On 6/20/07, Jens Frank jf@mormo.org wrote:
On Wed, Jun 20, 2007 at 10:15:22PM -0600, Daniel Cannon wrote:
There seem to be more Wikimedia server problems. No sysadmins could be
found
on IRC, so I'm posting here to see if someone gets the message.
We're getting the standard Wikimedia Error on all wikis with a message
to
the effect of:
Request: GET http://de.wikipedia.org/, from 66.230.200.146 via sq18.wikimedia.org (squid/2.6.STABLE12) to () Error: ERR_CANNOT_FORWARD, errno (11) Resource temporarily unavailable
at
Thu, 21 Jun 2007 04:13:11 GMT
It's been reported on multiple squids, and at present I am completely
unable
to load any page of any wiki without getting it. If a sysadmin could
kindly
assist -- thank you.
Works for me. Nagios looks OK. Seems to be fixed.
Yes, it seems to have just sorted itself a few minutes after I made that post .. just my luck =D. JeLuF's taking a look over the logs on the squids, etc., and can't seem to find anything that caused this. I'm not particularly concerned, but did notice this relevant log entry:
21:07 brion: took live a change to SquidUpdate.php which should skip HTTP purges instead of doing both HTTP and HTCP (old code didn't exit the HTTP function when subcalling to the HTCP function)
May or may not be related (probably not, but it's the only thing I could find pertaining to the squids), but just tossing that out there. In any case the crash effected every user of every wiki, causing a flood of traffic to IRC. Seeing as it's quite a slow hour right now, it would definitely be good to have this looked into and hopefully prevented before the morning rush :).
Anyway, I'll try to report here if this happens again.