Problems with Proliant 1600 (memory errors, boot fail)

[Mat2000]

Hello all.

I'm looking for some help, ideally someone will have seen the following behaviour before and can say "Oh, I've seen this it's....."

Equpiment:
Compaq Proliant 1600r
2x Pentium3 550MHz Processors
5x 9.1Gb drives
Smart 2DH raid controller
Intergrated management screen

Problem 1:
Press the power button, the LCD shows it's welcome text (something about a version number..) for about 1 second, then goes blank. Nothing further happens, system wont boot.

Problem 2:
Press the power button, the LCD comes on with it's text and stays there, the lcd never gets to the "System Init.... etc" section. System wont boot.

Problem 3:
Press the power button, nothing happens at all, no sign of life.


So, it's a little odd, when it does boot it's fine (well, it will be when I fit a new scsi cable to the raid array) I've checked obvious things, like making sure that all cards and cables are seated correctly. I've removed and reseated the processor cage, and the expansion card tray.

I cant seem to find much of a pattern to these problems, the only thing I can see so far is that if I remove the processor cage with the machine turned off, but the power cables still plugged in, then when I go to turn it back on, I get problem 3 above.

I've been pulling my hair out trying to make sense of this, so sorry for the long post but I just hope someone has had these problems before and can tell me what is causing them and what to do about it.

Many thanks in advance for any help.

 

[Mat2000]

Replying to my own thead

I spent a chunk of the day at work reading a great many posts here and noticed some similarities. One post mentioned that the processor cage is not a fragile item and should be hit with a hammer. Well perhaps it didnt exactly say that.

So I removed and refitted the processor cage with more "enthusiasm" and sure enough - the machine booted.

It's currently running memory tests with memtest86 boot cd. I didnt mention the ram problems in the inital post, but I'm hoping they were linked to the cage not being seated correctly...

 

[marvhuffaker]

I've heard of some cases where that processor cage will just fail all of a sudden for no real reason.. One of my clients with numerous 1600's said that this happens quite frequently to them.

In their case, they just replace them as bad. But maybe after the server runs a while, the cage gets loose, or the connectors corrode or something.

Marvin Huffaker Consulting
Home of the BLUEGUNK Punk

http://www.bluegunk.com

 

[Mat2000]

Thanks for the info, that seems to be what's happened to me. It ran the memory tests just fine so I put it back online and it ran for about 40 hours and stalled, no display and IMD blank. Would not reboot after that.

I've tried all sorts of connections and nothing seems to help - I guess I need a new processor board. I dont suppose anyone knows which part actually fails on these boards? It sounds like a fuse or something, the board moves in the socket, there's a power spike and something blows?