Activity
From 2007-08-21 to 2007-08-27
2007-08-27
- 21:57 Bug #1322 (Fixed): etags checked in http-header-glue.c even when unconfigured
- 21:57 Bug #1322: etags checked in http-header-glue.c even when unconfigured
- fixed in r1971
- 21:56 Revision 523336eb: - added note about bug #1322
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1972 152afb58-edef-0310-8abb-c4023f1b3aa9
- 21:56 Revision 1972 (svn): - added note about bug #1322
- 21:54 Revision 15d83784: - make sure the etag buffer is non empty if we compare it with if-non-match
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1971 152afb58-edef-0310-8abb-c4023f1b3aa9
- 21:54 Revision 1971 (svn): - make sure the etag buffer is non empty if we compare it with if-non-match
- 21:50 Revision 1970 (svn): protect buffer_is_empty() from NULL buffers
- 21:49 Revision 53634de0: protect buffer_is_empty() from NULL buffers
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1969 152afb58-edef-0310-8abb-c4023f1b3aa9
- 21:49 Revision 1969 (svn): protect buffer_is_empty() from NULL buffers
- 21:48 Revision ea872bea: - added testcase for bug #1322
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1968 152afb58-edef-0310-8abb-c4023f1b3aa9
- 21:48 Revision 1968 (svn): - added testcase for bug #1322
- 21:10 Revision 74c303a7: - a more portable version for the SAPI breakage in php 5.2.4
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1967 152afb58-edef-0310-8abb-c4023f1b3aa9
- 21:10 Revision 1967 (svn): - a more portable version for the SAPI breakage in php 5.2.4
- 20:26 Bug #1320: check return value of calloc in mod_simple_vhost.c
- > just put assert() around it and let it die.
Is assert supposed to be used like that? I don't think so.
- 14:53 Feature #822: Pipe support to error log(server.errorlog), support of multiple workers
- an implementation for using pipes for error log is available from TRAC 296 :
http://trac.lighttpd.net/trac/ticke... - 13:13 Bug #949: fastcgi, cgi, flush, php5 problem.
- PHP does not need much memory for this loop - it just creates content.
Did anyone test this and did _not_ have this... - 12:16 Revision 1966 (svn): added a msg to run autogen.sh
- 11:40 Bug #1164: mod_extforward breaks (at least) URL matching
- Replying to [[lighttpd:comment:9|cahe]]:
> yep, doesn't work for me either. please fix this bug, it's killing me.. :... - 11:19 Bug #1164: mod_extforward breaks (at least) URL matching
- yep, doesn't work for me either. please fix this bug, it's killing me.. :(
-- cahe
2007-08-26
- 23:24 Bug #1164: mod_extforward breaks (at least) URL matching
- lighttpd-1.4.17.r1956.tar.gz - still doesn't work for me
-- misiolap - 14:57 Feature #1325: mod_core_proxy doesn't modify (parts of) "Destination" in WebDAV COPY/MOVE requests
- After reading through the code, there's a solution for it:
- 09:00 Feature #1327 (Obsolete): Add mimetypes to documentation and sample configuration files
- One or more of the sample configuration files in the released version could include the document formats common on FO...
- 03:29 Bug #1171: mod_compress supresses Etag and Last-Modified headers
- For some reason, this is the behavior when you don't turn on compress.cache-dir. Set this to a valid server path and ...
2007-08-25
- 23:54 Bug #1326 (Invalid): lighttpd.conf logic does not work as expected.
- The logic in lighttpd is totally borked 1.4.14 - 1.4.16. (versions i've tried) It is entirely too difficult to get th...
- 10:52 Bug #841: Segmentation fault using fastcgi / php5-cgi
- confirmed for 1.4.16
The problem is, that it is not clear in the source, if it is ok to not specify "host".
The c... - 07:27 Bug #1262 (Need Feedback): mod_uploadprogress bug
- Not sure, not sure...
Revision 1964 - seems it works fine if server.max-worker = 1:...
2007-08-24
- 23:49 Bug #577: lighttpd terminally ceases serving connections when server.max-connections state reached
- Still there in 1.4.16; tested on a Debian system.
I think it just increases load; if you want you can try my patch ... - 22:55 Bug #1322: etags checked in http-header-glue.c even when unconfigured
- The above patch doesn't catch the error actually, a new patch that check con->physical.etag->ptr has been attached, ...
- 12:42 Feature #1325: mod_core_proxy doesn't modify (parts of) "Destination" in WebDAV COPY/MOVE requests
- sorry, the WikiFormatting didn't like what I pasted, here is it again:...
- 12:40 Feature #1325 (Obsolete): mod_core_proxy doesn't modify (parts of) "Destination" in WebDAV COPY/MOVE requests
- ... and this leads to a "502 Bad Gateway" error.
This is the HTTP Request as captured with wireshark:
***********... - 11:27 Feature #44: add mod_log_spread
- I have this truly horrible patch for spread logging.
Adding:
accesslog.spread_location = "port@domain"
accessl... - 09:02 Bug #1324 (Fixed): authorization blocks OPTIONS
- If a client sends a OPTIONS method for a resource that is under authorization, then mod_auth blocks the other modules...
- 06:53 Bug #1308: lighty.env['request.method'] and lighty.env['request.protocol'] always return nil
- since we are talking about LUA which is very flexible.
Why not just return the request.request(or request.request_... - 02:40 Feature #1288: SSL Client Certificate validation.
- Moving the call to https_add_ssl_entries below plugins_call_handle_uri_raw in response.c fixes the issue I reported (...
2007-08-23
- 22:22 Feature #44: add mod_log_spread
- Did this get added?
-- joe - 20:37 Bug #1323: lighttpd proxy
- lol, dupe of #1302. sorry :(
- 20:35 Bug #1323 (Duplicate): lighttpd proxy
- I use r-1857, but afaik mod_proxy_core cannot generate a request for an external proxy server, it will always send a ...
- 18:18 Bug #1322 (Fixed): etags checked in http-header-glue.c even when unconfigured
- I have etags disabled via the new 1.4.16 feature, but when a if-none-match header is received it still tries to issue...
- 18:15 Bug #1311: fcgi fails for first request
- The PID is indeed still running after the error.
-- keturn <kevin - 12:09 Bug #1311 (Assigned): fcgi fails for first request
- The trac process seems to die before it sends a response.
Check that the PID is still running after the error is ... - 17:36 Bug #1306 (Fixed): mysql_query() call might block if updates are done to the table (or a WRITE table lock is done)
- 17:20 Bug #1312 (Fixed): $HTTP["remoteip"] compared to subnet does not work
- closing as of last comment.
- 16:47 Bug #1312: $HTTP["remoteip"] compared to subnet does not work
- With the 1.4.17 release candidate it seems to work as expected.
-- john.eckerdal - 12:10 Bug #1312: $HTTP["remoteip"] compared to subnet does not work
- Yep, this is a duplicate of #41
Close as duplicate when done.
- 17:00 Bug #405: PHP SCRIPT_NAME and PHP_SELF truncated when inside a userdir and using PATH_INFO
- the last test was run with HEAD of the 1.4.x branch using the tests/lighttpd.conf plus the following patch:...
- 16:54 Bug #405: PHP SCRIPT_NAME and PHP_SELF truncated when inside a userdir and using PATH_INFO
- The test script:...
- 16:52 Feature #1321 (Fixed): Reloading configuration of lighttpd
- Hi there,
I'm running lighttpd 1.4.16, a very good software. Is it possible to reload the lighttpd configuration w... - 16:28 Bug #1320 (Assigned): check return value of calloc in mod_simple_vhost.c
- just put assert() around it and let it die. We can't handle it graceful anyway.
The buffer.c code already does that.
- 14:56 Bug #1320: check return value of calloc in mod_simple_vhost.c
- this problem affects more calls to malloc(and friends).
- 13:45 Bug #1320 (Fixed): check return value of calloc in mod_simple_vhost.c
- check return value of calloc in mod_simple_vhost.c
- 15:23 Bug #1197: SIGABRT on MacOS X + select
- Replying to [[lighttpd:comment:2|jan]]:
> does the same happen with kqueue or poll as fdevent handler ?
I'm sorry... - 15:15 Bug #1197 (Assigned): SIGABRT on MacOS X + select
- 15:14 Bug #1197: SIGABRT on MacOS X + select
- does the same happen with kqueue or poll as fdevent handler ?
- 13:42 Bug #1319 (Invalid): simple-vhost.default-host and invalid hostnames
- Shouldn't simple-vhost.default-host be used only when
request is HTTP1.0 and Request-URI is not an absoluteURI, i.e.... - 13:23 Bug #1164: mod_extforward breaks (at least) URL matching
- can you please test with
http://zen.sh.nu/~darix/lighttpd-1.4.17.r1956.tar.gz
- 13:16 Bug #1164: mod_extforward breaks (at least) URL matching
- can you please please please fix this issue... it's the most annoying of the bugs right now.
- 12:56 Bug #1316 (Need Feedback): off_t is more than 4 bytes but we can't parse it with strtol()
- Well I did run make clean and ./configure, hoping it was enough.
Is it possible to fix makefiles so that ./autogen... - 12:07 Bug #1316 (Fixed): off_t is more than 4 bytes but we can't parse it with strtol()
- 12:01 Bug #1316: off_t is more than 4 bytes but we can't parse it with strtol()
- please update to HEAD and rerun autogen.sh.
see http://trac.lighttpd.net/trac/wiki/Devel/Subversion how to get trunk... - 08:34 Bug #1316 (Obsolete): off_t is more than 4 bytes but we can't parse it with strtol()
- I have LFS and strtoll.
In sys-strings.h this is triggered.
#error off_t is more than 4 bytes but we can't parse ... - 12:48 Bug #1314: http_request_parse: clear errno before str_to_off_t
- System calls do not modify errno on successful call.
- 12:06 Bug #1314 (Assigned): http_request_parse: clear errno before str_to_off_t
- don't strtol() and stroll() reset the errno themself ?
- 08:29 Bug #1314 (Obsolete): http_request_parse: clear errno before str_to_off_t
- if earlier syscall sets errno to ERANGE,
http_request_parse does not detect between overflow and r == STR_OFF_T_MAX.
- 12:28 Bug #1199 (Assigned): Problem with 1.5.0-r1857 on Debian.
- 12:27 Bug #1294 (Fixed): file "compile" missing in all 1.4.14+ distributions
- 12:07 Bug #1309 (Fixed): changeset r1919, mod_mysql_vhost: compile error
- 12:02 Bug #1309: changeset r1919, mod_mysql_vhost: compile error
- fixed in r1965
- 12:02 Revision f7df9a9e: MYSQL_OPT_RECONNECT is available from 5.0.13 on (fixes #1309)
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1965 152afb58-edef-0310-8abb-c4023f1b3aa9
- 12:02 Revision 1965 (svn): MYSQL_OPT_RECONNECT is available from 5.0.13 on (fixes #1309)
- 11:58 Bug #1315 (Fixed): stat_cache_get_entry_internal: fd leak on error path
- applied in r1964
- 08:32 Bug #1315 (Fixed): stat_cache_get_entry_internal: fd leak on error path
- stat_cache_get_entry_internal: fd leak on error path
- 11:58 Revision 1964 (svn): fixed possible fd-leak (fixes #1315)
- 11:57 Revision 1963 (svn): - added AM_PROG_CC_C_O as autotools suggest
- 11:06 Feature #1318 (Obsolete): dd schema match to lighty
- Hi
Please add $HTTP\[[scheme]] to lighty. This makes http->https redirection easy as shown in http://trac.lighttpd.n... - 08:55 Bug #1317 (Fixed): move fails for properties when new file exists
- When performing a MOVE the properties database is adjusted with an )"UPDATE properites SET resource = "new" WHERE res...
- 05:18 Feature #1313: Allow conditional config based on local IP and port
- $HTTP[ "host" ] can partly replace $HTTP[ "localip" ], but again we may got different "Host:" headers even on the sam...
2007-08-22
- 22:38 Feature #1313: Allow conditional config based on local IP and port
- you can do that already
$HTTP[[host]] is evaluate within https hosts too.
sadly it has downsides.
- 18:45 Feature #1313 (Need Feedback): Allow conditional config based on local IP and port
- OK, you got a valid workaround to avoid duplication of config, but I am not convinced that this is the more preferred...
- 17:46 Feature #1313 (Fixed): Allow conditional config based on local IP and port
- you dont have to duplicate stuff. you just put them into a seperate file and use:...
- 17:44 Feature #1313 (Wontfix): Allow conditional config based on local IP and port
- ...
- 17:44 Bug #1312: $HTTP["remoteip"] compared to subnet does not work
- can you please test if http://zen.sh.nu/~darix/lighttpd-1.4.17.r1956.tar.gz
fixes your problem? this is a 1.4.17 re... - 17:33 Bug #1312 (Invalid): $HTTP["remoteip"] compared to subnet does not work
- I use lighttpd as a personal webserver and on this server I also have a wiki. I want to block access to the wiki dire...
- 09:04 Revision 1962 (svn): check for strtoll
- 02:00 Bug #1311 (Obsolete): fcgi fails for first request
- lighttpd-1.4.13, trac 0.10.3.1
I have lighttpd starting trac with a configuration like this:...
2007-08-21
- 23:37 Bug #1310: ifdef for prctl is checking wrong constant
- and in r1961 in 1.5.0
- 23:35 Bug #1310 (Fixed): ifdef for prctl is checking wrong constant
- fixed in r1958
- 21:06 Bug #1310 (Fixed): ifdef for prctl is checking wrong constant
- In 1.4.16, and possibly earlier, the constant in config.h for prctl was renamed to HAVE_SYS_PRCTL_H, but server.c is ...
- 23:37 Revision 1961 (svn): fixed prctl() usage (#1310)
- 23:36 Revision 1960 (svn): use str_to_off_t instead of strtoll() which might not be available on
- all platforms
- 23:36 Revision 1959 (svn): added logging of the PID and UID of the sending process for SIGTERM and
- SIGINT
- 23:35 Revision 1e1a8e59: use HAVE_SYS_PRCTL_H as HAVE_PRCTL is not checked for
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1958 152afb58-edef-0310-8abb-c4023f1b3aa9
- 23:35 Revision 1958 (svn): use HAVE_SYS_PRCTL_H as HAVE_PRCTL is not checked for
- 18:31 Bug #1244: lighttpd does not compile on Digital UNIX
- ...and I never said that.
-- gabor - 17:53 Bug #1244: lighttpd does not compile on Digital UNIX
- and so far lighttpd runs on hpux, solaris (even the old ones) and irix, bsds, linux, osx, windows. that doesnt sounds...
- 17:52 Bug #1244: lighttpd does not compile on Digital UNIX
- how does it help if e.g. many types from stdint.h are missing?
including db.h is just an ugly workaround. imho.
a... - 17:37 Bug #1244: lighttpd does not compile on Digital UNIX
- Yes, but most (portable, e.g.: not linux-specific) stuff runs on it, even modern ones. It will also help on Tru64 (DU...
- 15:33 Bug #1244: lighttpd does not compile on Digital UNIX
- This is Digital Unix 4.0b, from 1996 ...
It will take some effort to get it running on this old lady.
- 17:40 Revision ef19baca: added sending UID and PID for SIGTERM and SIGINT to the logs
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1957 152afb58-edef-0310-8abb-c4023f1b3aa9
- 17:40 Revision 1957 (svn): added sending UID and PID for SIGTERM and SIGINT to the logs
- 15:05 Revision 30748f20: fixed timestamps for 32bit time_t which wraps at 2038
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1956 152afb58-edef-0310-8abb-c4023f1b3aa9
- 15:05 Revision 1956 (svn): fixed timestamps for 32bit time_t which wraps at 2038
- 11:02 Revision 470d32b8: - added AM_PROG_CC_C_O as suggested by autotools
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1955 152afb58-edef-0310-8abb-c4023f1b3aa9
- 11:02 Revision 1955 (svn): - added AM_PROG_CC_C_O as suggested by autotools
- 10:56 Revision 6719b15c: - added missing Makefile.am entries
- git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-1.4.x@1954 152afb58-edef-0310-8abb-c4023f1b3aa9
- 10:56 Revision 1954 (svn): - added missing Makefile.am entries
- 10:29 Bug #1090: mod_auth ldap fails after LDAP restart
- Todd, are you using conditional evaluation in the lighttd.conf?
Also available in: Atom