IBM raid driver deactivating device during automatic bootup.

[bbowers]

Netware 6.5 SP6

About a week ago we had an issue with the server running at unusually high utilization. We rebooted the server and it hung up on a few errors:

CDM deactivated IBM Raid due to device failure
filesys.nlm could not be located
Device driver deactivated IBM Raid due to device failure

These are not word for word, but very similar to what we are getting. Now for the kicker. If we boot up with F8 and do a server -na -ns -nl and manually load startup.ncf all the stages 1-5 and the autoexec.ncf it will boot just fine and everything operates properly. I have updated to the latest bios, raid firmware, and raid drivers. It is an IBM xSeries 226 with IBM Raid device. It almost seems like the process it witch it auto boots. Any ideas???

 

[marvhuffaker]

To me, if I get a RAID failure, I call IBM and get a tech out to start replacing parts. Especially when you are specifically saying that it reported a RAID failure.

A lot of times when you have hardware failures, rebooting will clear the error temporarily, but it's sure to come back and bite you at the worst possible time. Get it replaced before this happens.

Marvin Huffaker, MCNE

http://www.redjuju.com

 

[bbowers]

IBM had me pull all the hardware logs and they show no problems with hardware, raid controller, drives, ect. The strange thing is it manually boots everything with no issues.

 

[marvhuffaker]

Doesn't matter. I'd push them till they replace it. Either that or provide you with the answer to why that's happening - it's their hardware. If they won't budge, I know some tricks you can do with paperclips.


Marvin Huffaker, MCNE

http://www.redjuju.com

 

[bbowers]

I ended up replacing the SCSIHD.CDM file on the c: drive, when I did this the sercfg.* files all changed to today's day. TID 10094155 says to try deleting these files. So that may have resolved it or the SCSIHD.CDM was corrupt. Issue has been resolved.

 

[sstoppel]

bbowers, thanks for posting the answer. Too often threads just die off without resolution.

 

[bbowers]

I just had another server do the exact same thing. Both of these had SP6 on them. Both these servers have been rebooted after the SP6 was installed with no issues. Both servers experianced high CPU utilization (reason we needed to reboot). When we rebooted they cam up with the above problem.

 

[bbowers]

Ok, I have a SCSIHD.CDM under c:\nwserver dated 3/29/06 and one under
c:\nwserver\drivers dated 6/2/06. The one under drivers is the SP6
version according to the file list. What one does the startup.ncf use
during the boot process? I am guessing it is trying to use the SP5
version under C:\nwserver

 

[marvhuffaker]

It depends on the load order.. Do this at the console and it should show which version loaded and where it got it from.

m scsihd.cdm