On Mon, 09 Nov 2015 15:59:47 -0500
zach villers <zachvatwork(a)gmail.com> wrote:
I'm waiting to see if the last person that worked on this ticket
is
still actively involved and trying to do some testing on my own.
However, the CentOS 7 VM I am using has a different version of squid.
I've left a diff of my squid.conf and kojipkgs' squid.conf and notes
on how I set up apache. Would anyone have the bandwidth to look at my
notes on the ticket
https://fedorahosted.org/fedora-infrastructure/ticket/4507 and see if
it looks like I am on the right track or give me some guidance on how
I could recreate the issue?
Well, thats the trick I guess. :)
I can see that the production instance is still doing this, so it
should be possible to setup a env thats as close as needed to that that
shows the same issue. ;(
I'm not sure why your local centos7 vm wouldn't show it unless the
difference in config is the part that causes it. :(
Thanks,
Zach
#aikidouke
PS. I remember a month or so ago there was an issue with squid
hitting 100% CPU on one of our boxes (sorry I can't remember which.)
I just saw this bug and thought it might be relevant. Not sure if the
issue was ever resolved or not. I didn't see any active tickets and
couldn't find any alerts that looked promising.