Hi again! I tried turning postfix and apache off and run shredder again. What I found out is really strange.
During this gap (again - it occurs right before each top level object is processed) there are no processes in mysql, i.e. the SHOW PROCESSLIST command in mysql shows sleeping connection from rt_user. However, when I run the top command in shell, there is a perl process that eats 50% of CPU!!! And it does that precisely during this gap (20 seconds average). When the Shredder warnings start appearing, the mysqld process takes over CPU and RAM. When the top level object is wiped out, before going to the next one, Shredder again somehow calls the perl process and so on. Ruslan, please, tell me why does this happen? I never mentioned that, but we do have a clone RT with only 500 tickets for scrip testing and on that instance the Shredder runs almost instantly. I would really appreciate your advice on 'perl' gap thing. Kostya. -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Konstantin Naryzhniy Sent: Friday, March 16, 2007 7:09 PM To: 'Ruslan Zakirov' Cc: [email protected] Subject: RE: [rt-users] RTx::Shredder time gap between tickets wipeout So, well, do you suggest that I reinstall Shredder? Not quite sure how to uninstall it once it's up... If you commit things once - what event do "warning <object-id> wiped out" lines refer to in the log? And how many times the dump file is written per one top level object? I also plan to try running Shredder with Postfix and Apache off. I slightly hope that that'll speed things up a bit, since there's going to be no queries committed simultaneously with Shredder. I'll report on results of this. Ruslan, thanks again for your great tool! Kostya -----Original Message----- From: Ruslan Zakirov [mailto:[EMAIL PROTECTED] Sent: Friday, March 16, 2007 6:56 PM To: Konstantin Naryzhniy Cc: [email protected] Subject: Re: [rt-users] RTx::Shredder time gap between tickets wipeout On 3/16/07, Konstantin Naryzhniy <[EMAIL PROTECTED]> wrote: > Hello Ruslan! > Thanks for your rapid reply. > > I have added one of the indices you've suggested (we do not have Custom > Fields yet and the ACL table is relatively small and doesn't appear in slow > queries log). > > I do not think that the delay is caused by the committing transaction. > Actually, the gap has to deal with the previous ticket, rather that the next > one. I mean that this delay first happens right after when I confirm the "Do > you want to proceed? [y/N]" prompt and before the first warning line > appears. Huh, that's kinda a ringing bell that's I've screwed up something and that's actually problem in the shredder. > > Could you please explain what happens before the first warning is sent to > output? I am no programmer to dig into the source code)) > > Are the DELETE statement in the Shredder committed one by one or are they > committed in a single "package" after all the records in DB tables were > found? I commit things once per ticket or per user, depends on what top level objects are. Top level objects are objects you select with a plugin. > > thanks, > Kostya _______________________________________________ http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users Community help: http://wiki.bestpractical.com Commercial support: [EMAIL PROTECTED] Discover RT's hidden secrets with RT Essentials from O'Reilly Media. Buy a copy at http://rtbook.bestpractical.com _______________________________________________ http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users Community help: http://wiki.bestpractical.com Commercial support: [EMAIL PROTECTED] Discover RT's hidden secrets with RT Essentials from O'Reilly Media. Buy a copy at http://rtbook.bestpractical.com
