Random quits.

[PP Mguire]

So all day I had my folding going and for some reason I had WU fails on my stock server (6000+ / 260) and randomly last night while I was sleeping my 580 and 465 (and 2500k) decided to stop folding yet the program said they were. Heat isn't an issue here as our apartment is ice cold and Texas hasen't started hitting the high 90's yet. All I had to do was reboot my main rig for them to fold again, but my server rebooting doesn't seem to work.

[werty316]

What does the folding log say?

[PP Mguire]

What does the folding log say?

This is for my main rig.


[13:35:40] FAH GPU Tracker v3.54 Startup
[13:35:40] --------------------------------------------------
[13:35:40] Detected OS: Windows 7
[13:35:41] Detected CPU: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
[13:35:41] 4 Cores, 4 Threads, 3,301 MHz
[13:35:49] GPU detection completed successfully
[13:36:27] Applied new config settings
[13:36:32] Starting SMP
[13:36:32] Resumed existing WU on SMP
[13:36:32] Starting GPU 0
[13:36:32] Resumed existing WU on GPU0
[16:05:42] GPU 0 completed a WU successfully
[16:05:42] Deleted WU on GPU0
[16:05:42] Starting GPU 0
[16:05:42] Updated GPU0 config, restarting client
[16:05:42] Starting GPU 0
[16:05:42] Started new WU on GPU0
[19:20:38] GPU 0 completed a WU successfully
[19:20:38] Deleted WU on GPU0
[19:20:38] Starting GPU 0
[19:20:38] Started new WU on GPU0
[20:32:44] SMP Client completed a WU successfully
[20:32:44] Deleted WU on SMP
[20:32:44] Starting SMP
[20:32:44] Updated SMP config, restarting client
[20:32:45] Starting SMP
[20:32:45] Started new WU on SMP
[22:31:50] GPU 0 completed a WU successfully
[22:31:50] Deleted WU on GPU0
[22:31:50] Starting GPU 0
[22:31:50] Started new WU on GPU0
[01:18:31] Applied new config settings
[01:18:43] Starting GPU 1
[01:18:43] Updated GPU1 config, restarting client
[01:18:44] Starting GPU 1
[01:18:44] Started new WU on GPU1
[01:19:14] Stopped GPU 1
[01:42:53] GPU 0 completed a WU successfully
[01:42:53] Deleted WU on GPU0
[01:42:53] Starting GPU 0
[01:42:53] Started new WU on GPU0
[03:52:34] SMP Client completed a WU successfully
[03:52:34] Deleted WU on SMP
[03:52:34] Starting SMP
[03:52:34] Started new WU on SMP
[07:17:15] GPU 0 completed a WU successfully
[07:17:15] Deleted WU on GPU0
[07:17:15] Starting GPU 0
[07:17:15] Started new WU on GPU0
[10:23:50] SMP Client completed a WU successfully
[10:23:50] Deleted WU on SMP
[10:23:50] Starting SMP
[10:23:50] Started new WU on SMP
[11:53:25] Starting GPU 1
[11:53:25] Resumed existing WU on GPU1
[11:53:31] Stopped GPU 1
[11:53:36] Applied new config settings
[11:57:20] Stopped SMP
[11:57:20] Stopped GPU 0
[11:57:20] Tracker exitingNoon is when I checked my PC today and noticed it wasn't doing anything. Program said I was folding but I had no load on either so shut it down and restarted my PC.

My server just says that the WU machine has failed (unstable) and GPU will restart in 4 hours. The 6000+ keeps chugging along but the 260 keeps getting that error. I tested it for 2 hours under Furmark and completely fine. Fan at 100% so temps not an issue.

[werty316]

Sorry I meant the folding log file and not what FAH Tracker displays. I am not familiar with FAH Tracker(I use VMware w/ Ubuntu and the GPU console client) so I do not know if it logs the actual folding logs(it should though).

Check the folder where FAH Tracker is installed and see if there is an actual folding log file; it should be "fahlog.txt" or some other text file.

[PP Mguire]

Sorry I meant the folding log file and not what FAH Tracker displays. I am not familiar with FAH Tracker(I use VMware w/ Ubuntu and the GPU console client) so I do not know if it logs the actual folding logs(it should though).

Check the folder where FAH Tracker is installed and see if there is an actual folding log file; it should be "fahlog.txt" or some other text file.Just checked and it only has May 3rd from noon on up until now because I restarted. Same for my server.

[werty316]

If you are going to try to fold again be sure to check the folding logs for any errors.

[Blacksmith1]

I'm assuming you used trackers auto setup to find the gpu's. make sure in the config that the card is set to g80 and if advance methods is on turn it off, if it's off turn it on. this will get you a different set of WU's, if this doesn't fix the problem try loading the 285.xx stable driver. the 260 I had would fold fine for awhile then get one WU it didn't like , over and over and over. Check all the settings and if nothing else works delete the fahcore11.exe file. that's the actual client. tracker will D/l a new one automaticly. if all else fails delete the WU and start again.

[thebluebumblebee]

What driver version are you using?

[PP Mguire]

I'm assuming you used trackers auto setup to find the gpu's. make sure in the config that the card is set to g80 and if advance methods is on turn it off, if it's off turn it on. this will get you a different set of WU's, if this doesn't fix the problem try loading the 285.xx stable driver. the 260 I had would fold fine for awhile then get one WU it didn't like , over and over and over. Check all the settings and if nothing else works delete the fahcore11.exe file. that's the actual client. tracker will D/l a new one automaticly. if all else fails delete the WU and start again.Using the latest WHQL which I think is the 296.10 for both setups. Been looking at logs and no failed WUs on the 260 anymore after I disabled CPU folding on the 6000+. It appears that whenever I go to bed my 580 stops folding for some reason. Only got 4 hours of sleep and when I came back it still said folding but the card had no load and it was obviously cooler in the place. Only thing I did was shut my monitors off.


What driver version are you using?See above.

[werty316]

Have you considered giving v7 a try?

[PP Mguire]

Have you considered giving v7 a try?V7 of Tracker?

[werty316]

No, v7 of the official client: https://fah-web.stanford.edu/projects/FAHClient/wiki/BetaRelease

[PP Mguire]

No, v7 of the official client: https://fah-web.stanford.edu/projects/FAHClient/wiki/BetaReleaseIs it as easy to use as this Tracker program?

[thebluebumblebee]

Just turn off your screen saver.

[PP Mguire]

Just turn off your screen saver.I don't use a screen saver or any power saving features.

[werty316]

Its easy to setup and use. You configure your clients during setup and then you are off to the races. They have a pictorial setup guide here: http://folding.stanford.edu/English/WinGuide

[foxmobouser]

Is it as easy to use as this Tracker program?

I've tried the V7 just a few days ago for a few minutes, it did not pick up all my gpu's for folding, had less user control options and interface was confusing, went back to tracker. Had a lot of faild WU's on my 8800gs and 9600 gso downclocking seemed to help temps went down from 90 to 45 but still seeing a sporadic WU fails.

[Pablo 54]

Dump the tracker and use V7. I have had only one issue since going this route.
1. My 460 would not fold on the same machine with my older 285 or 8800 gtx. Fermi does not play well with other cores regardless of the tweeks you make.

The setup is easier, in my opinion than the tracker and it seems to be a bit more stable with multi gpu’s plus smp.

That being said, the room that houses my folding boxes was 81F this morning (with the AC in the house set a 70F to combat humidity and pollen). One of the boxes stopped because of heat buildup. Going to have to fold with only one box for awhile. It should not be this warm and humid in Ohio this early. We need a good old fashion ice age.

[PP Mguire]

Dump the tracker and use V7. I have had only one issue since going this route.
1. My 460 would not fold on the same machine with my older 285 or 8800 gtx. Fermi does not play well with other cores regardless of the tweeks you make.

The setup is easier, in my opinion than the tracker and it seems to be a bit more stable with multi gpu’s plus smp.

That being said, the room that houses my folding boxes was 81F this morning (with the AC in the house set a 70F to combat humidity and pollen). One of the boxes stopped because of heat buildup. Going to have to fold with only one box for awhile. It should not be this warm and humid in Ohio this early. We need a good old fashion ice age.I was folding on 2 separate machines. 260 was in my server, 580 being in my main rig. I actually had to drop out of the folding contest today because of the heat. I had the same problems but we have 2 room mates here and the AC trying to keep up froze the bedrooms out while the main part of the apartment was roasting.

[Pablo 54]

Amazing how much heat these things can kick off when running full blast.

[PP Mguire]

Amazing how much heat these things can kick off when running full blast.Yea I decided to completely leave my server out of folding and just did the 580. Still way too hot. It's just very funny to me cause I can game on both machines all day and it will never get this hot.

[Blacksmith1]

Which proves that games do not use any where near 100% of your cards. that's why I use furmark and folding to generate heat in cases for testing.

[PP Mguire]

Which proves that games do not use any where near 100% of your cards. that's why I use furmark and folding to generate heat in cases for testing.Yet, they load your GPU, VRAM, and VRAM controller up to 100% (some)? Folding only uses 1 part of your card for a long period of time. The heat from that just builds up over time where as when you are gaming it isn't loading your system for that long.