| Summary: | No such process: cgid daemon is gone; is Apache terminating? | ||
|---|---|---|---|
| Product: | Apache httpd-2 | Reporter: | David Boldt <dboldt> |
| Component: | mod_cgi | Assignee: | Apache HTTPD Bugs Mailing List <bugs> |
| Status: | CLOSED DUPLICATE | ||
| Severity: | major | ||
| Priority: | P3 | ||
| Version: | 2.0.47 | ||
| Target Milestone: | --- | ||
| Hardware: | Sun | ||
| OS: | Solaris | ||
| URL: | http://water.usgs.gov/ | ||
Resolving the problem that caused the cgid daemon to exit: The upcoming 2.0.48 resolves one particular bug that can cause the cgid daemon to exit on Solaris. Please let us know after moving to 2.0.48 whether or not the cgid daemon still exits intermittently. Recovering after such a problem: The patch below causes the cgid daemon to restart automatically after such a problem. The patch has been committed to Apache 2.1-dev but not yet merged into the stable tree because it relies on a behavior specific to the worker MPM. I would not hesitate to use the patch on Solaris with the worker MPM though. This should resolve the long-term affects of whatever problem causes the cgid daemon to exit. http://cvs.apache.org/viewcvs.cgi/httpd-2.0/modules/generators/mod_cgid.c.diff?r1=1.151&r2=1.152 Should that patch be used on top of 2.0.48, or is that change already in 2.0.48? The patch is still not in stable branch (2.0.x) because it has an inappropriate dependency on some behavior of the worker MPM. If you can't get the patch to apply to the latest mod_cgid from stable branch, let me know and I'll post a new one. patch pointed to previously is still appropriate for use with stable branch this problem was fixed in Apache 2.1-dev some time back with PR 19849, but a couple of issues prevented it from being merged into stable branch... those are now resolved, so hopefully the fix can appear in a 2.0 release before too terribly long *** This bug has been marked as a duplicate of 19849 *** |
Solaris 8 Server version: Apache/2.0.47 Server built: Jul 21 2003 00:32:49 Server's Module Magic Number: 20020903:4 Architecture: 32-bit Server compiled with.... -D APACHE_MPM_DIR="server/mpm/worker" -D APR_HAS_MMAP -D APR_HAVE_IPV6 (IPv4-mapped addresses enabled) -D APR_USE_PROC_PTHREAD_SERIALIZE -D APR_USE_PTHREAD_SERIALIZE -D SINGLE_LISTEN_UNSERIALIZED_ACCEPT -D APR_HAS_OTHER_CHILD -D AP_HAVE_RELIABLE_PIPED_LOGS -D HTTPD_ROOT="/usr/opt/apache" -D SUEXEC_BIN="/usr/opt/apache/bin/suexec" -D DEFAULT_SCOREBOARD="logs/apache_runtime_status" -D DEFAULT_ERRORLOG="logs/error_log" -D AP_TYPES_CONFIG_FILE="conf/mime.types" -D SERVER_CONFIG_FILE="conf/httpd.conf" Compiled in modules: core.c mod_access.c mod_auth.c mod_include.c mod_log_config.c mod_logio.c mod_setenvif.c worker.c http_core.c mod_mime.c mod_autoindex.c mod_cgid.c mod_negotiation.c mod_dir.c mod_userdir.c mod_alias.c mod_rewrite.c mod_so.c Web site routes all HTML documents through a parsing program written in Perl, (written before php was born, for similar reasons). Action text/html /cgi-bin/h2o.cgi Action text/html;charset=iso-8859-1 /cgi-bin/h2o.cgi The site receives an average of roughly 300K hits per day, and roughly every other day CGI processing crashes with messages such as these (messages trimmed to one per crash): water.usgs.gov-error_log-2003-09-10:[Wed Sep 10 14:47:21 2003] [error] [client 66.123.2.43] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/lookup/get, referer: http://pubs.water.usgs.gov/TWRI3A18/ water.usgs.gov-error_log-2003-09-12:[Fri Sep 12 20:36:37 2003] [error] [client 68.211.74.103] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/h2o.cgi water.usgs.gov-error_log-2003-09-14:[Sun Sep 14 22:57:22 2003] [error] [client 216.88.158.142] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/h2o.cgi water.usgs.gov-error_log-2003-09-16:[Tue Sep 16 04:10:50 2003] [error] [client 218.5.149.36] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/h2o.cgi water.usgs.gov-error_log-2003-09-28:[Sun Sep 28 16:13:10 2003] [error] [client 65.81.33.33] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/h2o.cgi water.usgs.gov-error_log-2003-09-29:[Mon Sep 29 10:54:20 2003] [error] [client 66.196.72.14] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/pointinhuc.pl water.usgs.gov-error_log-2003-09-29:[Mon Sep 29 13:30:58 2003] [error] [client 64.68.82.143] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/h2o.cgi water.usgs.gov-error_log-2003-09-29:[Mon Sep 29 13:30:59 2003] [error] [client 130.11.160.96] (3)No such process: cgid daemon is gone; is Apache terminating?: /www/errors/include/bottom.html, referer: http://water.usgs.gov/usgs/wrd_table.html water.usgs.gov-error_log-2003-09-29:[Mon Sep 29 13:31:00 2003] [error] [client 64.68.82.170] (3)No such process: cgid daemon is gone; is Apache terminating?: /afs/usgs.gov/www/water/cgi-bin/h2o.cgi Since all web pages filter through CGI, this effectively crashes the site.