[1943] in Release_7.7_team
hung Indy data
daemon@ATHENA.MIT.EDU (Robert A Basch)
Wed Sep 29 18:37:18 1999
Message-Id: <199909292237.SAA38489@aupair.mit.edu>
To: release-team@MIT.EDU
Date: Wed, 29 Sep 1999 18:37:00 -0400
From: Robert A Basch <rbasch@MIT.EDU>
I spent some time looking at asteroids, an Indy in the test cluster
that was hung in reactivate. By booting off the install server, and
stat'ing various files, I found:
1) It apparently had not completed reactivate; the first two files
in configfiles (/.cshrc, /.login) were trashed, but all others were
untouched. The trashed files were written at the same second as the
hang occurred.
2) passwd/shadow/group were OK, as were CellServDB and SuidCells, so
the machine rebooted fine. The new versions of these files were
written 9-10 seconds before the hang.
3) config_afs (run in the background by reactivate) apparently
completed about 4 seconds before the hang, based on the access time
of /bin/athena/fs.
4) The most recently written file in the AFS cache directory was
CacheItems, written 48 seconds prior to the hang.
asteroids is an R4600 Indy, so that seems to rule out an R5000-specific
problem.
Bob
