Proliant DL380 G3 random reboots

Running Windows Server 2003 SP2 we cannot figure our why a Compaq DL380 G3
server is randomly rebooting and I would like some advice on diagnostics and
troubleshooting before I call Compaq.

This was our exchange 2003 server and we are now moving all services off
this unit so we can do extended testing.

The Windows event log gives no clues to the reboots and it's happened 20
times in the last week so we've got lots of data yet nothing which says
points to anything specific. NO errors occur before the reboot in any of the
windows event logs or in the HP event viewer logs.

The only change is we bought some non-HP RAM about three weeks before the
trouble started, but then we switched back to the HP RAM and the trouble
continues.

Needless to say we cannot recreate the error. We have Compaq 24x7x4 support
so we could have the thing replaced if we can only recreate the problem.

All we have discovered so far is the CD-ROM drive is not working, as it will
not eject via the operating system or the eject button on the drive itself
(we had to open it with a paper clip using the release slot). And it will
not read a CD-ROM. Although the light does flash when you press the eject
button. At least this is repeatable, but I fear this is not the cause.

We have six 72GB SCSI drives, four are in a RAID 5+1 and two are in a RAID
1.

If you have any suggestion please let me know.

Thanks!
-Bob

"just bob" [email protected] wrote in message
...
Running Windows Server 2003 SP2 we cannot figure our why a Compaq DL380 G3
server is randomly rebooting and I would like some advice on diagnostics
and

You didn't describe your setup so there are / can be lots of variables.

Also consider the simple things,

1-Possible intermittent AC line power to the server.

2-A UPS with bad batteries and the AC-line power flickers and you could get
a reboot.

Just a thought ..

"Phil" wrote in message
. ..

"just bob" [email protected] wrote in message
...
Running Windows Server 2003 SP2 we cannot figure our why a Compaq DL380
G3 server is randomly rebooting and I would like some advice on
diagnostics and

You didn't describe your setup so there are / can be lots of variables.

Also consider the simple things,

1-Possible intermittent AC line power to the server.

2-A UPS with bad batteries and the AC-line power flickers and you could
get a reboot.

Just a thought ..


Not sure what more I could have told you about the setup. I gave you the
model, OS, drive layout and RAM. It has two CPU's. No third party software
or hardware, save the OS and Exchange 2003 and anti-virus. When the server
reboots there is nothing relevant in the event logs, before or after the
reboot.

Our UPS system if fully monitored and we get alerts for everything. owing
up. And we have redundant power supplies on in the servers and the HP alerts
are working for those too. If we pull one out, or disconnect the power
cable(s), we immediately get alerts.

Thanks for the suggestion.

"just bob" [email protected] wrote in message
m...

"Phil" wrote in message
. ..

"just bob" [email protected] wrote in message
...
Running Windows Server 2003 SP2 we cannot figure our why a Compaq DL380
G3 server is randomly rebooting and I would like some advice on
diagnostics and

You didn't describe your setup so there are / can be lots of variables.

Also consider the simple things,

1-Possible intermittent AC line power to the server.

2-A UPS with bad batteries and the AC-line power flickers and you could
get a reboot.

Just a thought ..


Not sure what more I could have told you about the setup. I gave you the
model, OS, drive layout and RAM. It has two CPU's. No third party software
or hardware, save the OS and Exchange 2003 and anti-virus. When the server
reboots there is nothing relevant in the event logs, before or after the
reboot.

Our UPS system if fully monitored and we get alerts for everything. owing
up. And we have redundant power supplies on in the servers and the HP
alerts are working for those too. If we pull one out, or disconnect the
power cable(s), we immediately get alerts.

Thanks for the suggestion.


Have you tried reseating all the internal plugs and boards inside the
server. Same with the ram, pull it out and reinsert it a few times. Since
the problems started with changing ram, something could have happened in
that process. These type problems can be a "bitch" to locate but the cure is
usually very simple.

"Phil" wrote in message
.. .

"just bob" [email protected] wrote in message
m...

"Phil" wrote in message
. ..

"just bob" [email protected] wrote in message
...
Running Windows Server 2003 SP2 we cannot figure our why a Compaq DL380
G3 server is randomly rebooting and I would like some advice on
diagnostics and

You didn't describe your setup so there are / can be lots of variables.

Also consider the simple things,

1-Possible intermittent AC line power to the server.

2-A UPS with bad batteries and the AC-line power flickers and you could
get a reboot.

Just a thought ..


Not sure what more I could have told you about the setup. I gave you the
model, OS, drive layout and RAM. It has two CPU's. No third party
software or hardware, save the OS and Exchange 2003 and anti-virus. When
the server reboots there is nothing relevant in the event logs, before or
after the reboot.

Our UPS system if fully monitored and we get alerts for everything. owing
up. And we have redundant power supplies on in the servers and the HP
alerts are working for those too. If we pull one out, or disconnect the
power cable(s), we immediately get alerts.

Thanks for the suggestion.


Have you tried reseating all the internal plugs and boards inside the
server. Same with the ram, pull it out and reinsert it a few times. Since
the problems started with changing ram, something could have happened in
that process. These type problems can be a "bitch" to locate but the cure
is usually very simple.


Another thought,
You said "Our UPS system if fully monitored"
Is it monitored by the server in question, or by another means. If power
fails from the UPS (or connection) and monitored by the server in question,
it would not be able to record the incident. Again, not knowing the setup
can only "guess"

If you don't have your servers plugged directly into the ups, a common power
strip could be at fault. I am assuming you are using only one ups.

We use two power busses feeding each cabinet with each on separate ups's. In
other words each power supply in the servers (and all equipment) are fed two
from different sources of independent (backed up) power. Also in my case we
use two active PDU's per cabinet that not only switch on/off/reboot, but
also monitor voltage, current, and temperature. I monitor and control all
power independent of the server racks.

Have you actually done a test of your ups (pulling power to it) to make sure
it's doing it's job and batteries holding up. Again, with an abrupt power
failure and being monitored by server in question, it will not record the
cause of failure.

Just thoughts ! Good luck.

"just bob" [email protected] wrote:

The Windows event log gives no clues

What's in the server's IDL (Insight Diagnostic Log)? - The one
accessible from ROM / your Smartstart CD / the system management
homepage?

Did you try running diagnostics from Smartstart?


cheers
Heimo

--
You never ask questions when God's on your side.