[1611] in SIPB-AFS-requests

home help back first fref pref prev next nref lref last post

Re: Cutover plans

daemon@ATHENA.MIT.EDU (Greg Hudson)
Tue Nov 15 22:56:06 1994

To: Matt Braun <mhbraun@MIT.EDU>
Cc: star-maintainers@MIT.EDU
In-Reply-To: Your message of "Tue, 15 Nov 1994 21:35:21 EST."
             <199411160235.VAA03915@medic.MIT.EDU> 
Date: Tue, 15 Nov 1994 22:55:33 EST
From: Greg Hudson <ghudson@MIT.EDU>


Okay, disk space:

We have a 2GB disk sitting on the third floor of E40 and a 1GB disk on
opus (old-beacon) doing nothing.

We also have 1GB of disk on bloom-picayune which is destined to be
freed up when RTFM is cut over to penguin-lust (penguin-lust already
has a 1GB external).

In order to do the news cutover, we may need to have two 1GB disks (we
haven't yet determined if we can just reattach the disks to a sun).
Therefore, if we do the AFS migration with the 2GB disk in E40 and the
1GB disk on opus, we may have to wait for the RTFM cutover before we
can cut over news.

Therefore, before deciding to use the 2+1 disks for the AFS migration,
I would like someone to find out whether we can move disks from an
Ultrix machine to a Sun rather than copying the data.  We should also
wait a few days to see whether purchasing actually orders the 4GB
disk, since that will make our lives a lot easier.

(Another possibility, if Purchasing is slow, is bringing up the third
AFS server with the 2GB disk, moving all of rosebud's RW data there,
moving all the RO data on rosebud to the internal disks, and freeing
up rosebud's external disk.  Then we can do a news cutover with
rosebud's and opus's disk, assuming we need to copy the news, put one
of old-bed's disks on new-rosebud, and copy the RO data there.  I
don't think this will turn out to be the best solution, though.)

I will not have time to work on anything until after next Tuesday
because of classwork.  Other people are encouraged to do things
without me.


home help back first fref pref prev next nref lref last post