Author: Wiki.Melancholie
Description:
Wikistats (located at http://dammit.lt/wikistats/) stopped with dump progress at 20080815-020000.
[[n:Template:Popular_articles]] (included on [[n:Main_Page]] is waiting for new some logs ;-)
Version: unspecified
Severity: normal
URL: http://dammit.lt/wikistats/
Wiki.Melancholie wrote:
Would https://wikitech.leuksman.com/ be a better place for reporting such things?
Wiki.Melancholie wrote:
Incomplete/corrupted logs between 20081021-090000 to 20081021-140001 (en, en.n etc. missing)!
Stopped working then with 20081021-150001.tmp
According to https://wikitech.leuksman.com/view/Server_admin_log the Squids should have been working well...?
Wiki.Melancholie wrote:
For completeness:
Disk space of the pagecounts collecting server seems to had been filled up somehow.
Data loss on October 21 between ~13:00 UTC and ~23:00 UTC, rest should resync...
Wiki.Melancholie wrote:
Working again, but it turned out that actually all the data between 20081021-090000 and 20081022-120000 got corrupted (because master daemon failed) or is even missing.
Wiki.Melancholie wrote:
Stopped working again with pagecounts-20090104-050000.gz; failing for 1½ days now.
It's weird, but any Squid/update problems do always occur when a major event is going on in the world.
Wiki.Melancholie wrote:
Stopped working again after pagecounts-20090201-060000.gz; 20 hours.
my box at datacenter started rebooting few times a day ( :((( ) - fixed some of scripts to recover more gracefully.
[mass-moving wikistats reports from Wikimedia→Statistics to Analytics→Wikistats to have stats issues under one Bugzilla product (see bug 42088) - sorry for the bugspam!]