Server rebooting every day un-intentionally

[WolfPC]

For the past couple of weeks, my server has been crashing and rebooting at the end of the work day (ranges from 5-9pm each time). 1 out 4 times I would get a blue screen of death (have to look for the codes) and it would just be locked up. The other times the server just reboots, in the middle of tax season for an accounting firm - not good karma for me.

I have a Dell PowerEdge, Windows TSE 4.0, Citrix MF 1.8, 1GB RAM, 10-20 users max w/ 10 average daily. I do have an external Compaq DLT drive and thought it was the cause, but the other night, I took a gamble, removed it and the b/u software and the next night the server rebooted as usual. Some days it won't reboot and will make me think things are fine.
I saved the 1GB .dmp files in dated directories. What do I look for to be the cause.
I did get a message in the event log that referred to the partition be corrupt or unusable. I read that I should make sure I have a backup of that partition (luckily not the boot partition) and wipe it and restore. Is this wise and a solution? Or will I have to totally rebuild the server.

 

[CitrixEngineer]

My guess is that it's probably a printer driver, although if you're seeing corrupt partition messages, then I'd certainly agree with the restoration of it - from a pre-corrupt backup, if possible.

What concerns me most is why it became corrupt in the first place. Maybe a thorough disk scan would show up a problem with the disk.

I'd suspect temporary files as well. Clearing out temp files from the multitude of locations that they get stored in can help avoid data corruption greatly.

I'm not an avid reader of memory dumps, but the last 10 or so lines may give a clue as to what happened - as may the first 10 or so.

I hope some of this helps CitrixEngineer@yahoo.co.uk

 

[Highland]

Do you have anything scheduled on the server for around the time of the failure.....??? Is anything appearing in the event logs prior to the re-boot....??? Has the server ever re-booted at any other time (outwith 5-9pm)..???

Did you get the blue screen error code....????

 

[ih8pcs]

I have to agree with CitrixEngineer about a printer driver being the problem. Anytime my Citrix server has crashed, it has always been a printer driver causing it. As far as the partition problem: I think that's a separate issue, but one that you will have to deal with as well.

 

[WolfPC]

Thanks all.

I do remember that this all started happening shortly after we hooked up 2 new Canon digital copiers that are also network printers.

I checked the partition and there were no problems with it at all. That probably popped up after the many reboots for some reason.

There are no messages in the event logs prior to the reboots at all. It never happens at the same time. Usually it has been rebooting in the early evening (6-8pm) and being an accounting firm, people are working that late.
I think I am going to disconnect the Canons for a week adn remove the drivers. Hopefully that is the cause. It has to be, because the server has been running fine for over a year and no new programs or devices have been installed, except for the Canons.

 

[Highland]

Yeah, does sound like it could be the Canon's fault...???

Would have thought the event logs would report an error code prior to re-booting if the printers had been at fault though...??? (STOP 0x00000050 was the error with HP drivers) Are the printers heavily used...???

Just out of interest what models are the Canons..???

 

[DevilDog]

I'll ask some basic questions... odds are that you already checked these out but I'll ask anyway. Can you identify the starting point of your problem? Were any changes made at that time (24-48 hr period)? Since it does not seem to crash during the day than does any one of your users initiate any unique job/task at the end of the day? And see if you can id the info on the 1st three lines of your blue screen. I agree that the corrupted partition may be a seperate issue. -- Devil Dog --

 

[WolfPC]

The models are Canon Imagerunner 5000.
One is heavily used and I had removed the other from the server since it is right next to a high speed Xerox printer. I actually just unplugged the second Canon and will concentrate on the heavily used printer.

I can honestly say that the servers started rebooting about 2 or 3 days after I plugged the Canons in and the tech came out to set them up.

I will look up how to pull the info from the .dmp file and see if any of that is helpful.

 

[WolfPC]

OK. Server just crashed at 12:45pm, first time in a while during the day.

The blue screen was there and I wrote down some of the #s.
The first line had these 4 #'s - 0x00000028 / 0x00000002 / 0x00000000 / 0x80133D18

Under the list of devices or whatever they are:
80015000 scsiport.sys
801ea000 aic78xx.sys
801f1000 pedge.sys
801f4000 aic78u2.sys
80207000 disk.sys


Does this mean the problem is with the scsiport.sys?? I can't remember where I found info for this.
Should I just go home and never come back?

 

[CitrixEngineer]

It could well be that the problem lies with the SCSI driver - try upgrading it with the most up to date driver from the manufacturer's website.

CitrixEngineer@yahoo.co.uk

 

[Highland]

Had a quick look for the error codes.....

One possibility seems to lie with the incorrect purging of autocreate printers when users are logging out...??? Don't know if you are using any...??? Do you have ur event log set-up to log logins and logoffs, it could be one user causing the fault when logging off...???

The other DID mention a SCSI driver error and to try installing most up-to-date driver.....

What Service Pack are you running...??? Might be worth a reinstall, u never know.....???!!!

Just a final thought, do you have any virus scanners running...???