[tech] disk replacement post-mortem
James Andrewartha
trs80 at ucc.gu.uwa.edu.au
Thu May 13 01:18:42 WST 2004
Well, that took longer than expected. The 9gig disk from EE (thanks Alex)
went in nicely, although this was masked by the fact that the tape drive
was not happy on either SCSI bus, a fact I did not determine until just
before 11pm. I also pulled out the 4gig drive that wasn't doing anything,
so now no drives are next to each other in the prioris case (the drive
that died most recently was between two other drives, and full-height to
boot). Actually, I think that drive might still be ok, but upon
disassembling the SBB we discovered that it's SCA internally, which
combined with the facts that home02 (the first drive to die) was also in
an SCA SBB, and SCA hard drives go for a significant discount to 68pin
drives of the same capacity. Anyway, I also compiled (well, linked) a new
kernel for morwong so we don't have to boot genvmunix all the time to
detect the xcr raid controller. Currently the mirrors are being resynced
(when the tape drive was on the xcr none of the drives were detected by
Tru64, hence they got put in the stale state), once that's done I'll
attach the new 9gig drive to homemirror02 and then we should be optimal
again. The tape drive should be tested on another box first, there is a
spare narrow scsi card floating about that we can use to do this.
Today's moral: never boast about how you're going to use hardware solely
to perform backups, it makes it feel unwanted and go on strike.
--
# TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \
# UCC Wheel Member http://trs80.ucc.asn.au/ #| what squirrels do best |
[ "There's nobody getting rich writing ]| -- Collect and hide your |
[ software that I know of" -- Bill Gates, 1980 ]\ nuts." -- Acid Reflux #231 /
More information about the tech
mailing list