[SunRescue] heat problems in an SS20

Mike Nicewonger rescue at sunhelp.org
Wed Nov 15 22:00:07 CST 2000


Why not just replace the caps with fresh ones?

Mike N
----- Original Message ----- 
From: "Paul Theodoropoulos" <paul at anastrophe.com>
To: <rescue at sunhelp.org>
Sent: Wednesday, November 15, 2000 10:49 PM
Subject: Re: [SunRescue] heat problems in an SS20


> 
> Something else is probably wrong. I have an SS20 with quad 100Mhz 
> hyper's, and dual 4G 7200RPM drives - all this machine is doing at 
> the moment is running four setiathome processes. It does a great 
> job of heating my office, and has had no crashes, no lockups. The 
> only extra cooling in the box is one very tiny (1 inch) fan blowing 
> out the sbus slot immediately adjacent to the processors.
> 
> i would imagine that this may be a noise/capacitance problem across 
> your Mbus. SM41's are pretty old (I have an integrix SS20 clone 
> with dual SM41's in the same stack - it serves this domain), and 
> the capacitors on the card definitely drift out of spec as the 
> years go by. Since they drift non-deterministically, you never know 
> which SM41 may 'age out' sooner than another. A dual CPU config 
> definitely stresses the capacitance on the Mbus even further. If 
> you can get ahold of another SM41, you have a good chance of 
> finding a combination of the two out of three that will work 
> reliably...
> 
> 
> At 07:21 PM 11/15/00, you wrote:
> >I've known that these things run hot, but I've never had any 
> >problems
> >before now.  I've got an SS20, with (what I assume) are 4x32MB 
> >simms,
> >dual SM41s, and dual 4GB 5400RPM drives.  I've got the extra 
> >cooling fan
> >for the drives, but that doesn't seem to have been enough.
> >This machine runs Linux, and I wanted a copy of MySQL on here for 
> >some
> >testing.  I downloaded the package, and fired up a compile.  The 
> >first
> >time, it ended on a Signal 11, so I took out the drive, thinking 
> >that
> >was the most effective way to reduct the heat 
> >output.  Unfortunately,
> >this wasn't enough, so I ended up pulling one of the SM41s.  While 
> >this
> >allowed the compile to succeed, it's DARN slow.  Short of adding 
> >in a
> >few fans on the far side of the case, is there anything I can do 
> >to help
> >alleviate these heat problems.  (I haven't found a local source 
> >for box
> >fans, yet).  Thanks,
> >
> >     Greg
> >_______________________________________________
> >Rescue maillist  -  Rescue at sunhelp.org
> >http://www.sunhelp.org/mailman/listinfo/rescue
> 
> -----------------------------------
> Paul Theodoropoulos   paul at atgi.net
> Senior Unix Systems Administrator
> Advanced Telcom Group, Inc.
> Santa Rosa, California
> Work: http://www.atgi.net
> Play: http://www.anastrophe.com
> Downtime is Not an Option
> 
> _______________________________________________
> Rescue maillist  -  Rescue at sunhelp.org
> http://www.sunhelp.org/mailman/listinfo/rescue
> 




More information about the rescue mailing list