ProLiant Servers (ML,DL,SL)
cancel
Showing results for 
Search instead for 
Did you mean: 

Blue Screens on DL580 and 6400R

 
Ayman Altounji
Valued Contributor

Blue Screens on DL580 and 6400R

I manage a Citrix Terminal server Server Farm (MetaFrame 1.8, NT 4.0, 4GB RAM, all current MS SP's, Citrix SPs, hotfixes, and compaq BIOS & support software) - with Smart array 5300 controllers and 3131 nics -
we've had 3 cpq 6400R's for about 2 years - they have had a problem with various blue-screens for quite awhile (I've been tracking for 12 months now), even after rebuilding the OS several times. The servers have been heavily loaded, and I thought that was the cause - I'm getting blue-screens approx every other day - sometimes multiple daily.

Well, 2 months ago we received 2 new DL580's w/ 3134 nics and 5300 array controllers. I've only had them online for about 4 weeks. I have started getting blue-screens on them now also.

Here is the kicker - I've got an older Cpq 5500 running Terminal Server also-same apps, but fewer users (slower cpu & less memory) Smart Array 2 controller and NetFlex nic - I haven't had a single blue screen on that server - NOT ONE!!!

I am totally at a loss to explain this. I am about to replace the NIC cards on the 6400 & DL580 with older ones from another server, and perhaps set the array controllers to read-only cache. I am willing to take any suggestions.
4 REPLIES 4
Ayman Altounji
Valued Contributor

Re: Blue Screens on DL580 and 6400R

We run Citrix on a 2000 platform with 1.8 Metaframe on both DL380s and 6400Rs and we had nothing but issues with blue screens under heavy load on the 6400s with NT 4 as the OS once we went to 2K the issues went away. We have a load of 70 users per box with 16 boxes running JDA apps which are not memory friendly by any means and as a matter of course we bounce the boxes once a day but they have gone for several weeks with issues. We have had issues with the surveyor utility and the web agent causing issues in the Citrix servers stealing CPU time until the server effectively hangs and requires a pull of the power cord and we also had issues with the the 64bit dual NICs causing memory exception errors when they were in load balancing mode. Once we disabled one of the ports since network bandwidth is just about never an issue on citrix that issue went away
Ayman Altounji
Valued Contributor

Re: Blue Screens on DL580 and 6400R

I recently found a MSKB article Q294196 refrencing some of the blue-screens we were getting - I called MS and they gave me the patch, and our blue-screens have been reduced, but we are still getting far too many, but with slightly different error codes. We do load all the compaq agents, and have dual port nics (but only 1 port is enabled)

I would like to go to Win2K on these servers, but I don't know when that will be possible

Compaq suggested I capture the crash dumps, which requires me to limit the memory on the servers - which is a performance problem on these production servers..
Ayman Altounji
Valued Contributor

Re: Blue Screens on DL580 and 6400R

A little more info for anyone interested - After applying the MS patch, I had left the servers configured to capture a crash dump file, so I had to reduce the physical memory to 2GB by the /MAXMEM switch. (had to limit memory so I would have disk space to capture the crash dump file). I had no blue-screens on the 2 DL580s for about 2 weeks. I decided the MS patch had done its job - and removed the /MAXMEM switch, returning the servers to 3GB of memory. Blue-screens resumed - but I didn't associate it with the memory. After a week or so of blue-screens, I again limited memory to 2GB via /MAXMEM - and I haven't had a single blue-screen on the 2 DL580s in 28 days. The cpq6400's are still blue-screening about twice a week collectively. The 6400's have 3GB of memory each.
I'm now suspecting some sort of Memory management issue..I've resized the partitions on my DL580's and will soon enable memory up to 3GB on them so I can try to capture crash dump files for examination.
Highlighted
Kipp Glover
Occasional Advisor

Re: Blue Screens on DL580 and 6400R

Did you every capture the dump once you repartitioned your disk? I am just curious if this pointed you in a direction as to what was causing the problem.