Created attachment 364932 [details] test script Description of problem: When I use Fedora::Bugzilla->get_bug() to get a big BZ (e.g. with 4MB attachment), it gets stuck after a while and top shows ~99% cpu usage for this script. I use bz504799 to trigger this. Version-Release number of selected component (if applicable): 0.10-1 - I tried 0.13 from CPAN and problem is still there. How reproducible: always Steps to Reproduce: run attached script Actual results: Script will never finish and consumes ~98% of cpu. Expected results: Script finishes.
Thanks for the reports! I'm working on a number of enhancements/fixes, and will address these shortly.
Interesting -- Im "not authorized to access bug#504799". Do you have another bug where you're seeing this?
(In reply to comment #2) > Interesting -- Im "not authorized to access bug#504799". Do you have another > bug where you're seeing this? Ups, this is private bz. I'll try to find another...
ok - here's the bz you can use (16MB attachment) - bz515677
Hmmm.... Looks like this is actually an XML::Twig issue, reportes as rt35672; there's a 3.33 "development" kicking around on xmltwig.com that resolves this. http://rt.cpan.org/Public/Bug/Display.html?id=35672 http://xmltwig.com/xmltwig/XML-Twig-3.33.tar.gz I've filed another tix asking that it be "officially" released to the CPAN; if that happens we can easily update perl-XML-Twig... If it doesn't, we should still look at updating perl-XML-Twig, as there are 2+ years of fixes/enhancements in there.
Right, I installed XML-Twig-3.33.tar.gz and the issue does not occur. Thanks Chris.
I believe these bug fixes are valuable and we should update XML::Twig. In Changes are mainly bugfixes so I hope it will be ok.
Ok, so no response from the author. I'm going to deploy 3.33 into production on a couple systems at work tomorrow and see what happens... Assuming everything goes OK (as I believe it will) I'll update our package here.
Um, I've already update to 0.33 in F-12 and rawhide. I believe with comparison to other Fedora updates, this is only small risk.
No problem. :) The author did just respond and state that he is going to cut a new GA release "ASAP", so hopefully we can switch over when that comes out. (I have a feeling it's also going to be called 3.33)
I'd like to close it as nextrelease (F-12).