[11924] in Hotline Meeting
eros: rel-eng AFS server: SCSI controller
daemon@ATHENA.MIT.EDU (Richard Basch)
Tue Nov 24 00:55:10 1992
Date: Tue, 24 Nov 1992 00:55:04 -0500
To: op@Athena.MIT.EDU, hotline@Athena.MIT.EDU
Cc: rel-eng@Athena.MIT.EDU
From: "Richard Basch" <basch@MIT.EDU>
We have experienced three occassions of moderate to severe filesystem
corruption on eros in the past 1.5 months. Since it is occurring with
different drives (2GB R^2 drives), I suspect there may be a problem with
the SCSI controller. I would like to request that we have the SCSI
controller replaced on this machine, as a precautionary measure.
At this point, the loss within our source trees has been over 300 files
in the past 1.5 months. Up until now, no recent work has been lost,
except now, the loss has now included very recent work; it destroyed
several files relating to the Sun port.
In addition, I have not yet bothered trying to recover the losses within
the 7.4 build volumes that have occurred...
I don't believe 3 times is a coincidence, and I would rather have a
precautionary hardware replacement of the most likely component that
could be failing. Self-tests did not reveal anything, but since we are
talking about our source tree, let's not play any games.
Please log a service call ASAP, and let me know what is going on.
Thanks,
-Richard
