[3234] in Release_Engineering
New bugs on 7.7 solaris -- umount/attach
daemon@ATHENA.MIT.EDU (Bill Cattey)
Fri Jun 17 11:57:15 1994
Date: Fri, 17 Jun 1994 11:57:06 -0400 (EDT)
From: Bill Cattey <wdc@MIT.EDU>
To: miki@MIT.EDU, vrt@MIT.EDU, rel-eng@MIT.EDU
Last night, when logging out, or perhaps when the machine was
deactivating, I noticed several messages on the console of the form:
/sbin/umountall: cut: not found
One theory is that this may be because the program cut is needed on the
local disk for umountall to do its work, and that it didn't get tracked
in, or put on /srvd.
This morning, logging in, I got a whole raft of error messages about the
various NFS filesystems I attach for image browsing in the Tulip Library
project:
11:34 fsid: server bulb.MIT.EDU not responding (for filesystem
bulb:/images/EA000043)
11:34 fsid: Ignoring host bulb.MIT.EDU for filesystem
bulb:/images/EA000034 due to previous host errors
11:34 fsid: Ignoring host bulb.MIT.EDU for filesystem
bulb:/images/EA000041 due to previous host errors
11:34 umount: warning: /images/EA000041 not in mnttab
11:34 umount: /images/EA000041 not mounted
When I do /etc/mount, these filesystems are shown not to be mounted.
When I do attach, they are shown to be there.
When I do ls, I find they are REALLY NOT THERE.
When I attempt attach, I am told they are already attached.
Inasmuch as we just gave the programs that use these filesystems to
users this past wednesday, and inasmuch as we want them to work across a
reactivate, I believe that this mount problem needs to be fixed before
we deploy 7.7 to real users.
-wdc
