0

I am seeing a strange error with my ubuntu dev machine. It is going on since almost a year. Meanwhile I have reinstalled ubuntu at every release (not just upgraded to the next release).

Every day when I turn on my machine I see this errors in apache's error log, and my local websites seems awfully slow:

[core:notice]  AH00051: child pid 8467 exit signal Segmentation fault (11), possible coredump in /etc/apache2
[core:error]  AH00546: no record of generation 0 of exiting child 8467
[core:notice]  AH00051: child pid 8469 exit signal Segmentation fault (11), possible coredump in /etc/apache2
[core:error]  AH00546: no record of generation 0 of exiting child 8469
[core:notice]  AH00051: child pid 8479 exit signal Segmentation fault (11), possible coredump in /etc/apache2
[core:error]  AH00546: no record of generation 0 of exiting child 8479
[core:notice]  AH00051: child pid 8481 exit signal Segmentation fault (11), possible coredump in /etc/apache2
[core:error]  AH00546: no record of generation 0 of exiting child 8481
[core:notice]  AH00051: child pid 8482 exit signal Segmentation fault (11), possible coredump in /etc/apache2
[core:error]  AH00546: no record of generation 0 of exiting child 8482
[core:notice]  AH00051: child pid 8485 exit signal Segmentation fault (11), possible coredump in /etc/apache2
[core:error]  AH00546: no record of generation 0 of exiting child 8485

This goes on until I do a sudo service apache2 restart: after this command apache starts normally and everything works.

I've said about midnight because it happens every day when I turn on my pc, but because of a huge amout of work I had to work for a few days after midnight, and I have noticed that even with my pc already on, this error comes exactly at midnight.

The only thing that comes to my mind is logrotate, which is set to run daily for apache2. I always edit this config file to make the apache's logs user readable and writeable (I need it for a PHP log reader).

/var/log/apache2/*.log {
    daily
    missingok
    rotate 14
    compress
    delaycompress
    notifempty
    create 777 root adm
    sharedscripts
    postrotate
                if invoke-rc.d apache2 status > /dev/null 2>&1; then \
                    invoke-rc.d apache2 reload > /dev/null 2>&1; \
                fi;
    endscript
    prerotate
        if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
            run-parts /etc/logrotate.d/httpd-prerotate; \
        fi; \
    endscript
}

Could this be the reason? The only thing that I have modified is that create 777 root adm line...

  • Check the cron table – Romeo Ninov Feb 10 at 8:50
  • It's empty, for both my user and root. – ToX 82 Feb 10 at 13:50
  • Set LogLevel to debug. If that doesn't reveal the problem, follow the Apache Debugging Guide. – Freddy Feb 10 at 15:34
0

Find the crash dump and get a stack trace with human readable symbols from it.

Install debug symbol packages for at least httpd. Install gdb.

Determine where core dumps go. See man core for more details.

cat /proc/sys/kernel/core_pattern

If prefixed with a pipe (|), you have a crash processing system and reports probably save to /var/crash/. Get the trace with apport-retrace --stdout

If a file name pattern, core dumps go there. Read the gdb manual on how to load a core file.

With the symbols, you can tell where the crash is, and can report this to your Apache HTTPD support channel.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.