[2404] in SIPB-AFS-requests
Greg Hudson: Defect noticed
daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Wed Jul 24 17:44:35 1996
From: Jonathon Weiss <jweiss@MIT.EDU>
To: op@MIT.EDU, sipb-afsreq@MIT.EDU
Date: Wed, 24 Jul 1996 16:34:24 EDT
FYI...
------- Forwarded Message
Received: from SOUTH-STATION-ANNEX.MIT.EDU by po10.MIT.EDU (5.61/4.7) id AA27177; Wed, 24 Jul 96 13:12:20 EDT
Received: from GLACIER.MIT.EDU by MIT.EDU with SMTP
id AA11463; Wed, 24 Jul 96 13:12:19 EDT
Received: (from ghudson@localhost) by glacier.MIT.EDU (8.6.12/8.6.12) id NAA07729; Wed, 24 Jul 1996 13:12:19 -0400
Date: Wed, 24 Jul 1996 13:12:19 -0400
Message-Id: <199607241712.NAA07729@glacier.MIT.EDU>
From: Greg Hudson <ghudson@MIT.EDU>
To: afsdev@MIT.EDU
Subject: Defect noticed
I was scanning through the 3.4.a.patches.2 fixed defects to find
something which might apply to the panics people have been seeing (I
didn't find any), and I noticed this one which might be relevant to
the afsuser locker:
3.4-7164-vos.remove.check.vldb.entry.before.deleting
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
A vos remove of a backup volume deleted an entry in the VLDB even if
given the wrong partition. VLDB entries are now checked before a
backup volume is deleted. If the volume is not listed in the VLDB but
is on disk, the volume is deleted from the disk instead of failing.
If the volume is in the VLDB but not on another server or
partition, the entry is removed from the VLDB. Additional checks are
now done when using vos remove to delete a volume.
I don't think it's worth redoing the tests to get this fix into the
afsuser locker (it's not a bug that destroys data or anything), but I
thought people should be aware that the bug exists.
------- End of Forwarded Message
