[Info-vax] Alphaserver service/repair recommendations

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax-bounces at info-vax.com] On Behalf Of
> rcyoung via Info-vax
> Sent: 13-Feb-16 8:01 PM
> To: info-vax at info-vax.com
> Cc: rcyoung <rcyoung at aliconsultants.com>
> Subject: Re: [New Info-vax] Alphaserver service/repair
> recommendations
> 
> On Saturday, February 13, 2016 at 6:14:47 PM UTC-5, rcyoung wrote:
> > On Saturday, February 13, 2016 at 3:41:43 PM UTC-5, Robert A. Brooks
> wrote:
> > > On 2/13/2016 2:36 PM, rcyoung wrote:
> > > > Any recommendations on where to obtain service on a 2100 4/233
> OpenVMS
> > > > system? South Carolina or Georgia would be best, but I'll look at
> more
> > > > distant providers too.
> > >
> > > I don't have an answer to your question, but you might improve your
> chances
> > > of getting help if you can describe the problem with the system.
> > >
> > > It's been said (although not in a while) that "terse questions beget
> terse answers".
> > >
> > > --
> > >                   -- Rob
> >
> > Sorry for any confusion... here are the details...
> >
> > I have an Alphaserver 2100 4/233 running OpenVMS giving the
> following error on power up.
> >
> > Interrupt through vector 660 on cpu 0
> > starting console on cpu 0
> > breakpoint at PC 130c98 desired, XDELTA not loaded
> >
> > It was just rolling the error msg again and again....but the last time I
> powered up I did not even get the error or a prompt of any kind. The LED
> display showed it doing internal tests however.
> >
> > I was able to move drives over to a second machine, but also noticed
> that one drive is not even being recognized on the second system ( but a
> replacement drive is), so something is bonkers with that disk too.
> >
> > When it rains it pours...
> 
> Update: error message noted before is back....infinitely
> repeating.......with no drives in system.

HW troubleshooting tips - assume maint guide available: 
(May have to defeat the cover switch to check these)

One approach to solve problem is to determine what is NOT the problem

- if volt meter available, measure DC outputs to confirm pwr supply ok.
Ideally measure DC output with oscilloscope for excessive ripple, but 
these scopes are not always available these days except in scientific
shops.

- dead fans are always interesting as CPU overheating can cause any type
of faults. Check fans all ok. 

- isolate to lowest common denominator by removing as many components 
as possible i.e. PCI peripherals, memory (check interleaving, leave min 
memory), leave CPU 0 (remove other Cpu), then see if pwr up error 
still there.

Should be able to get to the >>> prompt with min cpu, memory and no 
disks.


Regards,

Kerry Main
Kerry dot main at starkgaming dot com