[2591] in SIPB-AFS-requests
Re: notes on software for ronald-ann
daemon@ATHENA.MIT.EDU (mhpower@MIT.EDU)
Mon Dec 23 22:07:26 1996
From: mhpower@MIT.EDU
To: jweiss@MIT.EDU
Cc: sipb-afsreq@MIT.EDU
In-Reply-To: "[2589] in SIPB-AFS-requests"
Date: Mon, 23 Dec 1996 22:05:58 EST
>I think the the change notes in sipb-afs transactions [2269], [2279],
>[2281], [2313], [2363], [2512], and [2552] would need to be followed.
>[2269] mentioned a number of files and directories under
>/afs/sipb/service/afs/sunos -- this directory tree could presumably be
>updated using the information from the other six transactions.
This has been done. I'm not sure if we want to just use the copy of
reynelda's kernel (in /afs/sipb/service/afs/sunos/vmunix) or whether
we should rebuild it locally on the new ronald-ann. Otherwise, the
information in [2269] should now be mostly usable as it is.
>Arguably the SIPB locker release script and the nightly cloning are
>services.
Ok, I've copied them to /afs/sipb/service/afs/sunos/var/local/sync for
now. I only noticed one part that obviously wouldn't work: synclocker
wouldn't have synctree in its path on SunOS. We appear to be running
this synctree binary on ronald-ann:
-rwxrwxr-x 1 root 170280 Sep 19 1991 /usr/local/bin/synctree
and I have no idea what source code it was built from, or if it
differs in any important way from the standard synctree binary.
Perhaps the author of synclocker would know.
For now, I added /usr/athena/etc to the path in the copy of synclocker
in /afs/sipb/service/afs/sunos/var/local/sync.
It might be a good idea to test the SunOS synclocker and release.sh at
a time when people are around to watch whether and how they break.
Matt
