[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]
Re: DS20L
- From: "Ingo T. Storm" <tux-axp computerbild de>
- To: <axp-list redhat com>
- Subject: Re: DS20L
- Date: Fri, 31 May 2002 07:26:55 +0200
to add some nightmare here:
In our CPU room we have roughly 150 Suns (from older Ultra1/2/5/10 through
E220, E250, E450, E3500, E4500, all types of Netras through (now) a few
V880R), a handful of HP, ~100 IA, ~10 Athlons - and of course my 5
Ruffians). Half of them with small to large (2 racks) disk arrays and some
small to large tape libraries. Add to that some racks with telco stuff.
Then last year thanks to a UPS test on a Sunday morning, 7pm, the AC went
down for some 6 hours. It took us 3 hours to get all the admins on board and
shut the stuff down... By then in different corners of the room we had
already reached temperatures from at least 30C to over 45C in my corner of
the room.
In the end, all that blew up was one Athlon and dozens of disk in the
smaller disk arrays in the hottest corner. ALL of except a handful of disks
came up fine after it had cooled down. Ok, density is reasonbly low: less
then 10% of the machines are rack mounted. But to be honest, we would have
expected at least half of our stuff would have died for good in this
scenario.
And I don't think that anyone would sweat more in a "sauna" (what's the
english word for that?) than us admins did that day...
Have you padded YOUR servers today? :o)
Ingo
- References:
- RE: DS20L
- From: "Christopher C. Chimelis" <chris@debian.org>
[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]
[]