[FQuest Alert] SAMSON Server

[sheila]

The SAMSON server has experienced a high loading condition. The techs tried to bring the server back under control, but ended up rebooting the server. The SAMSON server is now back up and responding.

Event Start: 9:17 PM
Event End: 9:32 PM
Total downtime: 15 minutes

We do apologize for any inconvenience caused by this issue.

[LokiSon]

Did you just have another Samson event? I couldn't access my account for 15-20 minutes until about 10 minutes ago.

[phppete]

Yes they did but no notice was posted (strange)
Another load spike:
12:33pm up 15:04, 3 users, load average: 39.52, 92.98, 48.96

[Terra]

Yes, SAMSON suffered another event the forced us to gracefully reboot the server...

The high loading is a red herring, as the true problem is with a semaphore not being released in the mmap() subsystem which in turn causes processes to pile up in the 'R' state... What is causing this race condition and/or why it is not being released is not known yet, however tonight's maintenance should help to reveal if this is a hardware or software issue...

http://www.aota.net/forums/showthread.php?t=21173

--
Terra
sysAdmin
FutureQuest

[phppete]

Thanks for the explanation :)

[LokiSon]

Um... having another one right now?

[Bob]

LokiSon,

http://www.aota.net/forums/showthread.php?postid=146822#post146822

-Bob

[LokiSon]

tx Bob. That'll teach me to bookmark Alerts without looking further. :-/