Long apache restart time after upgrading to plesk 10.4.4

[pichlo]

Hey Guys,

after upgrading to Plesk 10.4.4 (and many other problems) there's still one big problem left. Everytime when I change settings regarding the apache configuration in plesk, apache is offline for several seconds. I've checked the apache error log /var/log/apache2/error.log while creating a new subdomain. Here's the result:

[Thu Feb 23 22:04:28 2012] [notice] caught SIGTERM, shutting down
(...)
[Thu Feb 23 22:05:05 2012] [notice] Apache/2.2.9 (Debian) DAV/2 mod_fcgid/2.3.6 mod_python/3.3.1 Python/2.5.2 mod_ssl/2.2.9 OpenSSL/0.9.8g mod_perl/2.0.4 Perl/v5.10.0 configured -- resuming normal operations

So apache or what else needs 37 seconds (!) to restart. So I ask myself what the system does in this time. So I take the time reconfiguring the same domain by httpdmng:

time /usr/local/psa/admin/bin/httpdmng --reconfigure-domain <domain>.de
real 0m37.544s

Maybe this are only two equal but indepedant numbers. But the other option is, that plesk stops apache before reconfiguring the domain. This would explain the systems behavour...... In this case Parallels have switch the order of these operations....

Could everyone confirm this behavour for others plesk installations?
Does anyone know the order of plesks internal operations?
Have Parallels changed these order for Plesk 10.4.4?
Does anyone know how I could reduce these time period?

Regards
Martin

 

[pichlo]

No Answer?

Hey Guys - especially from parallels,

no ideas, no answer?

Martin

 

[IgorG]

Try to enable Apache graceful restart instead of typical one as per KB article http://kb.parallels.com/en/112020
Check also Apache restart interval in Plesk Control Panel GUI: Tools and settings -> Server settings -> Apache restart interval

 

[pichlo]

This seems to work. Thank you.