Re: Server keeps dying silently

Now that's a good idea, having narrowed down the obvious hardware suspects.
Mobo's are so hard to faultfind, but a leaking capacitor is an avenue I
haven't tried. In with the torch then....

"ratman and bobbin" wrote:


"thedrumdoctor" <thedrumdoctor@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:A5146C1F-AECD-472F-A6BA-B85A8F9E9641@xxxxxxxxxxxxxxxx
I've run out of ideas on this one and it's a good job it's in a home
environment rather than in production!

I purchased a used server (ok alarm bells ring but don't diss the purchase
just yet). It had been in a production environment running 2000 SBS. The
internal specs are:

Gigabyte GA-8IK1100(Rev 1.x) Mobo
P4 2.4 CPU
1GB DDR RAM
2 x IDE Maxtor 76GB HDD (Mirrored boot volume on same controller)
Tekram DC-315 SCSI controller hosting HP C5683A tape drive
Nvidia GForce MX440 display (running standard Microsoft drivers for
troubleshooting purposes)
On-board SATA RAID with 2 Seagate Barracuda 80GB drives mirrored (these
have
been removed and the onboard RAID and SATA support has been disabled for
troubleshooting)
BIOS is up to date with the latest release for the board

Win 2003 standard server SP1 is installed, running as a PDC and running
WSUS. I've put off upgrading to SP2 for the moment.

*PROBLEM*
The server will randomly die, that is to say, lock-up, giving no BSOD. All
the lights stay on, the PSU carries on and the drives spin. There is no
display and I simply have to power off the server and power on again.

I've ran memory tests for hours and everything passes. I've ran CPU stress
tests for over 8 hours and everything passes. I've ran HDD checks and they
pass. I've used motherboard monitor to check the temperature but the CPU
doesn't go over 75 and the internal sensors all report normal cooling. The
BIOS has been set to fail-safe defaults and I've set the CPU temperature
to
alarm if it reaches 80. But nothing really heavy happens on this server to
actually work the CPU. Even when SQL is running during the WSUS update it
hardly hits 50% (and hyperthreading has been enabled). I've seen dodgy
PSU's
take down boxes before so I changed it for a new one but it still dies on
me!

I haven't loaded any system drivers that have not been digitally signed by
Microsoft and have even run Driver Verifier to try and find a culprit, but
nothing shows up.

The only thing that seems to have any repition with the failures is that
they nearly all happen during a scheduled anti-virus scan (Symantec AV
Corporate edition 8.0). Now there's no reported compatability issues with
this product according to Symantec on 32 bit versions of Server 2003 so I
can't see this being a problem, unless it's trying to access locked files
that Server 2003 doesn't want it to. However, I have never had any of
these
problems with 2000 Server. Having tested probably every hardware
possiblity
I'm at a loss as to what could be causing the lock-up without giving me a
BSOD. I'd actually LOVE a BSOD as it least it would give me a clue that
it's
either hardware or driver related! Oh yes, one final thing, the event logs
are clean, really clean. They throw up information warnings about Symantec
anti-virus not being able to access certain files in the Documents and
Settings folder and its subfolders, but that's it. No hardware errors or
warnings that a disc or device is about to fail. Has anyone experienced
this
sort of thing before? I can't see how an anti-virus scan could bring down
a
server as it's never happened to any of the 2000 servers I've had in the
past, or is 2003 server just too fussy?

Could it possibly be bad capacitors? I had a similar problem about a year
ago on a P3 server running W2K and Exchange 5.5. It would run for longer
than 8-10 minutes before completely freezing. It ran for longer in safe mode
but would still eventually freeze. Maybe and eyeball inspection to check for
bulging/leaking capacitors.

hth
Geoff



.

Relevant Pages

  • Re: Performance of processing
    ... level of aggregations in the cubes ... I have an Intel server where all individual benchmarks (disk, memory, cpu) ... the cubes) will take on a particular hardware. ...
    (microsoft.public.sqlserver.olap)
  • Re: Performance of processing
    ... level of aggregations in the cubes ... I have an Intel server where all individual benchmarks (disk, memory, cpu) ... the cubes) will take on a particular hardware. ...
    (microsoft.public.sqlserver.olap)
  • Re: 2008 SBS no longer boots
    ... driver issue initially, the fact that it seems to happen randomly now ... weird with a hardware problem that consistently doesn't get through boot ... other customer, even a 2 week old one, the problem immediately returned. ... There was a time when a server was fired up with a specialized OS and had a battery of tests run against each component to purposefully stress it. ...
    (microsoft.public.windows.server.sbs)
  • Re: Dealing with SSL processing via hardware
    ... If the "hardware" is basically just a single CPU computer with Linux and OpenSSL installed on ROM wrapped in a sealed box, then I would suppose you could emulate the exact same performance characteristics by adding another CPU to your main server computer and only using that CPU for running the same SSL and socket software you would have inside the sealed box. ...
    (borland.public.delphi.non-technical)
  • RE: Recommendation for 1000BASE-SX card?
    ... supported and drivers that aren't. ... on current versions of hardware. ... Cheapo crappy desktop gear, and expensive high quality ... into the high-end server market in a big way. ...
    (freebsd-questions)
Loading