Nicolas Ross (Nicolas.Ross@cybercat.qc.ca)
Thu, 13 Jan 2000 16:29:38 -0500
Peter A. Castro said...
> Nicolas Ross said ...
> > Peter A. Castro said...
> > > Nicolas Ross said ...
> > > >
> > > > Hi there !
> > > >
> > > > Recently, we bought a 200 MHz CPU upgrade card from one who was
selling
> > > > them on this list.
> > > Did the person you bought them from test them first?
> > Nope, they were brand new from the box, factory sealed...
>
> Hmm...
>
> > > > We have a ANS 700 w/ 150 MHz CPU, 112 megs ram, 5 IBM 4 gigs disks
in
> > > > software raid, running YDL 1.1
> > > >
> > > > So, yesterday morning we replaced the 150 MHz CPU card with the 200
MHz one.
> > > > The server went OK for about 8 hours. And then started crashing like
hell.
> > > > Crashed 3 times in less than one hour. I finally removed the 200 MHz
cpu,
> > > > and but back the old one.
> > > >
> > > > Any similar case ? Any solutions ?
> > >
> > > We have three ANS's and I upgraded the last two with 200 MHz CPU cards
> > > last week. So far no problems at all. Can you describe the crashes?
> > > Are programs dumping? Are you getting kernel panics? Machine lockup?
> > > Anything suspicious in you syslogs? More info please!
> >
> > It varies...
> >
> > The first time, the console was frozen, nothing to do. Three finger
split
> > (ctrl - opt - start)
> > Second time, httpd crashed, and caused a kernel panic.
> > Third time, console froze again.
> > There were nothing in the log files saying abnormal things...
>
> Ok. If you pop the 150MHz card back in is it stable again? If so,
> then something on the bus can't handle the faster access time. My first
> suspect would be the memory. As an experiment, pull all but the first
> bank of memory out (ya, I know it's a pain to do). Also, if you have any
> cards plugged into the PCI bus, pull them too, as well as any harddisks
> which aren't needed to boot the machine. If that works, swap those
> modules with the next set of memory. The objective is to find which
> modules are marginal. If all memory passes, try re-inserting the drives,
> one by one, followed by the cards.
> Good luck.
This is ... hmmm ... a productions erver, so a down time of more than a
reboot time is not possible. Maybe at 4 o'clock sunday morning.
But anyway, the bus stays at 50 MHz when switching from 150 to 200, so this
theory doesn't stand...
> > > > Nicolas Ross, ing. stag.
>
> --
> Peter A. Castro (doctor@fruitbat.org) or (pcastro@us.oracle.com)
This archive was generated by hypermail 2.0b3 on Thu Jan 13 2000 - 21:32:43 GMT