Message boards :
Number crunching :
BSOD with Seti, tried reinstall of everything...
Message board moderation
Author | Message |
---|---|
lupo Send message Joined: 29 Aug 10 Posts: 91 Credit: 4,736,407 RAC: 0 |
So I was getting regular BSOD when running SETI with the GPU tasks. I decided I would just format my machine (it needed it) and do a clean reinstall. I zipped up all my WU and set them aside and put them back in place once I put BOINC and Lunatic's back on the machine. I noticed that SETI will not process these 3K or so WU I saved. Maybe it has something with them being associated with the different version of software, I'm not sure. Does anyone know how to get the WU working again under this new install of BOINC. I put them back in the same folder as before but BOINC does not see them nor it is downloading nay new WU (guess servers are down again).. Thanks, Adam Edit: Figured It would be a good idea to post my messages: 12/22/2010 2:22:06 PM Starting BOINC client version 6.10.58 for windows_x86_64 12/22/2010 2:22:06 PM log flags: file_xfer, sched_ops, task 12/22/2010 2:22:06 PM Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3 12/22/2010 2:22:06 PM Data directory: E:\ProgramData\BOINC 12/22/2010 2:22:06 PM Running under account lupo 12/22/2010 2:22:06 PM Processor: 12 GenuineIntel Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz [Family 6 Model 44 Stepping 2] 12/22/2010 2:22:06 PM Processor: 256.00 KB cache 12/22/2010 2:22:06 PM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx tm2 popcnt aes pbe 12/22/2010 2:22:06 PM OS: Microsoft Windows 7: Enterprise x64 Edition, (06.01.7600.00) 12/22/2010 2:22:06 PM Memory: 5.99 GB physical, 11.98 GB virtual 12/22/2010 2:22:06 PM Disk: 186.31 GB total, 170.04 GB free 12/22/2010 2:22:06 PM Local time is UTC -5 hours 12/22/2010 2:22:06 PM NVIDIA GPU 0: GeForce GTX 470 (driver version 26099, CUDA version 3020, compute capability 2.0, 1249MB, 726 GFLOPS peak) 12/22/2010 2:22:06 PM NVIDIA GPU 1: GeForce GTX 470 (driver version 26099, CUDA version 3020, compute capability 2.0, 1249MB, 726 GFLOPS peak) 12/22/2010 2:22:06 PM NVIDIA GPU 2: GeForce GTX 470 (driver version 26099, CUDA version 3020, compute capability 2.0, 1248MB, 726 GFLOPS peak) 12/22/2010 2:22:06 PM SETI@home Found app_info.xml; using anonymous platform 12/22/2010 2:22:06 PM AQUA@home URL http://aqua.dwavesys.com/; Computer ID 82466; resource share 100 12/22/2010 2:22:06 PM SETI@home URL http://setiathome.berkeley.edu/; Computer ID 5707955; resource share 100 12/22/2010 2:22:06 PM SETI@home General prefs: from SETI@home (last modified 21-Dec-2010 23:08:43) 12/22/2010 2:22:06 PM SETI@home Host location: none 12/22/2010 2:22:06 PM SETI@home General prefs: using your defaults 12/22/2010 2:22:06 PM Reading preferences override file 12/22/2010 2:22:06 PM Preferences: 12/22/2010 2:22:06 PM max memory usage when active: 4908.06MB 12/22/2010 2:22:06 PM max memory usage when idle: 5521.57MB 12/22/2010 2:22:06 PM max disk usage: 50.00GB 12/22/2010 2:22:06 PM (to change preferences, visit the web site of an attached project, or select Preferences in the Manager) 12/22/2010 2:22:06 PM Not using a proxy 12/22/2010 2:22:06 PM Suspending computation - user request |
Gundolf Jahn Send message Joined: 19 Sep 00 Posts: 3184 Credit: 446,358 RAC: 0 |
I zipped up all my WU and set them aside and put them back in place once I put BOINC and Lunatic's back on the machine. I hope you backed up your whole data directory, otherwise the stored WUs are lost. Gruß, Gundolf |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
If you get the BSOD, then there is something strange going on in your own computer. To miss some of the earlier work, shouldn't be a bother, just wait to get some new WUs. |
SciManStev Send message Joined: 20 Jun 99 Posts: 6652 Credit: 121,090,076 RAC: 0 |
My 980 kept blue screening like that with two 480's. It didn't matter what speed things were runnig at. It would still Blue Screen. I turned off hyperthreading, and the problem went away. Steve Warning, addicted to SETI crunching! Crunching as a member of GPU Users Group. GPUUG Website |
gizbar Send message Joined: 7 Jan 01 Posts: 586 Credit: 21,087,774 RAC: 0 |
What power supply are you running with that? According to your post, you have 3 GTX470's in that rig? It may be stressing a rail too much, trying to run all 3. If it only BSOD's on running GPU units, I'd start looking there. Try running one card at a time and seeing if it BSOD's, then 2, then 3. If it runs 2 successfully, then try a different combination, but still 2. Then try all the 2 card combinations, then try 3 cards again. Even if I get a WU fail for some reason, it only gives me a computation error. It does not give me a BSOD. Yes, Seti is in a downtime at the moment. They are doing the regular backup and trying to re-stripe Oscar to a different size. It's taking a while, because all the data has to be copied to Carolyn, then Oscar has his stripes changed, then all the data has to be copied back again. NOT a trivial operation. Giz. A proud GPU User Server Donor! |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
I zipped up all my WU and set them aside and put them back in place once I put BOINC and Lunatic's back on the machine. Well the OP has a new host now, so i guess not, He could try dropping his old Wu's into his project folder, and then Merge his hosts, the project might resend his lost tasks, it would save downloading the Wu's again, Claggy |
Gundolf Jahn Send message Joined: 19 Sep 00 Posts: 3184 Credit: 446,358 RAC: 0 |
He could try dropping his old Wu's into his project folder... You cannot drop old WUs into the project folder if you don't have saved the client_state.xml too. And if you've saved it, you don't need to merge, since the host info is saved too. Gruß, Gundolf |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
He could try dropping his old Wu's into his project folder... But if he Merges the two hosts into one, the Wu's from the old host will become part of the new host, Will the Server resend the Wu's from the old host?, or will they be listed as timed out? Claggy |
SAHBster Send message Joined: 6 Jan 03 Posts: 12 Credit: 4,753,184 RAC: 0 |
I too have a variation on this, Black Screen Of Death. Only happens when running GPU w/u's. Comp. switched to never sleep or hibernate. Have been running on this new pc for nearly a year and it has only started happening this last 7 days or so. Card is Nvidia GT220. Using optimised Lunatics client. Have rolled back driver for card but makes no difference. When coming back to pc, even after just 5 - 10 minutes, screen just comes up black, then no signal detected and then just goes black again. Have to hold 'off button' to shut down. Restarts with options for safe mode et. al. But alway starts up fine 'normally'. Device manager tells me graphics cards working fine. Any pointers welcome. Thanks Chaps !! 22/12/2010 21:04:58 Starting BOINC client version 6.10.58 for windows_x86_64 22/12/2010 21:04:58 log flags: file_xfer, sched_ops, task 22/12/2010 21:04:58 Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3 22/12/2010 21:04:58 Data directory: C:\ProgramData\BOINC 22/12/2010 21:04:58 Running under account 22/12/2010 21:04:58 Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz [Family 6 Model 23 Stepping 10] 22/12/2010 21:04:58 Processor: 2.00 MB cache 22/12/2010 21:04:58 Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 syscall nx lm vmx tm2 pbe 22/12/2010 21:04:58 OS: Microsoft Windows 7: Home Premium x64 Edition, (06.01.7600.00) 22/12/2010 21:04:58 Memory: 6.00 GB physical, 12.00 GB virtual 22/12/2010 21:04:58 Disk: 931.51 GB total, 875.74 GB free 22/12/2010 21:04:58 Local time is UTC +0 hours 22/12/2010 21:04:59 NVIDIA GPU 0: GeForce GT 220 (driver version 19107, CUDA version 2030, compute capability 1.2, 1024MB, 115 GFLOPS peak) 22/12/2010 21:04:59 SETI@home Found app_info.xml; using anonymous platform |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
Try disabling all ACPI and power-saving features from BIOS first. Does it re-occur? |
SAHBster Send message Joined: 6 Jan 03 Posts: 12 Credit: 4,753,184 RAC: 0 |
Sorry Olli for being an idiot. Please can you explain to me where I can find the ACPI and BIOS Thank you for your quick response and patience. Ok Olli, will google the ACPI and BIOS stuff. Additionally, just set GPU to crunch whilst PC in use, screen shut down within 30 seconds. Had to reboot to get back on here !! Machine will crunch CPU work fine and can be left for hours, have just reset GPU only to run when machine idle. From that it doesn't look as if it is a power saving problem does it ? |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
So, the clue is somewhere within the seven days you have not been able to run it ok. What happened seven days ago? |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
It still can be about power saving. Try turning off any power saving features from your computer's screen saver. |
SAHBster Send message Joined: 6 Jan 03 Posts: 12 Credit: 4,753,184 RAC: 0 |
Thanks Olli, will try the power saving route first. Then try to go through error logs to see if I can get any clues. Thanks for your assistance, much appreciated. |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
Still, have you got GPU-Z in use?. It can monitor graphic adapter's temperatures. High temps might be the issue. When is the last time you have opened your computer case and vacuumed all the fans and dissipators properly? Excess dust? I do it twice a month. |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
Here is CPU-Z, if you don't already have it. |
skildude Send message Joined: 4 Oct 00 Posts: 9541 Credit: 50,759,529 RAC: 60 |
You may have already mentioned this but you'll need to turn off the screensaver. Just set it to go blank after 20 minutes. You definitely can't run a BOINC/seti screensaver if you are crunching the WU on your GPU In a rich man's house there is no place to spit but his face. Diogenes Of Sinope |
lupo Send message Joined: 29 Aug 10 Posts: 91 Credit: 4,736,407 RAC: 0 |
I did a fresh install of windows 7 and wiped BOINC and reinstalled BOINC and lunatic. Im hoping that the issue has now gone away. I guess Ill know for sure once they start distributing WU. I got 3 tonight which I ripped through in like 30 minute so I'll need something more substantial. IIRC, they are back online tomorrow. For tonight I'll juts run the GPU's doing dnet. That will stress then hard. I'm running a 1200W PSU and I can tell from the APC I ave it plugged into that it;s not drawing more than 700W at any given time so it's not that. Thank for the help thought guys. Appreciate it. |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
If you try merging your two hosts, there's a possibility that the server will resend the Wu's from your old host. Claggy |
SAHBster Send message Joined: 6 Jan 03 Posts: 12 Credit: 4,753,184 RAC: 0 |
Olli, Thanks for the pointer, downloaded GPUZ, running at 64 - 65 C, fan speed 50pct. A curious thing though, my nephew is in local government IT support here in the U.K. He was visiting over the holidays, reseated GPU and memory modules. All appears to be running fine now. Thanks for your patience Olli. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.