[11.25 Current] apache restarting every hour

[iscandr]

Hello everyone,

After an update to 11.25 apache gracefully restarts every hour. Log rotation and service monitor are disabled. I am unable to track what is causing these restarts.

/usr/local/apache/logs/error_log

[notice] Graceful restart requested, doing restart
[notice] mod_bw : Memory Allocated 0 bytes (each conf takes 28 bytes)
[notice] mod_bw : Version 0.8 - Initialized [0 Confs]
[notice] Apache/2.2.13 (Unix) mod_ssl/2.2.13 OpenSSL/0.9.8e-fips-rhel5 mod_auth_passthrough/2.1 mod_bwlimited/1.4 FrontPage/5.0.2.2635 PHP/5.2.11 mod_perl/2.0.4 Perl/v5.8.8 configured -- resuming normal operations

Do you have any suggestions?

[thewebhosting]

Please make sure that you have not set the cronjobs to restart apache every hour?

[iscandr]

No such cronjob exists.

[bvierra]

any updates to subdomains, adding removal of accounts? anything that would touch apache?

[iscandr]

Nothing touched apache.

/scripts/fixeverything seems to have fixed the issue.

apache is now up 2+ hours, running smooth.

[Reado]

Parent Server Generation: 324

324 graceful restarts in 4 days !

[Reado]

[Mon Oct 12 22:27:55 2009] [notice] Graceful restart requested, doing restart
[Mon Oct 12 22:27:57 2009] [notice] Apache/2.2.13 (Unix) mod_ssl/2.2.13 OpenSSL/0.9.8e-fips-rhel5 mod_bwlimited/1.4 configured -- resuming normal operations
[Mon Oct 12 22:44:36 2009] [notice] Graceful restart requested, doing restart
[Mon Oct 12 22:44:37 2009] [notice] Apache/2.2.13 (Unix) mod_ssl/2.2.13 OpenSSL/0.9.8e-fips-rhel5 mod_bwlimited/1.4 configured -- resuming normal operations

I haven't requested a graceful restart !!!

Seems to be doing this frequently, every 15 minutes or so!

[Reado]

This is causing me serious problems - according to Google Analytics Sunday we received lower-than-usual visitors to all sites. Yesterday (which is usually our busiest day) we received ZERO visitors.

After checking the Apache logs it appears to be restarting every 8 to 15 minutes.

The restarts I experienced yesterday weren't just graceful restarts, but:

[Tue Oct 13 09:08:43 2009] [notice] Graceful restart requested, doing restart
[Tue Oct 13 09:08:43 2009] [error] (9)Bad file descriptor: apr_socket_accept: (client socket)
[Tue Oct 13 09:08:44 2009] [notice] Apache/2.2.13 (Unix) mod_ssl/2.2.13 OpenSSL/0.9.8e-fips-rhel5 mod_bwlimited/1.4 configured -- resuming normal operations
[Tue Oct 13 09:17:04 2009] [notice] Graceful restart requested, doing restart
[Tue Oct 13 09:17:05 2009] [notice] Apache/2.2.13 (Unix) mod_ssl/2.2.13 OpenSSL/0.9.8e-fips-rhel5 mod_bwlimited/1.4 configured -- resuming normal operations
[Tue Oct 13 09:17:06 2009] [notice] Graceful restart requested, doing restart
[Tue Oct 13 09:17:06 2009] [error] (9)Bad file descriptor: apr_socket_accept: (client socket)
[Tue Oct 13 09:17:07 2009] [notice] Apache/2.2.13 (Unix) mod_ssl/2.2.13 OpenSSL/0.9.8e-fips-rhel5 mod_bwlimited/1.4 configured -- resuming normal operations

[Infopro]

I have this going on here as well. Searching my logs for "Graceful restart requested" it's found in every syscheck email since upgrading to 11.25 on Oct 10.

[cPanelKenneth]

We should have a fix for this in tonights 11.25 build ( as long as all goes as planned ).

[hostmedic]

was the update in last nights build - and if so ...
do we need to be on edge to grab it -
and ... if so ...
do we need to recompile or do anything else afterwards

Thanks for all your hard work in fixing this issue btw -

[cPanelNick]

This is now published in CURRENT 40255. We expect to be able to reduce the restarts some more before RELEASE.

[Cristi4n]

As far as I can see they are doing a graceful restart for every cpanel account. So, now I have for example 800 apache restarts in a few hours when cpanellogd comes is for all my servers with cPanel current.
I am very happy with this, this is everything I ever wanted from cPanel, every night my servers with cPanel current becomes inaccesible.
I love you guys.

Give us a fix for this asap please, I can't take it anymore. Even that stupid, bad written Plesk is better in processing logs, maby you can take a look at how they did it. Come on!

I already have 40255 and a lot of graceful restarts, as usual.

I replaced /usr/local/cpanel/bin/safeapacherestart , maby this is better than having the server offline.

[Reado]

As far as I can see they are doing a graceful restart for every cpanel account. So, now I have for example 800 apache restarts in a few hours when cpanellogd comes is for all my servers with cPanel current.
I am very happy with this, this is everything I ever wanted from cPanel, every night my servers with cPanel current becomes inaccesible.
I love you guys.

Give us a fix for this asap please, I can't take it anymore. Even that stupid, bad written Plesk is better in processing logs, maby you can take a look at how they did it. Come on!

I already have 40255 and a lot of graceful restarts, as usual.

There's nothing to stop you from moving to RELEASE is there? I did it as there were other issues with CURRENT and I no longer have the problem.

[Infopro]

I have this going on here as well. Searching my logs for "Graceful restart requested" it's found in every syscheck email since upgrading to 11.25 on Oct 10.

Since midnight till this post (about 4:30PM) I've found this 11 times today. Yesterday, 14. Day before, 13, total.

Why not downgrade to release or stable and stay outta trouble? CURRENT is always expected to have a bit more going on.