[2991] in Release_Engineering
Re: power hit affected eros
daemon@ATHENA.MIT.EDU (Jeffrey I. Schiller)
Sat Mar 13 22:04:54 1993
Date: Sat, 13 Mar 93 22:04:48 -0500
From: Jeffrey I. Schiller <jis@mit.edu>
To: rel-eng@Athena.MIT.EDU, op@Athena.MIT.EDU, basch@mit.edu, jon@mit.edu,
In-Reply-To: kim@Athena.MIT.EDU's message of Sat, 13 Mar 93 20:33:02 EST <9303140133.AA03812@box.MIT.EDU>
From: kim@Athena.MIT.EDU
Date: Sat, 13 Mar 93 20:33:02 EST
There was a power hit this afternoon in E40. I think that
Tom found eros halted with fsck errors. He probably commented
out the bad partition and booted the machine (at least that's
what he did with our servers that were halted with fsck errors.)
I don't really know the "normal" state of eros, so I'm
hesistant to do anything with it...
I checked on eros's rz6 (the disk with the fsck problem). Fsck reports
many DUP blocks on many inodes. Looks like something bashed some critical
disk i/o's to the ilist region of the disk. All the affected inodes appear
to be in two groups of contiguous inodes. In any event, I left it alone.
So as of now eros is up, but all volumes on rz6 are unavailable.
-Jeff
