them as log levels without checking to be correct).
2011-08-18Upstream: properly allocate memory for tried flags.Maxim Dounin1-3/+8
Previous allocation only took into account number of non-backup servers, and this caused memory corruption with many backup servers. See report here: http://mailman.nginx.org/pipermail/nginx/2011-May/026531.html
2011-08-18Fixing cpu hog with all upstream servers marked "down".Maxim Dounin1-1/+5
The following configuration causes nginx to hog cpu due to infinite loop in ngx_http_upstream_get_peer(): upstream backend { server 127.0.0.1:8080 down; server 127.0.0.1:8080 down; } server { ... location / { proxy_pass http://backend; } } Make sure we don't loop infinitely in ngx_http_upstream_get_peer() but stop after resetting peer weights once. Return 0 if we are stuck. This is guaranteed to work as peer 0 always exists, and eventually ngx_http_upstream_get_round_robin_peer() will do the right thing falling back to backup servers or returning NGX_BUSY.
2011-08-18Fixing proxy_set_body and proxy_pass_request_body with SSL.Maxim Dounin1-2/+1
Flush flag wasn't set in constructed buffer and this prevented any data from being actually sent to upstream due to SSL buffering. Make sure we always set flush in the last buffer we are going to sent. See here for report: http://nginx.org/pipermail/nginx-ru/2011-June/041552.html
2011-08-18Fix names of the referer hash size directives introduced in r3940.Igor Sysoev1-1/+1
2011-08-18Fix body with request_body_in_single_buf.Maxim Dounin1-1/+3
If there were preread data and request body was big enough first part of the request body was duplicated. See report here: http://mailman.nginx.org/pipermail/nginx/2011-July/027756.html
2011-08-18Correctly set body if it's preread and there are extra data.Maxim Dounin1-0/+1
Previously all available data was used as body, resulting in garbage after real body e.g. in case of pipelined requests. Make sure to use only as many bytes as request's Content-Length specifies.
2011-08-05fix gzip quantity: "q=0." and "q=1." are valid values according to RFCIgor Sysoev1-1/+1
2011-08-04refactor gzip quantity introduced in r3981: it ignored "q=1.000"Igor Sysoev1-27/+36
2011-08-04A new fix for the case when ssl_session_cache defined, but ssl is notIgor Sysoev5-10/+7
enabled in any server. The previous r1033 does not help when unused zone becomes used after reconfiguration, so it is backed out. The initial thought was to make SSL modules independed from SSL implementation and to keep OpenSSL code dependance as much as in separate files.
2011-08-03removal of error message about %name log_format parameters,Igor Sysoev1-19/+0
they have been deleted long ago in 0.5.0-RELEASE
2011-08-02fix typo introduced in r3985Igor Sysoev1-1/+1
2011-08-02bump versionIgor Sysoev2-3/+3
2011-08-01fix r3981 again for case "Accept-Encoding: gzip"Igor Sysoev1-1/+1
2011-08-01always set timer in discard body handler, this fixes the casesIgor Sysoev1-1/+1
when request for static file is redirected by error_page to an SSI page patch by Maxim Dounin
2011-08-01lingering_close "off|on|always"Igor Sysoev3-4/+30
patch by Maxim Dounin
2011-08-01do not send RST on normal lingering close read timeout,Igor Sysoev1-1/+0
if reset_timedout_connection is on patch by Maxim Dounin
2011-08-01fix r3981 for case "Accept-Encoding: gzip"Igor Sysoev1-1/+1
2011-08-01enable lingering close for pipelined requestsIgor Sysoev1-1/+5
patch by Maxim Dounin
2011-08-01Accept-Encoding refactoring: "gzip; q=0" supportIgor Sysoev1-15/+136
2011-07-30Accept-Encoding refactoring: test first the most common case "gzip,"Igor Sysoev1-7/+27
2011-07-30Accept-Encoding refactoring: remove ancient MSIE 4.x test for gzipIgor Sysoev3-13/+1
2011-07-30test length of proxy_pass with variablesIgor Sysoev1-4/+6
patch by Lanshun Zhou
2011-07-29update r3945 with more descriptive error messageIgor Sysoev6-3/+14
2011-07-29loader_files, loader_sleep, and loader_thresholdIgor Sysoev2-12/+71
2011-07-25set correct configuration file values while adding pathIgor Sysoev1-0/+2
patch by Maxim Dounin
2011-07-25rename ngx_http_file_cache_manager_sleep() to ngx_http_file_cache_loader_sleep()Igor Sysoev1-4/+4
and do not use it all in cache manager: this is a vestige of the times when cache manager loaded cache
2011-07-24do not close connection if cache file is too small: replace it with valid oneIgor Sysoev1-1/+1
2011-07-24elimination of reading cache files by cache loaderIgor Sysoev5-42/+13
2011-07-24fix r3968Igor Sysoev1-1/+1
2011-07-24update the previous commit:Igor Sysoev1-2/+1
removing dependencies on file uniq since WIN32_FIND_DATA has no such field
2011-07-24The cache loader performs two tasks: inserting cache objects in inactivityIgor Sysoev1-11/+6
list and evaluating total cache size. Reading just directory is enough for this purpose. Elimination of reading cache files saves at least one disk I/O operation per file. Preparation for elimination of reading cache files by cache loader: removing dependencies on the reading: *) cache node valid_sec and valid_msec are used only for caching errors; *) upstream buffer size can be used instead of cache node body_start.
2011-07-24fuse two if's in one conditionIgor Sysoev1-11/+2
2011-07-24style fixIgor Sysoev1-1/+1
2011-07-22do not try to reuse and save a SSL session for a peer created on the flyIgor Sysoev1-4/+25
by ngx_http_upstream_create_round_robin_peer(), since the peer lives only during request so the saved SSL session will never be used again and just causes memory leak patch by Maxim Dounin
2011-07-22finalizing with rc == 0 in unbuffered proxy mode caused nginx to waitIgor Sysoev1-1/+2
for another send_timeout before actually closing client's connection if client timed out while still talking to upstream server patch by Maxim Dounin
2011-07-22fix SSL connection issues on platforms with 32-bit off_tIgor Sysoev2-3/+5
patch by Maxim Dounin
2011-07-22fix build by gcc46 with -Wunused-value optionIgor Sysoev1-3/+7
patch by Maxim Dounin
2011-07-20ECDHE supportIgor Sysoev6-2/+72
patch by Adrian Kotelba
2011-07-20MSIE export versions are rare now, so RSA 512 key is generated on demandIgor Sysoev4-25/+11
and is shared among all hosts instead of pregenerating for every HTTPS host on configuraiton phase. This decreases start time for configuration with large number of HTTPS hosts.
2011-07-20bump version: 1.1.0 development versionIgor Sysoev2-3/+3
2011-07-19fix segfault if cache key is larger than upstream buffer sizeIgor Sysoev1-0/+9
patch by Lanshun Zhou
2011-07-19$uid_resetIgor Sysoev1-18/+63
2011-07-19fix r3756: release lock to allow other process to delete cache nodeIgor Sysoev1-1/+2
patch by Maxim Dounin
2011-07-18fix building by MSVCIgor Sysoev1-26/+26
generated by cgit (git 2.34.1) at 2025-12-06 21:24:22 +0000