Review of method to diagnose bluescreens?
Hi Experts,
I work at an IT repair shop and we are constantly (as you can imagine) running across bluescreen issues. My normal process for diagnosing these goes like this.
1. test ram with windows memory diag
2. analyze dump files with bluescreenview
3. test hard drive for bad sectors
4. scan for viruses
if all the above come out good/clean and the .dmp files don't point to anything definitive i continue with a few more steps
5. update all drivers from manufactures website
6. uninstall any unecessary apps
7. uninstall/reinstall antivirus software
8. test run for 12 hrs to see if bluescreen occurs - if no bluescreen's we give back to customer - if bluescreens occur again for customer or one occurs during 12 hr run then i continue with a few more steps
9. hook up a new power supply of at least 100 watts greater than one currently installed and let it run for 12 hrs - if no bluescreens we give it back to customer - if it still bluescreens we replace the ram and give it back to customer
10. if we still see bluescreens after replacing the ram we backup data and wipe the system
What do you think of what i'm doing here and can anyone provide input on a better way to be more thorough on reading dump files? It seems like i'm running constantly into ntfs.sys, hal.dll, ntoskrnl.exe bluescreens and from what i read on these they don't point to anything definitive...
I work at an IT repair shop and we are constantly (as you can imagine) running across bluescreen issues. My normal process for diagnosing these goes like this.
1. test ram with windows memory diag
2. analyze dump files with bluescreenview
3. test hard drive for bad sectors
4. scan for viruses
if all the above come out good/clean and the .dmp files don't point to anything definitive i continue with a few more steps
5. update all drivers from manufactures website
6. uninstall any unecessary apps
7. uninstall/reinstall antivirus software
8. test run for 12 hrs to see if bluescreen occurs - if no bluescreen's we give back to customer - if bluescreens occur again for customer or one occurs during 12 hr run then i continue with a few more steps
9. hook up a new power supply of at least 100 watts greater than one currently installed and let it run for 12 hrs - if no bluescreens we give it back to customer - if it still bluescreens we replace the ram and give it back to customer
10. if we still see bluescreens after replacing the ram we backup data and wipe the system
What do you think of what i'm doing here and can anyone provide input on a better way to be more thorough on reading dump files? It seems like i'm running constantly into ntfs.sys, hal.dll, ntoskrnl.exe bluescreens and from what i read on these they don't point to anything definitive...
If you need help with a particular issue we can help you ...upload the dmp file. ....You can use microsoft debugging tools for proper anlaysis of dump files.
http://msdn.microsoft.com/en-us/windows/hardware/gg463009
Ded9
http://msdn.microsoft.com/en-us/windows/hardware/gg463009
Ded9
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Side note: Replacing hardware seems like a bigger cost that wiping the system. Not sure why you would try replacing the power supply or ram before a wipe. Especially if you ran hardware diagnostics and the system was running fine previously.
Also if you can not easily reproduce the blue screen then trouble shooting it within the os with drivers is going to be hard and time consuming. A wipe and reload will take less time and resolve any software issues 99% of the time.
Also if you can not easily reproduce the blue screen then trouble shooting it within the os with drivers is going to be hard and time consuming. A wipe and reload will take less time and resolve any software issues 99% of the time.
ASKER
Thank you all for the info, that about clears things up for me,
Here are the 3 most recent bluescreen dumps from a machine that pushed me to ask this question.
I ended up replacing the ram on this one even though it tested good with windows memory diag.
I couldn't find anything definitive that the bluescreens pointed to - do you see anything different on those?
032811-16567-01.dmp 032811-28594-01.dmp 032411-14274-01.dmp
Here are the 3 most recent bluescreen dumps from a machine that pushed me to ask this question.
I ended up replacing the ram on this one even though it tested good with windows memory diag.
I couldn't find anything definitive that the bluescreens pointed to - do you see anything different on those?
032811-16567-01.dmp 032811-28594-01.dmp 032411-14274-01.dmp
Thanks Working on it
Ded9
Ded9
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Could just be the chipset drivers
Wish I could edit. Check you bios version as well.
ASKER
ded9 - On that system we replaced the ram with known-good and it's been 3 days now with no bluescreens. They were previously getting at least 1 a day. Hopefully that resolved it but if not i've never run across a way to test processors, do you recommend something?
BCipollone - i updated the chipset drivers and bios and we still had a bluescreen after...thanks for the input.
BCipollone - i updated the chipset drivers and bios and we still had a bluescreen after...thanks for the input.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thank you all for the input, very good advice on how to approach bluescreens.
Glad to have posted.
Glad to have posted.
This article describes much of why this happens.
http://support.microsoft.com/kb/314477