[1891] in Release_7.7_team
Re: 8.3 release, part deux
daemon@ATHENA.MIT.EDU (Ted McCabe)
Wed Jul 28 14:06:13 1999
Mime-Version: 1.0
Message-Id: <v0420540ab3c4f358d68e@[18.177.1.78]>
In-Reply-To: <199907281246.IAA21374@mary-kay-commandos.mit.edu>
Date: Wed, 28 Jul 1999 14:06:03 -0400
To: Garry Zacheiss <zacheiss@mit.edu>, release-team@mit.edu, ops@mit.edu
From: Ted McCabe <ted@MIT.EDU>
Content-Type: text/plain; charset="us-ascii" ; format="flowed"
I checked the volume access counts for these first two nights of the release.
The sun4 srvd volumes with 5 RO sites all had normal levels of
accesses, the release didn't seem to spike the daily usage
appreciably. The average I've seen historically is in the 3-4
million/day/RO range.
The sun4 os RO's, and the sgi srvd RO's had reasonable access counts
for a release happening, each with about ~1 million/day.
The sgi_65 os volumes had a suprising spike for the insight subvolume
(2 RO's). 8.3 million total access counts since this morning, but no
apparent change since about noon (which is simply when I started
checking). For comparison, the inst subvolume, which is the one I
expected to get most of the load, has had only 1.3 million accesses
so far today.
Was this usage of the insight volume expected?
It's hard to accurately correlate access counts to data xferred, but
I would imagine that this was a significant part of the network load
observed overnight.
--Ted