[42423] in North American Network Operators' Group
Re: Re[2]: telehouse - 25 broadway
daemon@ATHENA.MIT.EDU (Brian Whalen)
Sun Sep 16 00:04:31 2001
Date: Sat, 15 Sep 2001 21:01:51 -0700 (PDT)
From: Brian Whalen <bri@sonicboom.org>
To: Sean Donelan <sean@donelan.com>
Cc: <nanog@merit.edu>
In-Reply-To: <20010916035610.18680.cpmta@c011.snv.cp.net>
Message-ID: <20010915210124.S1414-100000@cx175057-a.ocnsd1.sdca.home.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: owner-nanog-outgoing@merit.edu
Actually I think Exodus will run em for a week or two as a test..
Brian "Sonic" Whalen
Success = Preparation + Opportunity
On 15 Sep 2001, Sean Donelan wrote:
>
> I'm sure a full analysis will be performed after the recovery
> efforts are completed. I'm also certain the operators of
> both 25 Broadway and 32 Old Slip are working very dilgently
> to get them running.
>
> But I would like to point out, no one regularly runs their
> generators for 48+ hours as part of a normal test. In addition,
> most standby generators are fitted only for "limited" duration
> runs. You should expect problems during any extended run of a
> generator plant. I'm a bit surprised that 25 Broadway and
> 32 Old Slip are the only ones we've heard about.
>
> Until I know a bit more about what happened, I can't say
> whether any alternative design could have performed better.
>
> On Sat, 15 September 2001, Joe McGuckin wrote:
> > Was this unit tested regularly? With a load bank?
> >
> > If there was a weekly test run, why wasn't this problem caught?
> >
> > It seems like there's a lesson to be learned here.
> >
> > My guess is that many sites' idea of a periodic test is to fire up the
> > generator (without a load) for 5 or ten minutes and assume everything's ok.
> >
> > How many folks actually perform a load transfer to the generator during
> > testing to check out the transfer switch ?
>
>
