Bug#493573: apache2: Restart and reload does not work
Hi, I was on computerless holiday so I could not follow the bug reports and do any debugging but it seems that upgrading apache2, php5, mysql-common and their dependencies to current testing solved the problem. Thanks for all help. I think this bug report may be closed, with notice to do an upgrade. -- Regards Vladislav Kurz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493573: apache2: Restart and reload does not work
Hello, Which processes are still running? The one running as root or those running as www-data or both? Both of them. One as root and several itch children running as www-data. You can also try (as root) lsof -i -P|egrep TCP.*:80 Result: apache2 11032 root3u IPv6 3029567 TCP *:80 (LISTEN) So it is apache. -- Regards Vladislav Kurz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493573: apache2: Restart and reload does not work
Hi, we did some more testing and googling and foud thet this issue is discussed on apache-dev mailing list. They mention that the problem is related to proxy setting, and yes, we use mod_proxy too. There are some patches too. http://www.gossamer-threads.com/lists/apache/dev/355044?page=last We did strace, the last line is: futex(0x4447a700, FUTEX_WAIT, 2, NULL if you want I can send the whole strace -- Regards Vladislav Kurz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493573: apache2: Restart and reload does not work
Hi, On Thursday 07 August 2008, Vladislav Kurz wrote: we did some more testing and googling and foud thet this issue is discussed on apache-dev mailing list. They mention that the problem is related to proxy setting, and yes, we use mod_proxy too. There are some patches too. http://www.gossamer-threads.com/lists/apache/dev/355044?page=last That thread is about a bug in the worker mpm in trunk. It says that 2.2.x is not affected. Therefore this is probably not your problem. We did strace, the last line is: futex(0x4447a700, FUTEX_WAIT, 2, NULL if you want I can send the whole strace When apache is not answering to requests, start top and type fy uwww-data What info is in the WCHAN column? Get the pid of a process that has not semtimedo in the WCHAN column (if any) and strace that process and try to do a request. It might also be interesting to see the strace output of a child process while you do a graceful restart. Cheers, Stefan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493573: apache2: Restart and reload does not work
Package: apache2.2-common Version: 2.2.9-6 Severity: important After upgrade to lenny apache started to fail after logrotate. I found that after reload or restart it was still running, listening on network socket but not answering reqests. The only way was to kill it with kill -9. Log for reload says just: [notice] Graceful restart requested, doing restart Log for restart says this: [error] child process 27290 still did not exit, sending a SIGKILL [error] child process 27291 still did not exit, sending a SIGKILL [error] child process 27292 still did not exit, sending a SIGKILL [error] child process 27293 still did not exit, sending a SIGKILL [error] child process 27294 still did not exit, sending a SIGKILL (the above lines were repeated 4 times) [notice] caught SIGTERM, shutting down After writing the notice about restart or shutdown, the old apache process was still running, but not processing requests. New apache process could not be started because ports were occupied: (98)Address already in use: make_sock: could not bind to address [::]:80 (98)Address already in use: make_sock: could not bind to address 0.0.0.0:80 no listening sockets available, shutting down Unable to open logs failed! I couldnt find any more info in logs. If you point me how to get more debug info I'l provide it. Regards Vladislav Kurz -- Package-specific info: List of enabled modules from 'apache2 -M': actions alias auth_basic authn_file authz_default authz_groupfile authz_host authz_user autoindex cgi deflate dir env expires ext_filter imagemap include mime_magic mime negotiation php5 proxy_connect proxy_ftp proxy_http proxy rewrite setenvif ssl status suexec userdir vhost_alias -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable'), (60, 'testing'), (30, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18-6-686-bigmem Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Versions of packages apache2 depends on: ii apache2-mpm-prefork 2.2.9-6Apache HTTP Server - traditional n apache2 recommends no packages. Versions of packages apache2.2-common depends on: ii apache2-utils 2.2.9-5 utility programs for webservers ii libapr1 1.2.7-8.2The Apache Portable Runtime Librar ii libaprutil1 1.2.12+dfsg-7The Apache Portable Runtime Utilit ii libc6 2.7-10 GNU C Library: Shared libraries ii libmagic1 4.17-5etch3 File type determination library us ii libssl0.9.8 0.9.8g-10SSL shared libraries ii lsb-base3.1-23.2etch1Linux Standard Base 3.1 init scrip ii mime-support3.39-1 MIME files 'mime.types' 'mailcap ii net-tools 1.60-17 The NET-3 networking toolkit ii perl5.10.0-11Larry Wall's Practical Extraction ii procps 1:3.2.7-3/proc file system utilities ii zlib1g 1:1.2.3.3.dfsg-5 compression library - runtime -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493573: apache2: Restart and reload does not work
On Sunday 03 August 2008, Vladislav Kurz wrote: Log for restart says this: [error] child process 27290 still did not exit, sending a SIGKILL [error] child process 27291 still did not exit, sending a SIGKILL [error] child process 27292 still did not exit, sending a SIGKILL [error] child process 27293 still did not exit, sending a SIGKILL [error] child process 27294 still did not exit, sending a SIGKILL (the above lines were repeated 4 times) [notice] caught SIGTERM, shutting down After writing the notice about restart or shutdown, the old apache process was still running, but not processing requests. Which processes are still running? The one running as root or those running as www-data or both? New apache process could not be started because ports were occupied: (98)Address already in use: make_sock: could not bind to address [::]:80 (98)Address already in use: make_sock: could not bind to address 0.0.0.0:80 no listening sockets available, shutting down Unable to open logs failed! I couldnt find any more info in logs. If you point me how to get more debug info I'l provide it. You can also try (as root) lsof -i -P|egrep TCP.*:80 to determine if it is really apache that is keeping the listening sockets open. If not you might be running into bug #366124. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
