[11482] in Hotline Meeting
Addendum for today's daily report
daemon@ATHENA.MIT.EDU (salemme@Athena.MIT.EDU)
Wed Oct 28 09:45:46 1992
From: salemme@Athena.MIT.EDU
Date: Wed, 28 Oct 92 09:45:15 -0500
To: op@Athena.MIT.EDU
Cc: athena-outage@Athena.MIT.EDU
Backup HIGH Due to evidently buggy behavior of 'cron' on persephone
(RS6000) after the time change, two cloning/expunging
operations ran simultaneously the nights of Oct 25/26
and Oct 26/27; this resulted in more than 2500 reported
volumes needing salvaging. [Normally, there are only
a few, if any. We found the numbers, 801 one day and
1787 the next, to be alarmingly high.] In the interests
of time and safety, we decided to run salvage operations
over all the servers where these volumes reside. This
resulted in outages ranging from a few minutes to about
half an hour for each server salvaged. [After we tracked
the problem down, rebooting persephone seemed to clear
up the problem. Last night's cloning/expunging worked
fine, with no reported problems.]
