[3292] in Release_Engineering
Re: It's a good thing we're not making any new releases for the vax.
daemon@ATHENA.MIT.EDU (jweiss@MIT.EDU)
Fri Jul 29 17:52:56 1994
From: jweiss@MIT.EDU
Date: Fri, 29 Jul 1994 17:52:49 -0400
To: Bill Cattey <wdc@MIT.EDU>
Cc: rel-eng@MIT.EDU, dkk@MIT.EDU
In-Reply-To: [3290]
Well I explained this to a number of people in person, but here's what
I believe happened:
1) Sometime a few days ago the vax systempack for 7.6 vaxen in beta
(friendly) test was removed from the rel-eng cell. (I believe this
was related to loosing a disk on eros, or making space to restore the
data that was on the dead disk, but I'm not positive.)
2) This was noticed by dkk, when his vax lost it's packs.
3) I don't remember exactly how I got involved in the discussion, but
we decided that the reasonable thing was to point the beta-vax cluster
at the same filsys as the stafftest-vs cluster. (This happens to
point to the packs in the athena cell).
3a) we removed all of the DECstations from the beta-vax cluster (no,
don't ask, let's just say an @sys link was involved)
4) That night the Hesiod (R) propogation occured
5) much to some peoples surprise many of the vaxen in the beta-vax
cluster autoupdated from 7.6D to 7.6E. Well it was much to the
surprise of the people who noticed it. (Hi Bill)
Appearantly the beta-vax system packs in the rel-eng cell were at a
*lower* version number than the public packs in the athena cell. As
far as I know, no one is sure how this happened.
According to Moira (R) the following hosts are in the beta-vax cluster:
a-kind-of-magic, andrew, euphrosyne, get-a-life, gre, lakota,
nicky-jr, toto
Of these, euphrosyne, and get-a-life don't respond to ping, and
a-kind-of-magic is not autoupdate=true. According to SNMP the other
machines all autoupdated to 7.6E. As far as I know no one has tried
to contact the owners of these machines.
Jonathon
