[3931] in Release_7.7_team
Re: Status of Solaris 9.2 public release
daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Tue Jul 15 19:10:53 2003
Message-Id: <200307152310.h6FNAoZU009963@traction.mit.edu>
From: Jonathon Weiss <jweiss@MIT.EDU>
To: Garry Zacheiss <zacheiss@MIT.EDU>
cc: release-team@MIT.EDU, ops@MIT.EDU
In-reply-to: Your message of "Tue, 15 Jul 2003 18:45:37 EDT."
<200307152245.h6FMjbHW020592@brad-majors.mit.edu>
Date: Tue, 15 Jul 2003 19:10:50 -0400
> >> I asked kretch about this, and at his request sent mail to netreq
> >> describing my observations. He said he would look into it this morning.
>
> Kretch looked into it for me this morning, and this was his analysis:
>
> > It looks like this 10 Mb/s repeater may have been dealing
> > with as much traffic as it can handle. The MRTG graph
> > shows a large amount of traffic onto the network during
> > the night and about 10 Mb/s coming from the network between
> > 5am and 3pm. Can't know for sure if it was hitting this
> > repeater but it's certainliy likely. Take a look at the
> > graph:
> >
> > http://web.mit.edu/mrtg/www/nw12-rtr-2-backbone.11.html
>
> and the heavy usage "square wave" in the top graph at that URL does
> correspond to when machines would've been updating, so I'm guessing the
> update was beating the repeater senseless.
>
> Since a check with athinfo indicates some reasonable number of
> Suns in w20 didn't finish updating until around noon today, I'm going to
> start the Linux update process earlier tonight to compensate.
That repeater definately servers parts of the cluster (and I think
nothing else). Given the way the various updates work, it may not be
necessary to start tonight's update earlier, but it certianly won't
hurt.
Jonathon