One day later...
Well, we have two identical squid servers (load balancing with DNS).
The patched squid doesn't present this, anymore.
On the other hand, the other one, unpatched, hunged some minutes ago, and after launching, it
presented again this message (hanged, killed and restarted with dirty cache).
I'll patch this server and see if, hopefully, with a little bit of luck, they won't hang anymore,
and the patch solved both problems.
Amos Jeffries wrote:
On 17/01/2012 3:00 a.m., Amos Jeffries wrote:
On 17/01/2012 2:43 a.m., Jose-Marcio Martins da Cruz wrote:
Amos Jeffries wrote:
On 11/01/2012 11:44 p.m., Jose-Marcio Martins da Cruz wrote:
I had this too. But with a 3.1.18 in a production server (under
solaris).
Sure that was 3.1.18? that could narrow the diagnosis down a lot if its
true.
Yes. It's on a 3.1.18 on a Solaris box.
Tell me if I can make something to help.
3.1 narrows it down to probably just these two patches:
www.squid-cache.org/Versions/v3/3.1/changesets/squid-3.1-10394.patch
www.squid-cache.org/Versions/v3/3.1/changesets/squid-3.1-10403.patch
The top one being more likely. If you are able to patch and test with
those separately, and see if it disappears (or not) it would help.
Amos
I mean reverse the patch (patch -R -p0 < squid-3.1-10xx.patch) and test
with them absent of course.
Amos
