interesting thread, didn't have time to read more than a few pages yet, I found this by googling event id 51.
I'll write down all details about my case, maybe it helps someone. sorry for the painstakingly detailed report.
I have three hard disks in my PC, two SATA (WD6400AAKS 640GB, SAMSUNG 130SJ 1TB) , one IDE (WD1200JB 120GB), I'm no longer using the IDE one much, I read a file off it maybe once in a month![:)]( ":)")
the system is 2 years old, both the PC and the XP installation.
the PC also has a asus P5L 1394 motherboard, intel core 2 duo E6620 or something like that, some 2x1GB DDR2 RAM, cheap low end GPU, DVD-RW, a 400W 2 year old PSU and that's it.
oh, actually I had 2 hdd's (one SATA, a WD5000AAKS 500GB, and the IDE WD120) until April this year. but then end of April when I wanted to back up the entire SATA (500GB) drive to an external (also SATA, via USB, self powered via its own AC adapter) drive, the problem started: a few mins into the backup process, the drive skipped reading files then crashed bringing the entire system down as XP was installed on this drive.
when I tried to reboot (after a manual shut down), I got "third master disk error" in the BIOS. then I unplugged the USB drive and after 10 mins, I retried and it booted, I was able to do the backup, the files that failed before, were read and copied fine too!! and all worked OK for a few weeks - until the previous week.
note, after the problem in April happened, I quickly put in another, brand new SATA drive (the Samsung 1TB one), but did not use it much yet. I maybe accessed it 2 times in 2 months. anyway, from here, I've had 3 disks installed in the PC. but note, no important software changes happened.
now, a few days ago - last friday - I went to the PC after I got up. note, the PC is on 24/7 and I leave it at the logon screen when I go to bed. now, the PC greeted me with "unknown hard error" messagebox (title: "Windows - System error"). it had no extra file paths noted in it. clicking it, it would not go away.
I logged into the system anyway and noticed that I could not open any new programs, not even taskmanager. one program was left open from last night, an IRC client, this was still running fine, but as I found out later, it had stopped logging the channels to disk, a few hours before according to timestamps. this all points to the drive stopping working / being dropped out.
I had to shut down, reboot, system worked OK for maybe 10-20 mins then it bluescreened. I did not note the exact error in it. I shut down, reboot, went to the toilet while Windows was booting, I got back and god, instead of logon screen I'd just see another BSOD. oh, and it could not dump the physical memory to disk even though it was trying to in both cases. I seem to recall that in one of these tries I had the same "third master disk error" first.
now I didn't try to boot again, instead I got a CD with a partition manager thingie, a recovery disk, I booted from that and copied partition C: from the WD500 to that new other Samsung drive. (one sector could not be copied, the rest of 25GB was copied fine.) I've been able to boot from this disk just fine without BSODs in the following days.
ok, after this partition copy, I then shut down the PC as I had to leave and I was paranoid about leaving it on like I usually do. next day, I booted the PC again and I tried to back up all the files off this old WD500 drive (2 years old) because I guessed the drive was dying - I was backing up to an external USB drive. during the backup (a lot of GB's have already been copied by then) it again started failing to read files and then the following thing happened instead of a system crash as the drive no longer had the system on it: the drive just dropped out, Windows could not see its partitions, in event log I got event id 12. along with several of 50, 51, 57 and 26...
(12: "The device 'WDC WD5000AAKS-00YGA0' (IDE\DiskWDC_WD5000AAKS-00YGA0___________________12.01C02\5&3003bd5e&0&0.0.0) disappeared from the system without first being prepared for removal."
26: "Application popup: Windows - Delayed Write Failed : Windows was unable to save all the data for the file D:\xxxxxxx" - for 26, I was getting the exact file paths too.
50: "{Delayed Write Failed} ... etc. etc."
51: "An error was detected on device \Device\Harddisk1\D\xxx during a paging operation"
57: "The system failed to flush data to the transaction log. Corruption may occur")
then I also tried to load a S.M.A.R.T program and it could not access the disk.
the system (Samsung) drive was still functioning fine at this point but I quickly shut down the entire system about 5mins (at most) after seeing the WD500 drive dropped like this. then I put the hdd in the fridge for 3 hours![:)]( ":)")
after that I stuck it in as an external drive via USB and I was able to finish the backup without any hiccups. I then shut down the old drive, it's sitting on the shelf now but I'm suspecting it actually doesn't have a problem...? see below why I think so... (I haven't tested the disk though!!!)
so um, this was on saturday. after the fridge incident, I was able to use the system on newer Samsung hdd just fine, no problems. note, I put in another SATA disk in place of the "dead" old one (this is the WD640). previously this disk was actually the external backup of that old WD. so again I had 3 hard disks, 2 SATA, 1 IDE.
so, on sunday I decided to backup both SATA drives (IDE already had a backup) to another external drive, because at this point I did not really have a copy of my data if we count the WD500 drive as "bad". and at this point I really believed everything was fine other than a bad old drive.
but noooo, after quite a few GB's got copied to the external USB drive off one of the new SATA drives (the WD640, not the system drive), I got skipped files again and then Windows lost the drive and then 1 minute later I noticed that the system drive (the Samsung) also got dropped!!! Windows was in the same state as on friday when the problems started with the hard error messagebox. I was able to continue using the IRC client that was already running, I was able to use the file manager that was already running but I could not open any new programs anymore and I could not access any of the disks.
at this time I didn't get much in the event log because the system drive must be functional to be able to write the log, obviously. actually, I did get one single entry (event id 57) into the event log before the system drive also got dropped.
I rebooted after a couple of mins, everything was OK. but I had to sleep soon, so turned the system off again soon.
then next day, monday, I booted the system OK, I continued the backup to external USB, and after 2 hours of lots of copying, the disk (still the WD640) started having file read errors. I cancelled the backup process. after 1-2mins I retried, it was fine for 30secs then again file read errors and I noticed the transfer speed also really slowed down, from 15-25MB/sec to 500K/sec or something like that. so at that point I cancelled again. I retried again, same thing, cancel.
at this point I decided to download another S.M.A.R.T program, by the way the one I'd used before never showed an error until now. but after this last incident I checked quickly and it showed changes in: Seek Error Rate, Spin-up Retry Count, Calibration Retry Count, the weird thing is that these start from values of 200, 100, 100 initially and decreased value means problem. instead, this disk updated these three values to 253. WHAT?!?! that's not even meaningful. and indeed, when I checked later, the second S.M.A.R.T program reported the disk (WD640) as perfectly healthy. (the first such program did notice that there was something awry, though.)
soo, I decided to download and install this second program but I could not even start the download - I opened a browser window and tried to load google in it and it could not load. this was about 3mins after the first file read failure on the WD640 disk. now, I checked in the file manager (already running in RAM), and none of the three disks were accessible anymore!!! neither the non-system SATA, nor the system disk SATA, nor the IDE disk!!! (I forgot to check the external USB drive that was still connected but I have a hunch that it was probably still OK.)
I was able to ask Windows to shut down, then I went out and bought a new PSU as I didn't have another idea at that moment. I read online and here too that this could be a PSU issue.
in my case it is clearly not a SATA-related issue, as you can see, even my IDE disk got dropped, and I doubt it is a software issue as I didn't update or change anything in the system and I don't think it is XP related anyway due to those intermittent disk errors in the BIOS on booting.
now the new PSU is in the PC, I've finished backup'ing that WD640 drive and part of the Samsung drive, and no problems yet, I will report back with an update on this later.
I'll write down all details about my case, maybe it helps someone. sorry for the painstakingly detailed report.
I have three hard disks in my PC, two SATA (WD6400AAKS 640GB, SAMSUNG 130SJ 1TB) , one IDE (WD1200JB 120GB), I'm no longer using the IDE one much, I read a file off it maybe once in a month
the system is 2 years old, both the PC and the XP installation. the PC also has a asus P5L 1394 motherboard, intel core 2 duo E6620 or something like that, some 2x1GB DDR2 RAM, cheap low end GPU, DVD-RW, a 400W 2 year old PSU and that's it.
oh, actually I had 2 hdd's (one SATA, a WD5000AAKS 500GB, and the IDE WD120) until April this year. but then end of April when I wanted to back up the entire SATA (500GB) drive to an external (also SATA, via USB, self powered via its own AC adapter) drive, the problem started: a few mins into the backup process, the drive skipped reading files then crashed bringing the entire system down as XP was installed on this drive.
when I tried to reboot (after a manual shut down), I got "third master disk error" in the BIOS. then I unplugged the USB drive and after 10 mins, I retried and it booted, I was able to do the backup, the files that failed before, were read and copied fine too!! and all worked OK for a few weeks - until the previous week.
note, after the problem in April happened, I quickly put in another, brand new SATA drive (the Samsung 1TB one), but did not use it much yet. I maybe accessed it 2 times in 2 months. anyway, from here, I've had 3 disks installed in the PC. but note, no important software changes happened.
now, a few days ago - last friday - I went to the PC after I got up. note, the PC is on 24/7 and I leave it at the logon screen when I go to bed. now, the PC greeted me with "unknown hard error" messagebox (title: "Windows - System error"). it had no extra file paths noted in it. clicking it, it would not go away.
I logged into the system anyway and noticed that I could not open any new programs, not even taskmanager. one program was left open from last night, an IRC client, this was still running fine, but as I found out later, it had stopped logging the channels to disk, a few hours before according to timestamps. this all points to the drive stopping working / being dropped out.
I had to shut down, reboot, system worked OK for maybe 10-20 mins then it bluescreened. I did not note the exact error in it. I shut down, reboot, went to the toilet while Windows was booting, I got back and god, instead of logon screen I'd just see another BSOD. oh, and it could not dump the physical memory to disk even though it was trying to in both cases. I seem to recall that in one of these tries I had the same "third master disk error" first.
now I didn't try to boot again, instead I got a CD with a partition manager thingie, a recovery disk, I booted from that and copied partition C: from the WD500 to that new other Samsung drive. (one sector could not be copied, the rest of 25GB was copied fine.) I've been able to boot from this disk just fine without BSODs in the following days.
ok, after this partition copy, I then shut down the PC as I had to leave and I was paranoid about leaving it on like I usually do. next day, I booted the PC again and I tried to back up all the files off this old WD500 drive (2 years old) because I guessed the drive was dying - I was backing up to an external USB drive. during the backup (a lot of GB's have already been copied by then) it again started failing to read files and then the following thing happened instead of a system crash as the drive no longer had the system on it: the drive just dropped out, Windows could not see its partitions, in event log I got event id 12. along with several of 50, 51, 57 and 26...
(12: "The device 'WDC WD5000AAKS-00YGA0' (IDE\DiskWDC_WD5000AAKS-00YGA0___________________12.01C02\5&3003bd5e&0&0.0.0) disappeared from the system without first being prepared for removal."
26: "Application popup: Windows - Delayed Write Failed : Windows was unable to save all the data for the file D:\xxxxxxx" - for 26, I was getting the exact file paths too.
50: "{Delayed Write Failed} ... etc. etc."
51: "An error was detected on device \Device\Harddisk1\D\xxx during a paging operation"
57: "The system failed to flush data to the transaction log. Corruption may occur")
then I also tried to load a S.M.A.R.T program and it could not access the disk.
the system (Samsung) drive was still functioning fine at this point but I quickly shut down the entire system about 5mins (at most) after seeing the WD500 drive dropped like this. then I put the hdd in the fridge for 3 hours
after that I stuck it in as an external drive via USB and I was able to finish the backup without any hiccups. I then shut down the old drive, it's sitting on the shelf now but I'm suspecting it actually doesn't have a problem...? see below why I think so... (I haven't tested the disk though!!!)so um, this was on saturday. after the fridge incident, I was able to use the system on newer Samsung hdd just fine, no problems. note, I put in another SATA disk in place of the "dead" old one (this is the WD640). previously this disk was actually the external backup of that old WD. so again I had 3 hard disks, 2 SATA, 1 IDE.
so, on sunday I decided to backup both SATA drives (IDE already had a backup) to another external drive, because at this point I did not really have a copy of my data if we count the WD500 drive as "bad". and at this point I really believed everything was fine other than a bad old drive.
but noooo, after quite a few GB's got copied to the external USB drive off one of the new SATA drives (the WD640, not the system drive), I got skipped files again and then Windows lost the drive and then 1 minute later I noticed that the system drive (the Samsung) also got dropped!!! Windows was in the same state as on friday when the problems started with the hard error messagebox. I was able to continue using the IRC client that was already running, I was able to use the file manager that was already running but I could not open any new programs anymore and I could not access any of the disks.
at this time I didn't get much in the event log because the system drive must be functional to be able to write the log, obviously. actually, I did get one single entry (event id 57) into the event log before the system drive also got dropped.
I rebooted after a couple of mins, everything was OK. but I had to sleep soon, so turned the system off again soon.
then next day, monday, I booted the system OK, I continued the backup to external USB, and after 2 hours of lots of copying, the disk (still the WD640) started having file read errors. I cancelled the backup process. after 1-2mins I retried, it was fine for 30secs then again file read errors and I noticed the transfer speed also really slowed down, from 15-25MB/sec to 500K/sec or something like that. so at that point I cancelled again. I retried again, same thing, cancel.
at this point I decided to download another S.M.A.R.T program, by the way the one I'd used before never showed an error until now. but after this last incident I checked quickly and it showed changes in: Seek Error Rate, Spin-up Retry Count, Calibration Retry Count, the weird thing is that these start from values of 200, 100, 100 initially and decreased value means problem. instead, this disk updated these three values to 253. WHAT?!?! that's not even meaningful. and indeed, when I checked later, the second S.M.A.R.T program reported the disk (WD640) as perfectly healthy. (the first such program did notice that there was something awry, though.)
soo, I decided to download and install this second program but I could not even start the download - I opened a browser window and tried to load google in it and it could not load. this was about 3mins after the first file read failure on the WD640 disk. now, I checked in the file manager (already running in RAM), and none of the three disks were accessible anymore!!! neither the non-system SATA, nor the system disk SATA, nor the IDE disk!!! (I forgot to check the external USB drive that was still connected but I have a hunch that it was probably still OK.)
I was able to ask Windows to shut down, then I went out and bought a new PSU as I didn't have another idea at that moment. I read online and here too that this could be a PSU issue.
in my case it is clearly not a SATA-related issue, as you can see, even my IDE disk got dropped, and I doubt it is a software issue as I didn't update or change anything in the system and I don't think it is XP related anyway due to those intermittent disk errors in the BIOS on booting.
now the new PSU is in the PC, I've finished backup'ing that WD640 drive and part of the Samsung drive, and no problems yet, I will report back with an update on this later.