Message boards :
Number crunching :
Since the last outage...
Message board moderation
Author | Message |
---|---|
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
Here we go again. Every time there is an outage...a new glitch seems to pop up. One of my crunchers (with GPU) is no longer getting any GPU workloads. Any thoughts? |
Mr. Kevvy Send message Joined: 15 May 99 Posts: 3806 Credit: 1,114,826,392 RAC: 3,319 |
Does the BOINC worklog say it's requesting work for both CPU and GPU? If so, GPU work (of course) tends to get assigned faster as GPUs are more powerful so clients pick up more units of it. Strangely after the outage I had one box that wouldn't get any CPU units for over a day. Had to pick up some Einstein work to keep it busy. |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
I see nothing that would indicate whether CPU or GPU workloads are being requested. 5/24/2012 11:20:31 AM | | [work_fetch] work fetch start 5/24/2012 11:20:31 AM | | [work_fetch] ------- start work fetch state ------- 5/24/2012 11:20:31 AM | | [work_fetch] target work buffer: 180.00 + 864000.00 sec 5/24/2012 11:20:31 AM | SETI@home | [work_fetch] REC 72901.511 priority -0.110393 (project backoff 234.64) 5/24/2012 11:20:31 AM | | [work_fetch] CPU: shortfall 191943533.83 nidle 0.00 saturated 35259.78 busy 0.00 5/24/2012 11:20:31 AM | SETI@home | [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) 5/24/2012 11:20:31 AM | | [work_fetch] ------- end work fetch state ------- 5/24/2012 11:20:31 AM | | [work_fetch] No project chosen for work fetch 5/24/2012 11:21:32 AM | | [work_fetch] work fetch start 5/24/2012 11:21:32 AM | | [work_fetch] ------- start work fetch state ------- 5/24/2012 11:21:32 AM | | [work_fetch] target work buffer: 180.00 + 864000.00 sec 5/24/2012 11:21:32 AM | SETI@home | [work_fetch] REC 72903.376 priority -0.110283 (project backoff 173.71) 5/24/2012 11:21:32 AM | | [work_fetch] CPU: shortfall 191973119.99 nidle 0.00 saturated 35126.69 busy 0.00 5/24/2012 11:21:32 AM | SETI@home | [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 0.00, inc 0.00) 5/24/2012 11:21:32 AM | | [work_fetch] ------- end work fetch state ------- 5/24/2012 11:21:32 AM | | [work_fetch] No project chosen for work fetch |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
Nice, the debug flags I had set seem to have been the problem. Which is pretty lame! I turned them off, restarted the project and voila! Suddenly I'm getting GPU tasks again. Grrr, lame code... 5/24/2012 11:35:40 AM | SETI@home | update requested by user 5/24/2012 11:35:44 AM | SETI@home | Sending scheduler request: Requested by user. 5/24/2012 11:35:44 AM | SETI@home | Requesting new tasks for CPU and NVIDIA 5/24/2012 11:35:48 AM | SETI@home | Scheduler request completed: got 20 new tasks 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.31_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.190_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.41_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.194_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.196_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.198_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 16my10aa.19157.6388.4.10.112_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.51_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 16my10aa.19157.6388.4.10.115_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.168_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 16my10aa.19157.6388.4.10.78_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.172_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.6_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 16my10aa.19157.6388.4.10.83_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 16my10aa.19157.6388.4.10.85_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 09ja11aa.19742.4727.6.10.179_0 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.16_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.19_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 05ap10ah.8419.80701.15.10.21_1 5/24/2012 11:35:48 AM | SETI@home | Resent lost task 16my10aa.19157.6388.4.10.94_0 5/24/2012 11:35:50 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.31 5/24/2012 11:35:50 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.190 5/24/2012 11:35:53 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.31 5/24/2012 11:35:53 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.190 5/24/2012 11:35:53 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.41 5/24/2012 11:35:53 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.194 5/24/2012 11:35:53 AM | SETI@home | Starting task 09ja11aa.19742.4727.6.10.190_1 using setiathome_enhanced version 610 (cuda_fermi) in slot 82 5/24/2012 11:35:58 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.41 5/24/2012 11:35:58 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.194 5/24/2012 11:35:58 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.196 5/24/2012 11:35:58 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.198 5/24/2012 11:36:01 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.196 5/24/2012 11:36:01 AM | SETI@home | Started download of 16my10aa.19157.6388.4.10.112 5/24/2012 11:36:03 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.198 5/24/2012 11:36:03 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.51 5/24/2012 11:36:04 AM | SETI@home | Finished download of 16my10aa.19157.6388.4.10.112 5/24/2012 11:36:04 AM | SETI@home | Started download of 16my10aa.19157.6388.4.10.115 5/24/2012 11:36:05 AM | SETI@home | Finished download of 16my10aa.19157.6388.4.10.115 5/24/2012 11:36:05 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.168 5/24/2012 11:36:06 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.51 5/24/2012 11:36:06 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.168 5/24/2012 11:36:06 AM | SETI@home | Started download of 16my10aa.19157.6388.4.10.78 5/24/2012 11:36:06 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.172 5/24/2012 11:36:11 AM | SETI@home | Finished download of 16my10aa.19157.6388.4.10.78 5/24/2012 11:36:11 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.172 5/24/2012 11:36:11 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.6 5/24/2012 11:36:11 AM | SETI@home | Started download of 16my10aa.19157.6388.4.10.83 5/24/2012 11:36:26 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.6 5/24/2012 11:36:26 AM | SETI@home | Finished download of 16my10aa.19157.6388.4.10.83 5/24/2012 11:36:26 AM | SETI@home | Started download of 16my10aa.19157.6388.4.10.85 5/24/2012 11:36:26 AM | SETI@home | Started download of 09ja11aa.19742.4727.6.10.179 5/24/2012 11:36:28 AM | SETI@home | Finished download of 16my10aa.19157.6388.4.10.85 5/24/2012 11:36:28 AM | SETI@home | Finished download of 09ja11aa.19742.4727.6.10.179 5/24/2012 11:36:28 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.16 5/24/2012 11:36:28 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.19 5/24/2012 11:36:30 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.16 5/24/2012 11:36:30 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.19 5/24/2012 11:36:30 AM | SETI@home | Started download of 05ap10ah.8419.80701.15.10.21 5/24/2012 11:36:30 AM | SETI@home | Started download of 16my10aa.19157.6388.4.10.94 5/24/2012 11:36:31 AM | SETI@home | Finished download of 05ap10ah.8419.80701.15.10.21 5/24/2012 11:36:31 AM | SETI@home | Finished download of 16my10aa.19157.6388.4.10.94 5/24/2012 11:37:28 AM | SETI@home | Restarting task 05ap10ah.32185.72930.11.10.202_1 using setiathome_enhanced version 603 in slot 98 5/24/2012 11:38:09 AM | SETI@home | Restarting task 09ja11aa.19742.1455.6.10.237_1 using setiathome_enhanced version 603 in slot 122 |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
5/24/2012 11:21:32 AM | | [work_fetch] target work buffer: 180.00 + 864000.00 sec All your "crunchers (with GPU)" have BOINC 7.0.25 The "target work buffer: 180.00 + 864000.00 sec" is not good for BOINC 7.x (= 0.002 + 10 days) Set it to e.g. 7 + 3 days ("debug flags" have nothing to do with your low work fetch, the 0.002 days is the problem) Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Jakesnake5 Send message Joined: 14 May 99 Posts: 65 Credit: 18,370,396 RAC: 0 |
I have a system with an older Nvidia (8600). It used the 6.08 cuda app. It hasn't been getting ANY GPU work since I updated to the newest BOINC. Is the 6.08 no longer supported? I'd hate to have to buy a new vid card just to use it with Seti. |
Gatekeeper Send message Joined: 14 Jul 04 Posts: 887 Credit: 176,479,616 RAC: 0 |
I have a system with an older Nvidia (8600). 6.08 is still good. I'd check the same settings as referenced above for the OP, and if you still aren't getting GPU work, change your default settings to accept only GPU tasks (no CPU) until you get a good workload, then switch back to accepting both. |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
I have a system with an older Nvidia (8600). Make sure you swap your cache settings around for Boinc 7.0.x as the scheduler has changed, use something like: Maintain enough tasks to keep busy for at least 4 days (max 10 days). ... and up to an additional 0.01 days Claggy |
Mr. GoodWrench Send message Joined: 26 Jun 99 Posts: 19 Credit: 8,937,626 RAC: 0 |
I seem to be having a different problem than others in this thread. Hoping someone can lead me in the right direction. For several weeks prior to the outage I was only able to get CUDA work units. Now, after the last outage I'm only getting CPU work during requests. Here's the startup info from BOINC: 5/28/2012 11:49:53 AM | | No config file found - using defaults 5/28/2012 11:49:53 AM | | Starting BOINC client version 7.0.25 for windows_x86_64 5/28/2012 11:49:53 AM | | log flags: file_xfer, sched_ops, task 5/28/2012 11:49:53 AM | | Libraries: libcurl/7.21.6 OpenSSL/1.0.0d zlib/1.2.5 5/28/2012 11:49:53 AM | | Data directory: C:\ProgramData\BOINC 5/28/2012 11:49:53 AM | | Processor: 2 GenuineIntel Intel(R) Core(TM)2 Duo CPU T8100 @ 2.10GHz [Family 6 Model 23 Stepping 6] 5/28/2012 11:49:53 AM | | Processor: 3.00 MB cache 5/28/2012 11:49:53 AM | | 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 5/28/2012 11:49:53 AM | | OS: Microsoft Windows Vista: Home Premium x64 Edition, Service Pack 2, (06.00.6002.00) 5/28/2012 11:49:53 AM | | Memory: 4.00 GB physical, 8.18 GB virtual 5/28/2012 11:49:53 AM | | Disk: 453.20 GB total, 282.67 GB free 5/28/2012 11:49:53 AM | | Local time is UTC -4 hours 5/28/2012 11:49:53 AM | | NVIDIA GPU 0: GeForce 8600M GT (driver version 301.42, CUDA version 4.20, compute capability 1.1, 256MB, 176MB available, 91 GFLOPS peak) 5/28/2012 11:49:53 AM | | OpenCL: NVIDIA GPU 0: GeForce 8600M GT (driver version 301.42, device version OpenCL 1.0 CUDA, 256MB, 176MB available) I just d/l the most current driver version release on 5/22/12. Here's a sample of what the log shows for the past few weeks: 5/28/2012 11:49:54 AM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 11:49:55 AM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 11:49:55 AM | SETI@home | No tasks sent 5/28/2012 11:49:55 AM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 12:02:05 PM | SETI@home | Sending scheduler request: To fetch work. 5/28/2012 12:02:05 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 12:02:07 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 12:02:07 PM | SETI@home | No tasks sent 5/28/2012 12:02:07 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 12:22:47 PM | SETI@home | Sending scheduler request: To fetch work. 5/28/2012 12:22:47 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 12:22:49 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 12:22:49 PM | SETI@home | No tasks sent 5/28/2012 12:22:49 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 12:35:55 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 12:35:57 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 12:35:57 PM | SETI@home | No tasks sent 5/28/2012 12:35:57 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 1:15:55 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 1:15:57 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 1:15:57 PM | SETI@home | Project has no tasks available 5/28/2012 1:25:02 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 1:25:04 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 1:25:04 PM | SETI@home | No tasks sent 5/28/2012 1:25:04 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 1:36:12 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 1:36:15 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 1:36:15 PM | SETI@home | Project has no tasks available 5/28/2012 2:19:29 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 2:19:31 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 2:19:31 PM | SETI@home | No tasks sent 5/28/2012 2:19:31 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 4:02:22 PM | SETI@home | Reporting 1 completed tasks, requesting new tasks for NVIDIA 5/28/2012 4:02:24 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 4:02:24 PM | SETI@home | No tasks sent 5/28/2012 4:02:24 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 10:05:01 PM | SETI@home | Sending scheduler request: Requested by user. 5/28/2012 10:05:01 PM | SETI@home | Reporting 1 completed tasks, requesting new tasks for NVIDIA 5/28/2012 10:05:03 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 10:05:03 PM | SETI@home | No tasks sent 5/28/2012 10:05:03 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 10:27:18 PM | SETI@home | Sending scheduler request: To fetch work. 5/28/2012 10:27:18 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 10:27:22 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 10:27:22 PM | SETI@home | No tasks sent 5/28/2012 10:27:22 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 10:33:43 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 10:33:45 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 10:33:45 PM | SETI@home | No tasks sent 5/28/2012 10:33:45 PM | SETI@home | No tasks are available for SETI@home Enhanced 5/28/2012 10:44:51 PM | SETI@home | Requesting new tasks for NVIDIA 5/28/2012 10:44:54 PM | SETI@home | Scheduler request completed: got 0 new tasks 5/28/2012 10:44:54 PM | SETI@home | No tasks sent 5/28/2012 10:44:54 PM | SETI@home | No tasks are available for SETI@home Enhanced I must be missing something simple. Can anyone help? Or are CUDA units in high demand since the outage? TIA |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
I must be missing something simple. Yes, you are missing that there is no such thing as "CUDA units", "GPU work units" (nor "CPU work units"). All the tasks at the server are the same. The only "exception" are the .vlar tasks which are artificially intentionally not send to NVIDIA GPUs (to avoid lags) So even if you were requesting new tasks for CPU at the same very moment the result would be the same: "No tasks are available for SETI@home Enhanced" But I can't imagine this outcome to be "for the past few weeks". Very probably the problem is this: 5/28/2012 11:49:53 AM | | OpenCL: NVIDIA GPU 0: GeForce 8600M GT (driver version 301.42, device version OpenCL 1.0 CUDA, 256MB, 176MB available) I will let people which actually use CUDA to suggest how to free some VRAM (e.g. disable Aero ('glass' effect)) (you need maybe 220MB VRAM available) Very probably to achieve this you'll need to return to older driver (190.x 191.x) Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Mr. GoodWrench Send message Joined: 26 Jun 99 Posts: 19 Credit: 8,937,626 RAC: 0 |
Thanks for the suggestion. I may try an older driver once things get settled again. Hoping it may fix itself. |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
Grrr, now after this power outtage, I am getting ONLY GPU workloads. Who writes this code??? Why do I have to keep babying this tool to get it to work consistently? |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
And now this: 6/1/2012 9:10:28 AM | SETI@home | update requested by user 6/1/2012 9:10:29 AM | SETI@home | Sending scheduler request: Requested by user. 6/1/2012 9:10:29 AM | SETI@home | Reporting 1 completed task, requesting new tasks for CPU and NVIDIA 6/1/2012 9:10:37 AM | SETI@home | Scheduler request completed: got 0 new tasks 6/1/2012 9:10:37 AM | SETI@home | Project has no tasks available /me shakes his head in disappointment 32 CPU cores and 300+ GPU cores on this machine and all I can do now is one GPU workload at a time...sad. |
Mike Send message Joined: 17 Feb 01 Posts: 34380 Credit: 79,922,639 RAC: 80 |
You are running Boinc 7.0.25 on this machine. Upgrade to 7.0.28. With each crime and every kindness we birth our future. |
Ex: "Socialist" Send message Joined: 12 Mar 12 Posts: 3433 Credit: 2,616,158 RAC: 2 |
And were back- to getting high-priorities by the buttload... #resist |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
??? 7.0.28? Look here: http://boinc.berkeley.edu/download.php 7.0.25 is the latest...unless only the cool kids are privy to newer versions... |
Ex: "Socialist" Send message Joined: 12 Mar 12 Posts: 3433 Credit: 2,616,158 RAC: 2 |
7.0.25 is the newest STABLE. 7.0.28 is development, and is preferred for oh so many reasons I wont be covering here. #resist |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
Sigh. Where can it be downloaded from? What is the secret handshake? Who do I have to kiss? |
Ex: "Socialist" Send message Joined: 12 Mar 12 Posts: 3433 Credit: 2,616,158 RAC: 2 |
Sigh. The same page you were looking at. Just needed to click "All versions" to get here: http://boinc.berkeley.edu/download_all.php ;-) #resist |
gcpeters Send message Joined: 20 May 99 Posts: 67 Credit: 109,352,237 RAC: 1 |
Thanks brother! |
©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.