SCSI drives and the BSOD#%(^*#^%#$^!!!

[tahoe2]

We have 3 SCSI drives in our database server. Yesterday, I noticed only the first HD was showing up in disk mgmt, so I shut down and restarted the server. I noticed the SCSI BIOS was not loading, and immediately after the mpkernel screen loaded, I got the blue screen of death, with the following information:

STOP: 0x0000001E (0xc0000005, 0xE136887F, 0x00000001, 0x6E80AF12)
KMODE EXCEPTION IRQ1 IE SYSVER 0xF0000565

I restarted the server, same results. I chose Last Known Good Config, Same results.

I restarted in VGA mode, same results.
I changed SCSI Cables, restarted, same results.
I put stuff back the same way I found it, KICKED THE SERVER,
and it FINALLY restarted in VGA mode.
Any ideas?
I'm tempted to think it's the SCSI controller card.

Corie

 

[jtb]

Hey! Even Dr. Who needed to kick his TARDIS sometimes to get it started...

I would recommend removing and reseating all the cards in any machine once a year--just to knock out any dust or grit or oxidation that might have formed... We used to have the same issues back in those wild and crazy S100 Bus days... (think Micromation, Cromemco, Morrow Designs, and CP/M on 8" floppies)

As to whether your SCSI card might be bad? It might be... But I wouldn't know where to find any diagnostics for that...

JTB
Solutions Architect
MCSE-NT4, MCP+I, MCP-W2K, CCNA, CCDA,
CTE, MCIWD, i-Net+, Network+
(MCSA, MCSE-W2K, MCIWA, SCSA, SCNA in progress)

 

[markharr]

I have posted a new FAQ on determining the cause of a Stop 0x0000001E. Please check this when it is reviewed and posted by the administrator.

In short though, I think it is most likely the SCSI card from your description. There is no easy way to say that is the case for sure. The quickest way is to replace the SCSI card and re-test.

 

[markharr]

The FAQ I posted is at this page:

http://www.tek-tips.com/faqs.cfm?spid=55&sfid=3498

 

[tahoe2]

Thanks, I found the FAQ to be very informative. I had a little trouble with the debuggers, but I'll work on that this week.
Since no new drivers or hardware (or fixes) were added to this machine in the last 6 months, I have some digging to do, but the most important fact is, I have a drawer full of good backups, and the machine is under warranty, so I'm off the hook, so to speak. (it just gets me that I can't fix it myself...)

Corie

 

[tahoe2]

OK, I have just finished about 4 hours of reading, and I am no closer to understanding debuggers than before.
(now for my lament...)
Why oh why is there no list of Stop or Exception errors posted anywhere?
(ok, done now)
Corie

Ps, since in my experience most stop screens are caused by faulty video frivers (and since I had to reinstall video drivers when the server finally came back up) I am going to assume (I KNOW!!) that its the video and replace the video card and restart. You'll know tomorrow.....

 

[markharr]

As I recall, the stop errors are actually in the Intel processor documentation. I don't have them handy at the moment to confirm that.

I don't think that would be all that useful anyway. It's one thing to know the definition of a stop 1E but another thing entirely to know what it means.

The most important thing to look at from your point of view is the first line of the stop error code and a list of the addresses of the various modules that are loaded.

The first parameter tells you the type of error with 0x00000005 being the most common which is usually caused by a null parameter being passed up to a function.

The second parameter is the address where the exception occurred. Use this to determine which module actually caused the system to stop.

Of course, the exception could be caused by a higher up function in the stack passing a bad parameter like a null pointer. You can look in the stack back trace to see what was passed up the stack chain.

That's about as far as you can be reasonably expected to go unless you have access to the source code and the expertise to interpret that.