[2899] in SIPB-AFS-requests
Re: backup dumpinfo problem (and upgrading rb)
daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Wed Nov 12 02:43:49 1997
From: Jonathon Weiss <jweiss@MIT.EDU>
To: Greg Hudson <ghudson@MIT.EDU>
Cc: Jonathon Weiss <jweiss@MIT.EDU>, sipb-afsreq@MIT.EDU
In-Reply-To: Your message of "Tue, 11 Nov 1997 02:43:31 EST."
<199711110743.CAA15488@the-light-fantastic.MIT.EDU>
Date: Wed, 12 Nov 1997 07:43:30 GMT
OK, I replaced buserver on ra and rc and all of the binaries in
/afs/sipb/project/newdump/@sys. I'm not sure what stat things were in
before, but now the backup binary we have can perform dumpinfo queries
against ra and rc, but will fail with incompatible interface for rb.
I think it is time for us to move forward fairly aggressively with the
upgrade for rb. Personally, I'd like to see the actual upgrade happen
over winter break or IAP. This will have the additional advantage of
increasing the free space in teh cell.
I believe the next step is to order disks for the machine. I believe
we were planning on getting 4 4G disks (possibly, we could reuse the 2
disks that are currently on rb, by moving them to rc when we're done
(the 4G disk could replace teh mycrapolis disk that we ar no longer
using, at least) but we don't need to figure out that detail now.)
Is there any reason we shouldn't just order 4 4G seagate baracudda
disks in external enclosures with centronics connectors, short scsi
cables and FPT terminators (otherwise known as the configureation ASO
is using in the dev cell) Should we order 48hour advanced service? If
not, what is our plan for dealing with dead disks.
I'd prefer that someone else dealt with the process of getting the
quotes and ordering disks (it isn't even very hard, but other people
should know how.) I can tell you what you need to do, and I suspect
that ghudson can too. Any volunteers?
Any problems? I'd really like to see this get moving.
Jonathon
