All of a sudden... BSOD and memory issues

[Charlie98]

My new build has been pretty much rock solid, even after a mild OC (in sig below) until yesterday. Now I've gotten 3 BSOD's just today. Because I'm an idiot, I didn't see what the BSOD error code was except one Bad Pool Header (once I remembered to look.) I not technically literate enough to know where to look for the other codes, assuming they were logged.

My system has always been LinX stable and now I can't make 2 passes.

This is pointing to the RAM, but if anyone has any other ideas, I'm all ears. I've got my new eVGA 560ti/448 coming and I need to get this sorted out before I can install it... :hmm:

 

[Puppies04]

It isn't unknown for SB chips to need a little bump in vcore after they have been used for a while. Change the volatge to 1.35 and see if the problem goes away, if so start dropping the voltage back down until you encounter instability again.

 

[Charlie98]

It isn't unknown for SB chips to need a little bump in vcore after they have been used for a while. Change the volatge to 1.35 and see if the problem goes away, if so start dropping the voltage back down until you encounter instability again.

...let's give it a try... :\

 

[pete1229]

Before anything else you should google known issues with the OCZ Agility 3 60GB SSD. This sounds like a problem I have read on various threads on Anandtech and other sites, and most end up being SSD issues.

 

[Charlie98]

The voltage thing isn't working... in fact it's getting worse. IE is crashing, Outlook is crashing...

I posted pretty much the same information over at OCZ Forum, I'll see what they say. In the meantime... BACKUP!

 

[Puppies04]

Before anything else you should google known issues with the OCZ Agility 3 60GB SSD. This sounds like a problem I have read on various threads on Anandtech and other sites, and most end up being SSD issues.

Good call I missed the SSD in the sig, any idea what firmware the drive is running on OP?

 

[Charlie98]

It's current... 2.15v from the factory.

 

[richaron]

I would set ram/cpu to stock (or even underclock at first), & use a linux live cd (to avoid ssd) to first the test ram (memtest86 from the boot manager) & afterwards the cpu using any number of stress tools. If either of these fail you know where to start looking, but it could also be your PSU (or motherboard).

Of course you also have the chkdsk utility in windows, which you can schedule from the command line or the install cd in recovery mode (unless this has changed in win7, never used it in recovery).

I think quite highly of the Phoronix dudes so this might be easiest. You could also do it with a regular Ubuntu cd like this guide

Edit: I have two Agility 3 ssds and my only problems with them has been flakey power causing corrupted files/fs.

 

[Charlie98]

I would set ram/cpu to stock (or even underclock at first), & use a linux live cd (to avoid ssd) to first the test ram (memtest86 from the boot manager) & afterwards the cpu using any number of stress tools. If either of these fail you know where to start looking, but it could also be your PSU (or motherboard).

Of course you also have the chkdsk utility in windows, which you can schedule from the command line or the install cd in recovery mode (unless this has changed in win7, never used it in recovery).

I think quite highly of the Phoronix dudes so this might be easiest. You could also do it with a regular Ubuntu cd like this guide

Edit: I have two Agility 3 ssds and my only problems with them has been flakey power causing corrupted files/fs.

Much of what you are suggesting is probably beyond my technical expertise... but I'll have to give it a try. :|

 

[Steltek]

Much of what you are suggesting is probably beyond my technical expertise... but I'll have to give it a try. :|

It might be easier to download an ISO copy of Hiren's Boot CD and burn it to an CD. Boot the machine with it. Look through the application menus, and find the Testing Tools section. Under the Testing Tools utilities, you'd first want to run Memtest 86+ to test your memory followed by a session of Prime95 to stress test your CPU.

 

[richaron]

It might be easier to download an ISO copy of Hiren's Boot CD and burn it to an CD. Boot the machine with it. Look through the application menus, and find the Testing Tools section. Under the Testing Tools utilities, you'd first want to run Memtest 86+ to test your memory followed by a session of Prime95 to stress test your CPU.

much more less confusing :thumbsup:

 

[Charlie98]

OK, soooooo....

I run Memtest as is, and I get one error. I pull one RAM card out, no errors. I swap cards, no errors. I put both cards back in, no errors... and I made 10 passes of LinX with no errors.

But there are still problems, much like before, and I suspect it's either a problem with the SSD or data corruption.

 

[Charlie98]

I have both the W7 install disk and I created a Windows repair disk back in February... would running either of these do me any good assuming my OS is corrupted? There is still the possibility of bad SSD firmware, but I believe the reflash will nuke any data on the SSD, and the last system image was made last week... certainly with any problems still rolled up in it.

As I say, all this stuff is a little over my head, I'd be glad to hear any suggestions or if I'm missing something obvious.

 

[richaron]

I find it odd there were any problems with memtest86 at all. This suggests a ram instability of some sort (maybe it just wasn't seated properly), but if it's passing fulls test now it should be fine.

Of course this could have corrupted your SSD, I would run chkdsk and see if that helps.

 

[bononos]

..... As I say, all this stuff is a little over my head, I'd be glad to hear any suggestions or if I'm missing something obvious.

The obvious things would be dust/loose connections and overclocking.

 

[Charlie98]

The obvious things would be dust/loose connections and overclocking.

Well... you know I have dust issues...

The BSOD's seem to have gone away for now, the big issue is the programs crashing. I can't afford to be putting a big wad of information into QuickBooks and have it crash in the middle of it.

groberts101 (over on OCZ forums) suggested I reset the CMOS and work up from stock again... I may try that, but the consensus was to secure erase the SSD, reflash the firmware and start over.

 

[richaron]

Sandforce drives have a stigma from the earlier firmwares, so now a lot of people are supposed experts. A simple format & reinstall will fix filesystem issues. CHKDSK could too.

It is dumb to flash anything unless you are sure it will help, or absolute last resort.

 

[mikeymikec]

You can check the event log (the system log specifically) for an entry (either classed as 'error' or 'critical') called 'BugCheck' IIRC. That will give you all the codes, or at least enough codes to google for.

 

[Charlie98]

Sandforce drives have a stigma from the earlier firmwares, so now a lot of people are supposed experts. A simple format & reinstall will fix filesystem issues. CHKDSK could too.

It is dumb to flash anything unless you are sure it will help, or absolute last resort.

I've always been a big supporter of my OCZ drive, even given my limited experience. Never had a problem with it up until now, and I'm still not sure it's just one thing that's the problem.

CHKDSK didn't show any errors that I saw, but I may not be looking in the right place.

 

[mikeymikec]

To view the results of the disk check, you need to load the Windows Event Viewer:
Click on the ‘Start’ menu, then (either click directly on ‘Control Panel’, or if that isn’t on your Start menu, then go into the ‘settings’ menu), and left-click on ‘Control Panel’.
In the Control Panel, double-click on ‘Administrative Tools’. Then double-click on ‘Event Viewer’.
In the Event Viewer (in Windows XP), in the left-hand pane, left-click on the ‘Application log’. The list is normally sorted by date, with the most recent events at the top of the list. Browse the list for the time that you did the disk check, and you should find an entry whose source is ‘winlogon’, and double-click on it. The window that opens will contain a record of what the disk checking utility found during its scan. Please be aware that at lot of the information reported is in a technical manner, but I would be concerned if any 'read failures' or more than zero kilobytes of bad sectors are mentioned, as that is normally a sign that there is a serious problem with the hard disk.
The way of accessing the Application Log is fairly similar in Windows Vista and 7 – but before you can see the Application Log, you have to double-click on 'Windows Logs', then left-click on 'Application', then look for an entry whose source is 'wininit', and double-click on it.

 

[Charlie98]

If I'm reading this correctly, everything is peachy...

Checking file system on C: The type of the file system is NTFS. Volume label is SSD Boot. A disk check has been scheduled. Windows will now check the disk. CHKDSK is verifying files (stage 1 of 5)... 173568 file records processed. File verification completed. 164 large file records processed. 0 bad file records processed. 0 EA records processed. 7675 reparse records processed. CHKDSK is verifying indexes (stage 2 of 5)... 222948 index entries processed. Index verification completed. 0 unindexed files scanned. 0 unindexed files recovered. CHKDSK is verifying security descriptors (stage 3 of 5)... 173568 file SDs/SIDs processed. Cleaning up 59 unused index entries from index $SII of file 0x9. Cleaning up 59 unused index entries from index $SDH of file 0x9. Cleaning up 59 unused security descriptors. Security descriptor verification completed. 24691 data files processed. CHKDSK is verifying Usn Journal... 34509584 USN bytes processed. Usn Journal verification completed. CHKDSK is verifying file data (stage 4 of 5)... 173552 files processed. File data verification completed. CHKDSK is verifying free space (stage 5 of 5)... 6197951 free clusters processed. Free space verification is complete. CHKDSK discovered free space marked as allocated in the master file table (MFT) bitmap. Windows has made corrections to the file system. 58510335 KB total disk space. 33357100 KB in 145259 files. 85568 KB in 24692 indexes. 0 KB in bad sectors. 275863 KB in use by the system. 65536 KB occupied by the log file. 24791804 KB available on disk. 4096 bytes in each allocation unit. 14627583 total allocation units on disk. 6197951 allocation units available on disk. Internal Info: 00 a6 02 00 ea 97 02 00 a4 ab 04 00 00 00 00 00 ................ da 03 00 00 fb 1d 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ Windows has finished checking your disk. Please wait while your computer restarts.

 

[mikeymikec]

Well, the results look ok, however it doesn't mean that the drive is definitely ok. I've seen disk checks turn up nothing and yet the drive is still dodgy. Any disk errors in the system log?

It should say which device. '\Device\Harddisk0' is usually the first hard disk in the system.

PS - I have almost zero experience of SSDs.