HP Proliant DL580 G4 crashes. Blue screen

[LRP]

I have 2 HP Proliant DL580 G4, both with 1 cpu Xeon 3.0, 4Gb Ram, 3 hdd sas (raid 0+1, hot spare).
They have windows server 2003 x64 installed and RAC, from oracle.
Both machines have the same drivers and firmware installed.
they work fine for 3 or 4 days and then node 2 crashes with the error: "The computer has rebooted from a bugcheck. The bugcheck was: 0x00000109
(0xa3a03a389148f4f0, 0x0000000000000000, 0xd89a905d7b98fd53, 0x0000000000000101). A dump was saved in: C:\WINDOWS\MEMORY.DMP."
Once in while the error occours in node 1 (it happens twice)
Can anyone help me?

[chserj]

Check Hard Drive.

[LRP]

I already run the HP Insight Dignostics and no errors where found. everything seams to be working fine.
Those servers are connected to a Storage, HP EVA4000.

[Pieter 't Hart]

you mention EVA4000 so you're using SAN-technology.
i'm no SAN expert, but did have a basic training (HP SAN storage essentials).
a SAN (the network) is very sensitive about the whole (!)software configuration.
Engineers at our site (both HP an non-HP) said the same when implementing the SAN.
so check
- firmware of eva
- firmware of san switches
- firmware of hba's
- version of hba-driver
- version of multipath software
- firmware version/software versions of other devices attached to the SAN.
you may try to find an overall compatibility-matrix.
I don't think you'll get instant answer on this forum but you should contact HP-support for this.

[Daymon McCartney]

I'm seeing the same problem on an HP BL460c G1. Did you ever find a resolution to this issue? There are serveral configuration similarities between your set up & mine:

* Full STOP code = 0x00000109 (0xA3A03A389107F761, 0x0000000000000000, 0x73E4FE9CEE997019, 0x0000000000000101)

* Running Win2003 x64 (SP2)

* Direct-attached SAS drives

* SAN-attached (via Qlogic HBAs)

* Running database software (in my case, SQL)

Note that the last parameter of the STOP code matches, but is not a documented value at http://msdn2.microsoft.com/EN-US/library/aa469190.aspx.

Any pointers would be appreciated...

[LRP]

After a meeting with Oracle peolpe, they decided to install all Oracle software so we also decided to install all software.
Now whe install Win 2003 server R2 X64,SP2, psp 7.70, MPIO and drivers for a backup device.
Now the machine is working for about 3 weeks and no errors where found.

[Daymon McCartney]

Curious: Why PSP 7.70 instead of 7.80, which would've been the latest version available at the time you rebuilt?

[LRP]

I have read in a forum that 7.80 could cause some crashes with that machine, so i decided to put 7.80 insted of 7.80 that was the latest psp at that time.

[Daymon McCartney]

Thanks for the info. Hopefully your system remains stable. Just as an FYI, though, my system was stable for close to two months when it crashed. At the time of crash there didn't appear to be much activity on the server based on the logs available to me.

Regards,

Daymon