cvs commit: apache-apr STATUS
manoj 99/07/28 15:53:14 Modified:.STATUS Log: Since when are temperatures in the 90's a massive heat wave? Not that I like it or anything... Revision ChangesPath 1.29 +11 -12apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.28 retrieving revision 1.29 diff -u -d -u -r1.28 -r1.29 --- STATUS1999/06/30 19:01:01 1.28 +++ STATUS1999/07/28 22:53:12 1.29 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/06/30 19:01:01 $] +Last modified at [$Date: 1999/07/28 22:53:12 $] Release: @@ -20,15 +20,11 @@ Available Patches: In progress: -Logging statements -We have some comments along the lines of log this here. -Let's actually make those logs at some point. - Modules Ensure thread safety. Does not include mod_proxy, yet. Modules that don't work: -mod_digest.c --Anybody have a client that can transmit Digest -Auth? +mod_digest.c -- Don't bother with this until the new + mod_digest code gets incorporated. Questionable modules: (Not tested.) mod_log_agent.c -- Are we keeping these in 2.0 after Ken's @@ -50,15 +46,18 @@ worker threads exit quickly and signal the sigwait thread before the sigwait() call has been entered, the thread never gets the signal. -Open issues: + Min and MaxSpareServers don't allow fine enough granularity to specify + the number of idle threads needed. For example, if ThreadsPerChild is + 200, we probably don't need all 200 threads to be idle, yet that is + the minimum allowed right now. A patch was committed to the pthread + MPM to use Min/MaxSpareThreads instead; it should be moved to this + repository soon if no one has problems. - * Whether every APR routine should return a status code - * What the status codes should look like - * Bringing in the libap stuff +Open issues: * Optimize Optimize Optimize - FEATURE SET FOR APR + FEATURE SET FOR HYBRID SERVER Closed issues: Accept Thread Serialization
cvs commit: apache-apr STATUS
manoj 99/06/28 14:58:28 Modified:.STATUS Log: I hate the sun. Revision ChangesPath 1.27 +7 -1 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.26 retrieving revision 1.27 diff -u -d -u -r1.26 -r1.27 --- STATUS1999/05/26 04:04:32 1.26 +++ STATUS1999/06/28 21:58:27 1.27 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/05/26 04:04:32 $] +Last modified at [$Date: 1999/06/28 21:58:27 $] Release: @@ -42,6 +42,12 @@ Everything Needs patch: + + If SINGLE_LISTEN_UNSERIALIZED_ACCEPT is active, multiple threads are + allowes to call poll() at the same time, with the same pollfd array. + The only solution Manoj can think of right now is to have one pollfd + array per thread, or to just always require serialization (bleh on + both counts). With AIX 4.2, if MaxRequestsPerChild is low (e.g. 5) and ThreadsPerChild is high (e.g. 64), the signal delivered to sigwait
cvs commit: apache-apr STATUS
manoj 99/05/25 21:04:32 Modified:.STATUS Log: 64 * 8 256. Linux was hitting user process limits. Revision ChangesPath 1.26 +7 -7 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.25 retrieving revision 1.26 diff -u -u -r1.25 -r1.26 --- STATUS1999/05/25 05:49:27 1.25 +++ STATUS1999/05/26 04:04:32 1.26 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/05/25 05:49:27 $] +Last modified at [$Date: 1999/05/26 04:04:32 $] Release: @@ -43,12 +43,12 @@ Needs patch: - With Linux 2.2.9 + glibc 2.1.1pre (from RH 6), if MaxRequestsPerChild - is low (e.g. 5) and ThreadsPerChild is high (e.g. 64), some processes - will hang in pthread_create. This is an insane configuration, though. - - With AIX 4.2, if MaxRequestsPerChild is low and ThreadsPerChild is - high, the signal delivered to sigwait gets dropped on the floor. + With AIX 4.2, if MaxRequestsPerChild is low (e.g. 5) and + ThreadsPerChild is high (e.g. 64), the signal delivered to sigwait + gets dropped on the floor. The hypothesis is that the signals blocked + by pthread_sigmask aren't kept pending as they should be, so when + worker threads exit quickly and signal the sigwait thread before the + sigwait() call has been entered, the thread never gets the signal. Open issues:
cvs commit: apache-apr STATUS
manoj 99/05/24 22:49:28 Modified:.STATUS Log: oh well... Revision ChangesPath 1.25 +8 -1 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.24 retrieving revision 1.25 diff -u -u -r1.24 -r1.25 --- STATUS1999/05/24 03:47:39 1.24 +++ STATUS1999/05/25 05:49:27 1.25 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/05/24 03:47:39 $] +Last modified at [$Date: 1999/05/25 05:49:27 $] Release: @@ -42,6 +42,13 @@ Everything Needs patch: + + With Linux 2.2.9 + glibc 2.1.1pre (from RH 6), if MaxRequestsPerChild + is low (e.g. 5) and ThreadsPerChild is high (e.g. 64), some processes + will hang in pthread_create. This is an insane configuration, though. + + With AIX 4.2, if MaxRequestsPerChild is low and ThreadsPerChild is + high, the signal delivered to sigwait gets dropped on the floor. Open issues:
cvs commit: apache-apr STATUS
manoj 99/05/23 20:47:39 Modified:.STATUS Log: I saw Rob Malda take apart the circuitry in his beer mug last night. Revision ChangesPath 1.24 +1 -19 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.23 retrieving revision 1.24 diff -u -u -r1.23 -r1.24 --- STATUS1999/04/29 19:34:19 1.23 +++ STATUS1999/05/24 03:47:39 1.24 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/04/29 19:34:19 $] +Last modified at [$Date: 1999/05/24 03:47:39 $] Release: @@ -42,24 +42,6 @@ Everything Needs patch: - -When the server runs at a high load, and the load later drops off, -the server tries to kill off some children. The problem is that the -children selected may have threads blocked on the accept lock, so -they will never exit. This wasn't a problem with 1.3, because each -process got kicked out of the lock with a signal. This could be solved -(Thanks, Dean) with a server-wide pipe that replaces the SIGWINCH signal. -This would wake up every thread when a single child death signal is sent -over the pipe. - -Related to the above: On Red Hat 5.2, when we SIGTERM the server and -the server is in this state, the children waiting on fcntl will not -die, even though nobody else is holding the lock, verified with lsof. -The children eventually will be SIGKILLed by the parent. This only -happens in USE_MULTI_ACCEPT mode, and the problem goes away when -USE_INTRAPROCESS_SERIALIZED_ACCEPT is undefined. Dean has a couple of -suggestions for tracing down this problem in -[EMAIL PROTECTED] Open issues:
cvs commit: apache-apr STATUS
manoj 99/04/22 00:02:52 Modified:.STATUS Log: saladfork.com is taken. Oh well. Revision ChangesPath 1.21 +14 -2 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.20 retrieving revision 1.21 diff -u -u -r1.20 -r1.21 --- STATUS1999/04/19 04:49:15 1.20 +++ STATUS1999/04/22 07:02:52 1.21 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/04/19 04:49:15 $] +Last modified at [$Date: 1999/04/22 07:02:52 $] Release: @@ -55,7 +55,19 @@ the server tries to kill off some children. The problem is that the children selected may have threads blocked on the accept lock, so they will never exit. This wasn't a problem with 1.3, because each -process got kicked out of the lock with a signal. +process got kicked out of the lock with a signal. This could be solved +(Thanks, Dean) with a server-wide pipe that replaces the SIGWINCH signal. +This would wake up every thread when a single child death signal is sent +over the pipe. + +Related to the above: On Red Hat 5.2, when we SIGTERM the server and +the server is in this state, the children waiting on fcntl will not +die, even though nobody else is holding the lock, verified with lsof. +The children eventually will be SIGKILLed by the parent. This only +happens in USE_MULTI_ACCEPT mode, and the problem goes away when +USE_INTRAPROCESS_SERIALIZED_ACCEPT is undefined. Dean has a couple of +suggestions for tracing down this problem in +[EMAIL PROTECTED] Open issues:
cvs commit: apache-apr STATUS
manoj 99/04/18 21:49:16 Modified:.STATUS Log: I like playing Alpha Centauri. Revision ChangesPath 1.20 +7 -1 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.19 retrieving revision 1.20 diff -u -u -r1.19 -r1.20 --- STATUS1999/04/09 02:31:21 1.19 +++ STATUS1999/04/19 04:49:15 1.20 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/04/09 02:31:21 $] +Last modified at [$Date: 1999/04/19 04:49:15 $] Release: @@ -50,6 +50,12 @@ to the parent process rapidly (but not SIGWINCHes). The suspicion is that this is either an error in how pthread calls are used in the server, or a libc bug (the last refuge of a desperate programer). + +When the server runs at a high load, and the load later drops off, +the server tries to kill off some children. The problem is that the +children selected may have threads blocked on the accept lock, so +they will never exit. This wasn't a problem with 1.3, because each +process got kicked out of the lock with a signal. Open issues:
cvs commit: apache-apr STATUS
manoj 99/04/08 01:53:47 Modified:.STATUS Log: Nasty hot weather is here. It's late this year. Revision ChangesPath 1.18 +13 -1 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.17 retrieving revision 1.18 diff -u -u -r1.17 -r1.18 --- STATUS1999/03/23 22:06:30 1.17 +++ STATUS1999/04/08 08:53:46 1.18 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/23 22:06:30 $] +Last modified at [$Date: 1999/04/08 08:53:46 $] Release: @@ -36,9 +36,21 @@ mod_log_referer.c -- mod_log_config changes. If yes, I'll test -- them later. +Warning scrubbing +Get rid of all reasonable warnings with the strictest compiler flags, +to help weed out bugs + Everything Needs patch: + +On Red Hat 5.2, with a very low MaxRequestsPerChild (e.g. 10), when the +server is banged on with ApacheBench, occasionally, the main thread of a +process hangs around in pthread_exit_process. The size of that process is +always around 11M. This bug is also exhibited by sending lots of SIGHUPs +to the parent process rapidly (but not SIGWINCHes). The suspicion is that +this is either an error in how pthread calls are used in the server, or a +libc bug (the last refuge of a desperate programer). Open issues:
cvs commit: apache-apr STATUS
rbb 99/03/23 05:38:50 Modified:.STATUS Log: Updating the STATUS file again. Revision ChangesPath 1.16 +2 -3 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.15 retrieving revision 1.16 diff -u -r1.15 -r1.16 --- STATUS1999/03/22 15:54:56 1.15 +++ STATUS1999/03/23 13:38:48 1.16 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/22 15:54:56 $] +Last modified at [$Date: 1999/03/23 13:38:48 $] Release: @@ -50,10 +50,8 @@ * Whether every APR routine should return a status code * What the status codes should look like - * Various modules aren't pthreads-ready (mod_cgi) * Bringing in the libap stuff - * Mod_proxy * Optimize Optimize Optimize FEATURE SET FOR APR @@ -94,4 +92,5 @@ mod_usertrack.c mod_unique_id.c mod_cern_meta.c mod_auth.c (I'm assuming mod_auth* also work) mod_mime_magic.c +Mod_proxy
cvs commit: apache-apr STATUS
rbb 99/03/23 14:06:33 Modified:.STATUS Log: Updating the STATUS file. The standard modules are looking pretty good right now for apache-apr. Revision ChangesPath 1.17 +3 -9 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.16 retrieving revision 1.17 diff -u -r1.16 -r1.17 --- STATUS1999/03/23 13:38:48 1.16 +++ STATUS1999/03/23 22:06:30 1.17 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/23 13:38:48 $] +Last modified at [$Date: 1999/03/23 22:06:30 $] Release: @@ -27,21 +27,15 @@ Modules Ensure thread safety. Does not include mod_proxy, yet. Modules that don't work: - mod_status.c --updating the look needs to be finished mod_digest.c --Anybody have a client that can transmit Digest Auth? mod_info.c --DOES NOT WORK AT ALL! - mod_proxy.c --Doesn't even load into the server right now Questionable modules: (Not tested.) mod_log_agent.c -- Are we keeping these in 2.0 after Ken's mod_log_referer.c -- mod_log_config changes. If yes, I'll test -- them later. - -Mod_status - Re-write mod_status to use updated scoreboard for hybrid server. -This is pretty close to completion. - + Everything Needs patch: @@ -91,6 +85,6 @@ mod_log_config.c mod_asis.c mod_rewrite.c mod_mime.c mod_usertrack.c mod_unique_id.c mod_cern_meta.c mod_auth.c (I'm assuming mod_auth* also work) -mod_mime_magic.c +mod_mime_magic.c mod_status.c Mod_proxy
cvs commit: apache-apr STATUS
rbb 99/03/22 07:54:57 Modified:.STATUS Log: Updating the STATUS file again. Revision ChangesPath 1.15 +12 -14apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.14 retrieving revision 1.15 diff -u -r1.14 -r1.15 --- STATUS1999/03/19 16:32:15 1.14 +++ STATUS1999/03/22 15:54:56 1.15 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/19 16:32:15 $] +Last modified at [$Date: 1999/03/22 15:54:56 $] Release: @@ -26,22 +26,11 @@ Modules Ensure thread safety. Does not include mod_proxy, yet. -Modules that work: -mod_env.c mod_negotiation.c mod_autoindex.c mod_dir.c mod_cgi.c - mod_actions.c mod_speling.c mod_userdir.c mod_alias.c -mod_expires.c mod_headers.c mod_so.c mod_setenvif.c -mod_auth_anon.c mod_access.c mod_include.c mod_imap.c -mod_log_config.c mod_asis.c mod_rewrite.c mod_mime.c -mod_usertrack.c mod_unique_id.c mod_cern_meta.c -mod_auth.c (I'm assuming mod_auth* also work) - Modules that don't work: - mod_status.c -- updating the look needs to be finished - mod_mime_magic.c --Doesn't seem to work yet. More testing needed. + mod_status.c --updating the look needs to be finished mod_digest.c --Anybody have a client that can transmit Digest Auth? - mod_info.c --DOES NOT WORK AT ALL! One of the worker -threads actually becomes a zombie! + mod_info.c --DOES NOT WORK AT ALL! mod_proxy.c --Doesn't even load into the server right now Questionable modules: (Not tested.) @@ -96,4 +85,13 @@ Timeouts. (only soft timeouts) Keepalive and Initail Conection timeouts are done. Are there others? CGI? +Modules that work: +mod_env.c mod_negotiation.c mod_autoindex.c mod_dir.c mod_cgi.c +mod_actions.c mod_speling.c mod_userdir.c mod_alias.c +mod_expires.c mod_headers.c mod_so.c mod_setenvif.c +mod_auth_anon.c mod_access.c mod_include.c mod_imap.c +mod_log_config.c mod_asis.c mod_rewrite.c mod_mime.c +mod_usertrack.c mod_unique_id.c mod_cern_meta.c +mod_auth.c (I'm assuming mod_auth* also work) +mod_mime_magic.c
cvs commit: apache-apr STATUS
rbb 99/03/19 08:32:16 Modified:.STATUS Log: I testedd mod_cern_meta again, and it actually does work if the user understands what the directives do. That's one more module down. Revision ChangesPath 1.14 +3 -4 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.13 retrieving revision 1.14 diff -u -r1.13 -r1.14 --- STATUS1999/03/18 21:25:40 1.13 +++ STATUS1999/03/19 16:32:15 1.14 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/18 21:25:40 $] +Last modified at [$Date: 1999/03/19 16:32:15 $] Release: @@ -32,12 +32,11 @@ mod_expires.c mod_headers.c mod_so.c mod_setenvif.c mod_auth_anon.c mod_access.c mod_include.c mod_imap.c mod_log_config.c mod_asis.c mod_rewrite.c mod_mime.c -mod_usertrack.c mod_unique_id.c mod_auth.c (I'm assuming mod_auth* - also work) +mod_usertrack.c mod_unique_id.c mod_cern_meta.c +mod_auth.c (I'm assuming mod_auth* also work) Modules that don't work: mod_status.c -- updating the look needs to be finished - mod_cern_meta.c --I can't get this to work on either 1.3 or apr. mod_mime_magic.c --Doesn't seem to work yet. More testing needed. mod_digest.c --Anybody have a client that can transmit Digest Auth?
cvs commit: apache-apr STATUS
rbb 99/03/18 08:55:29 Modified:.STATUS Log: Updating which modules work and which don't. Revision ChangesPath 1.11 +25 -2 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.10 retrieving revision 1.11 diff -u -r1.10 -r1.11 --- STATUS1999/02/26 19:59:13 1.10 +++ STATUS1999/03/18 16:55:29 1.11 @@ -1,5 +1,5 @@ -Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/26 19:59:13 $] +2Apache Portable Runtime STATUS: +Last modified at [$Date: 1999/03/18 16:55:29 $] Release: @@ -22,6 +22,29 @@ In progress: Modules Ensure thread safety. Does not include mod_proxy, yet. +Modules that work: +mod_env.c mod_negotiation.c mod_autoindex.c mod_dir.c mod_cgi.c + mod_actions.c mod_speling.c mod_userdir.c mod_alias.c +mod_expires.c mod_headers.c mod_so.c mod_setenvif.c +mod_auth_anon.c mod_access.c mod_include.c mod_imap.c +mod_log_config.c mod_asis.c mod_rewrite.c mod_mime.c +mod_usertrack.c mod_unique_id.c mod_auth.c (I'm assuming mod_auth* + also work) + + Modules that don't work: + mod_status.c -- updating the look needs to be finished + mod_cern_meta.c --I can't get this to work on either 1.3 or apr. + mod_mime_magic.c --Doesn't seem to work yet. More testing needed. +mod_digest.c --Anybody have a client that can transmit Digest +Auth? + mod_info.c --DOES NOT WORK AT ALL! One of the worker +threads actually becomes a zombie! + mod_proxy.c --Doesn't even load into the server right now + + Questionable modules: (Not tested.) + mod_log_agent.c -- Are we keeping these in 2.0 after Ken's +mod_log_referer.c -- mod_log_config changes. If yes, I'll test + -- them later. Mod_status Re-write mod_status to use updated scoreboard for hybrid server.
cvs commit: apache-apr STATUS
rbb 99/03/18 10:15:41 Modified:.STATUS Log: Removing a stray 2 from the status file. Oops. Revision ChangesPath 1.12 +2 -2 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.11 retrieving revision 1.12 diff -u -r1.11 -r1.12 --- STATUS1999/03/18 16:55:29 1.11 +++ STATUS1999/03/18 18:15:40 1.12 @@ -1,5 +1,5 @@ -2Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/18 16:55:29 $] +Apache Portable Runtime STATUS: +Last modified at [$Date: 1999/03/18 18:15:40 $] Release:
cvs commit: apache-apr STATUS
rbb 99/03/18 13:25:40 Modified:.STATUS Log: Just updating the STATUS file. Revision ChangesPath 1.13 +5 -6 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.12 retrieving revision 1.13 diff -u -r1.12 -r1.13 --- STATUS1999/03/18 18:15:40 1.12 +++ STATUS1999/03/18 21:25:40 1.13 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/03/18 18:15:40 $] +Last modified at [$Date: 1999/03/18 21:25:40 $] Release: @@ -20,6 +20,10 @@ Available Patches: In progress: +Logging statements +We have some comments along the lines of log this here. +Let's actually make those logs at some point. + Modules Ensure thread safety. Does not include mod_proxy, yet. Modules that work: @@ -50,14 +54,9 @@ Re-write mod_status to use updated scoreboard for hybrid server. This is pretty close to completion. -Check misc thread safety. Make sure UNIX code path mirrors NT path where - appropriate - Everything Needs patch: - * A patch to make the alloc_mutex's scope per-process. It seems to be -causing trouble on Linux at least. Open issues:
cvs commit: apache-apr STATUS
rbb 99/02/26 11:59:13 Modified:.STATUS Log: Update the STATUS file Revision ChangesPath 1.10 +6 -9 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.9 retrieving revision 1.10 diff -u -r1.9 -r1.10 --- STATUS1999/02/17 08:13:13 1.9 +++ STATUS1999/02/26 19:59:13 1.10 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/17 08:13:13 $] +Last modified at [$Date: 1999/02/26 19:59:13 $] Release: @@ -27,13 +27,6 @@ Re-write mod_status to use updated scoreboard for hybrid server. This is pretty close to completion. -Accept Serialization across Processes - Currently, we use one mutex total, we need to change this to be - one mutex per socket. Possibly, we could attach the mutex to the - listen_rec, and pass it down to the accept loop. - -Timeouts. (only soft timeouts) - Check misc thread safety. Make sure UNIX code path mirrors NT path where appropriate @@ -77,5 +70,9 @@ Put (back) in the logic to manage the number of processes. Managing threads per process other than with a configuration constant is in doubt. - +Accept Serialization across Processes + We now use one mutex per socket. +Timeouts. (only soft timeouts) + Keepalive and Initail Conection timeouts are done. Are there others? + CGI?
cvs commit: apache-apr STATUS
manoj 99/02/16 17:08:19 Modified:.STATUS Log: I really have to get this song out of my head. Revision ChangesPath 1.8 +2 -4 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.7 retrieving revision 1.8 diff -u -u -r1.7 -r1.8 --- STATUS1999/02/16 16:04:10 1.7 +++ STATUS1999/02/17 01:08:18 1.8 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/16 16:04:10 $] +Last modified at [$Date: 1999/02/17 01:08:18 $] Release: @@ -25,14 +25,12 @@ Mod_status Re-write mod_status to use updated scoreboard for hybrid server. +This is pretty close to completion. Accept Serialization across Processes Currently, we use one mutex total, we need to change this to be one mutex per socket. Possibly, we could attach the mutex to the listen_rec, and pass it down to the accept loop. - -Memory leakage - It looks like we are leaking, we need to find out why and where. Timeouts. (only soft timeouts)
cvs commit: apache-apr STATUS
rbb 99/02/16 08:04:11 Modified:.STATUS Log: Update the STATUS file.Update the STATUS file. Revision ChangesPath 1.7 +24 -13apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.6 retrieving revision 1.7 diff -u -r1.6 -r1.7 --- STATUS1999/02/09 22:10:30 1.6 +++ STATUS1999/02/16 16:04:10 1.7 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/09 22:10:30 $] +Last modified at [$Date: 1999/02/16 16:04:10 $] Release: @@ -20,21 +20,24 @@ Available Patches: In progress: -Signal Handling -Investigate signal handling in Apache-pthreads. SIGUSR1 is not a - usable signal on Linux 2.0, because pthreads uses it to communicate - between threads. - Modules Ensure thread safety. Does not include mod_proxy, yet. -Process/thread management -Put (back) in the logic to manage the number of processes. - Managing threads per process other than with a configuration - constant is in doubt. +Mod_status + Re-write mod_status to use updated scoreboard for hybrid server. + +Accept Serialization across Processes + Currently, we use one mutex total, we need to change this to be + one mutex per socket. Possibly, we could attach the mutex to the + listen_rec, and pass it down to the accept loop. + +Memory leakage + It looks like we are leaking, we need to find out why and where. Timeouts. (only soft timeouts) +Check misc thread safety. Make sure UNIX code path mirrors NT path where + appropriate Everything @@ -48,14 +51,12 @@ * Bringing in the libap stuff * Mod_proxy - * Check misc thread safety. Make sure UNIX code path mirrors NT path where -appropriate * Optimize Optimize Optimize FEATURE SET FOR APR Closed issues: -Accept Serialization +Accept Thread Serialization Implemented as a blocking queue. 1 accept thread passes connection to queue (size of queue is config directive, default # of threads). All other threads block on queue. @@ -68,3 +69,13 @@ Scoreboard Design the scoreboard appropriately for a hybrid process web server. This includes adding a tid field to the scoreboard. +Signal Handling +Investigate signal handling in Apache-pthreads. SIGUSR1 is not a + usable signal on Linux 2.0, because pthreads uses it to communicate + between threads. SIGWINCH was used for graceful restarts. +Process/thread management +Put (back) in the logic to manage the number of processes. + Managing threads per process other than with a configuration + constant is in doubt. + +
cvs commit: apache-apr STATUS
rbb 99/02/09 14:10:30 Modified:.STATUS Log: Update to the status file. Revision ChangesPath 1.6 +10 -9 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.5 retrieving revision 1.6 diff -u -r1.5 -r1.6 --- STATUS1999/02/08 06:32:10 1.5 +++ STATUS1999/02/09 22:10:30 1.6 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/08 06:32:10 $] +Last modified at [$Date: 1999/02/09 22:10:30 $] Release: @@ -21,21 +21,21 @@ Available Patches: In progress: Signal Handling -Investigate signal handling in Apache-pthreads. Currently, we kill -the process when we encounter SIGPIPE, obviously bad. +Investigate signal handling in Apache-pthreads. SIGUSR1 is not a + usable signal on Linux 2.0, because pthreads uses it to communicate + between threads. Modules Ensure thread safety. Does not include mod_proxy, yet. -Scoreboard - Resign the scoreboard appropriately for a hybrid process web - server. This includes adding a tid field to the scoreboard. - Process/thread management Put (back) in the logic to manage the number of processes. Managing threads per process other than with a configuration constant is in doubt. +Timeouts. (only soft timeouts) + + Everything Needs patch: @@ -48,7 +48,6 @@ * Bringing in the libap stuff * Mod_proxy - * Timeouts. (only soft timeouts) * Check misc thread safety. Make sure UNIX code path mirrors NT path where appropriate * Optimize Optimize Optimize @@ -66,4 +65,6 @@ it again. Bringing this code up-to-date with the changes in the 1.3 branch. It's now up-to-date as of the morning of Feb 6th, 1999. - +Scoreboard + Design the scoreboard appropriately for a hybrid process web + server. This includes adding a tid field to the scoreboard.
cvs commit: apache-apr STATUS
manoj 99/02/07 22:32:11 Modified:.STATUS Log: I want winter back. Revision ChangesPath 1.5 +5 -5 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.4 retrieving revision 1.5 diff -u -u -r1.4 -r1.5 --- STATUS1999/02/06 08:01:16 1.4 +++ STATUS1999/02/08 06:32:10 1.5 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/06 08:01:16 $] +Last modified at [$Date: 1999/02/08 06:32:10 $] Release: @@ -46,12 +46,11 @@ * What the status codes should look like * Various modules aren't pthreads-ready (mod_cgi) * Bringing in the libap stuff - * Bringing this code up-to-date with the changes in the 1.3 branch. -The trees appear to have diverged in November(?!) * Mod_proxy * Timeouts. (only soft timeouts) - * Check misc thread safety. Make sure UNIX code path mirrors NT path. + * Check misc thread safety. Make sure UNIX code path mirrors NT path where +appropriate * Optimize Optimize Optimize FEATURE SET FOR APR @@ -65,5 +64,6 @@ put monitor process back into place. Was removed for initial port, it's time we make sure the server is running, and if not, start it again. - +Bringing this code up-to-date with the changes in the 1.3 branch. +It's now up-to-date as of the morning of Feb 6th, 1999.
cvs commit: apache-apr STATUS
manoj 99/02/06 00:01:17 Modified:.STATUS Log: No Scooby-Doo on TV tonight. Revision ChangesPath 1.4 +10 -5 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.3 retrieving revision 1.4 diff -u -u -r1.3 -r1.4 --- STATUS1999/02/05 14:58:46 1.3 +++ STATUS1999/02/06 08:01:16 1.4 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/02/05 14:58:46 $] +Last modified at [$Date: 1999/02/06 08:01:16 $] Release: @@ -25,13 +25,16 @@ the process when we encounter SIGPIPE, obviously bad. Modules -Ensure thread safety. Does not include mod_proxy. +Ensure thread safety. Does not include mod_proxy, yet. Scoreboard - Update the scoreboard for each thread, not per process. + Resign the scoreboard appropriately for a hybrid process web + server. This includes adding a tid field to the scoreboard. -Hybrid server - Use the updated server to determine when to launch new processes. +Process/thread management +Put (back) in the logic to manage the number of processes. + Managing threads per process other than with a configuration + constant is in doubt. Everything @@ -43,6 +46,8 @@ * What the status codes should look like * Various modules aren't pthreads-ready (mod_cgi) * Bringing in the libap stuff + * Bringing this code up-to-date with the changes in the 1.3 branch. +The trees appear to have diverged in November(?!) * Mod_proxy * Timeouts. (only soft timeouts)
cvs commit: apache-apr STATUS
rbb 99/02/05 06:58:48 Modified:.STATUS Log: Update to the STATUS file to reflect the work done recently on making this a hybrid server. Revision ChangesPath 1.3 +16 -9 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.2 retrieving revision 1.3 diff -u -r1.2 -r1.3 --- STATUS1999/01/26 20:28:30 1.2 +++ STATUS1999/02/05 14:58:46 1.3 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/01/26 20:28:30 $] +Last modified at [$Date: 1999/02/05 14:58:46 $] Release: @@ -20,11 +20,6 @@ Available Patches: In progress: -Accept Serialization -Implemented as a blocking queue. 1 accept thread passes connection -to queue (size of queue is config directive, default # of threads). -All other threads block on queue. - Signal Handling Investigate signal handling in Apache-pthreads. Currently, we kill the process when we encounter SIGPIPE, obviously bad. @@ -32,6 +27,12 @@ Modules Ensure thread safety. Does not include mod_proxy. +Scoreboard + Update the scoreboard for each thread, not per process. + +Hybrid server + Use the updated server to determine when to launch new processes. + Everything Needs patch: @@ -46,12 +47,18 @@ * Mod_proxy * Timeouts. (only soft timeouts) * Check misc thread safety. Make sure UNIX code path mirrors NT path. - * Process/Thread hybridization - * put monitor process back into place. Was removed for initial port, it's -time we make sure the server is running, and if not, start it again. * Optimize Optimize Optimize FEATURE SET FOR APR Closed issues: +Accept Serialization +Implemented as a blocking queue. 1 accept thread passes connection +to queue (size of queue is config directive, default # of threads). +All other threads block on queue. +Process/Thread hybridization +put monitor process back into place. Was removed for initial port, + it's time we make sure the server is running, and if not, start + it again. +
cvs commit: apache-apr STATUS
rbb 99/01/26 12:28:31 Modified:.STATUS Log: Update STATUS file for work being done in pthreads port Revision ChangesPath 1.2 +20 -1 apache-apr/STATUS Index: STATUS === RCS file: /home/cvs/apache-apr/STATUS,v retrieving revision 1.1 retrieving revision 1.2 diff -u -r1.1 -r1.2 --- STATUS1999/01/21 23:14:25 1.1 +++ STATUS1999/01/26 20:28:30 1.2 @@ -1,5 +1,5 @@ Apache Portable Runtime STATUS: -Last modified at [$Date: 1999/01/21 23:14:25 $] +Last modified at [$Date: 1999/01/26 20:28:30 $] Release: @@ -20,7 +20,18 @@ Available Patches: In progress: +Accept Serialization +Implemented as a blocking queue. 1 accept thread passes connection +to queue (size of queue is config directive, default # of threads). +All other threads block on queue. + +Signal Handling +Investigate signal handling in Apache-pthreads. Currently, we kill +the process when we encounter SIGPIPE, obviously bad. +Modules +Ensure thread safety. Does not include mod_proxy. + Everything Needs patch: @@ -31,6 +42,14 @@ * What the status codes should look like * Various modules aren't pthreads-ready (mod_cgi) * Bringing in the libap stuff + + * Mod_proxy + * Timeouts. (only soft timeouts) + * Check misc thread safety. Make sure UNIX code path mirrors NT path. + * Process/Thread hybridization + * put monitor process back into place. Was removed for initial port, it's +time we make sure the server is running, and if not, start it again. + * Optimize Optimize Optimize FEATURE SET FOR APR
cvs commit: apache-apr STATUS
coar99/01/21 15:14:26 Added: .STATUS Log: Add a STATUS file for recording what goes on with the APR stuff. Revision ChangesPath 1.1 apache-apr/STATUS Index: STATUS === Apache Portable Runtime STATUS: Last modified at [$Date: 1999/01/21 23:14:25 $] Release: 0.1 : In pre-alpha development Plan: Everyone discuss and wrangle out the issues on new-httpd, and commit changes as appropriate. Showstoppers: Committed Code Changes: . docs/ directory created and initial function pre-spec added . include/ directory created and start of APRStatus macros added . pthreads/ directory tree created and populated with latest work on Apache 1.3.5-dev port to pthreads inserted Available Patches: In progress: Everything Needs patch: Open issues: * Whether every APR routine should return a status code * What the status codes should look like * Various modules aren't pthreads-ready (mod_cgi) * Bringing in the libap stuff FEATURE SET FOR APR Closed issues:
