Can't get WUs for NVidia

Message boards : Graphics cards (GPUs) : Can't get WUs for NVidia

Author	Message
Beyond Send message Joined: 23 Nov 08 Posts: 1112 Credit: 6,162,416,256 RAC: 0 Level Scientific publications	Message 8944 - Posted: 26 Apr 2009 \| 6:18:22 UTC
	This has been going on for a few hours: Host Project Date Message GPUGRID 4/26/2009 1:13:06 AM Sending scheduler request: To fetch work. GPUGRID 4/26/2009 1:13:06 AM Requesting new tasks GPUGRID 4/26/2009 1:13:11 AM Scheduler request completed: got 0 new tasks GPUGRID 4/26/2009 1:13:11 AM Message from server: No work sent GPUGRID 4/26/2009 1:13:11 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. GPUGRID 4/26/2009 1:13:11 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested GPUGRID 4/26/2009 1:13:47 AM Sending scheduler request: To fetch work. GPUGRID 4/26/2009 1:13:47 AM Requesting new tasks GPUGRID 4/26/2009 1:13:52 AM Scheduler request completed: got 0 new tasks GPUGRID 4/26/2009 1:13:52 AM Message from server: No work sent GPUGRID 4/26/2009 1:13:52 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. GPUGRID 4/26/2009 1:13:52 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested GPUGRID 4/26/2009 1:15:27 AM Sending scheduler request: To fetch work. GPUGRID 4/26/2009 1:15:27 AM Requesting new tasks GPUGRID 4/26/2009 1:15:32 AM Scheduler request completed: got 0 new tasks GPUGRID 4/26/2009 1:15:32 AM Message from server: No work sent GPUGRID 4/26/2009 1:15:32 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. GPUGRID 4/26/2009 1:15:32 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested Tried v6.6.23 then v6.6.24, no difference.
	ID: 8944 \| Rating: 0 \| rate: / Reply Quote

MarkJ Volunteer moderator Volunteer tester Send message Joined: 24 Dec 08 Posts: 738 Credit: 200,909,904 RAC: 0 Level Scientific publications	Message 8946 - Posted: 26 Apr 2009 \| 7:04:19 UTC - in response to Message 8944. Last modified: 26 Apr 2009 \| 7:06:22 UTC
	This has been going on for a few hours: Host Project Date Message GPUGRID 4/26/2009 1:13:06 AM Sending scheduler request: To fetch work. GPUGRID 4/26/2009 1:13:06 AM Requesting new tasks GPUGRID 4/26/2009 1:13:11 AM Scheduler request completed: got 0 new tasks GPUGRID 4/26/2009 1:13:11 AM Message from server: No work sent GPUGRID 4/26/2009 1:13:11 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. GPUGRID 4/26/2009 1:13:11 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested GPUGRID 4/26/2009 1:13:47 AM Sending scheduler request: To fetch work. GPUGRID 4/26/2009 1:13:47 AM Requesting new tasks GPUGRID 4/26/2009 1:13:52 AM Scheduler request completed: got 0 new tasks GPUGRID 4/26/2009 1:13:52 AM Message from server: No work sent GPUGRID 4/26/2009 1:13:52 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. GPUGRID 4/26/2009 1:13:52 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested GPUGRID 4/26/2009 1:15:27 AM Sending scheduler request: To fetch work. GPUGRID 4/26/2009 1:15:27 AM Requesting new tasks GPUGRID 4/26/2009 1:15:32 AM Scheduler request completed: got 0 new tasks GPUGRID 4/26/2009 1:15:32 AM Message from server: No work sent GPUGRID 4/26/2009 1:15:32 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. GPUGRID 4/26/2009 1:15:32 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested Tried v6.6.23 then v6.6.24, no difference. Reset your debts via the <zero_debts> flag. I'd suggest you also turn on the <sched_op_debug> flag so you can see what its asking for. Basically its asking for cpu work and getting told there isn't any. This is correct. It needs to ask for cuda work. ____________ BOINC blog
	ID: 8946 \| Rating: 0 \| rate: / Reply Quote

Paul D. Buck Send message Joined: 9 Jun 08 Posts: 1050 Credit: 37,321,185 RAC: 0 Level Scientific publications	Message 8964 - Posted: 26 Apr 2009 \| 21:23:58 UTC - in response to Message 8946.
	Reset your debts via the <zero_debts> flag. I'd suggest you also turn on the <sched_op_debug> flag so you can see what its asking for. Basically its asking for cpu work and getting told there isn't any. This is correct. It needs to ask for cuda work. work fetch debug flag ...
	ID: 8964 \| Rating: 0 \| rate: / Reply Quote

Beyond Send message Joined: 23 Nov 08 Posts: 1112 Credit: 6,162,416,256 RAC: 0 Level Scientific publications	Message 8967 - Posted: 26 Apr 2009 \| 22:51:46 UTC - in response to Message 8964. Last modified: 26 Apr 2009 \| 22:52:31 UTC
	A few hours later it started getting work again. During this time the GPU WUs ready to send was stuck at 53, then 52 so it was most likely a server problem. Ran a few Seti WUs in the meantime.
	ID: 8967 \| Rating: 0 \| rate: / Reply Quote

MarkJ Volunteer moderator Volunteer tester Send message Joined: 24 Dec 08 Posts: 738 Credit: 200,909,904 RAC: 0 Level Scientific publications	Message 8979 - Posted: 27 Apr 2009 \| 9:23:35 UTC - in response to Message 8964.
	Reset your debts via the <zero_debts> flag. I'd suggest you also turn on the <sched_op_debug> flag so you can see what its asking for. Basically its asking for cpu work and getting told there isn't any. This is correct. It needs to ask for cuda work. work fetch debug flag ... No sched_op_debug displays the scheduler request/result. As you can see it was silly enough to ask for cpu work from GPUgrid. Nothing to worry about it will backoff the resource and should try a cuda request sooner or later. Host Project Date Message Anakin GPUGRID 27/04/2009 1:00:47 AM [sched_op_debug] Starting scheduler request Anakin GPUGRID 27/04/2009 1:00:47 AM Sending scheduler request: To fetch work. Anakin GPUGRID 27/04/2009 1:00:47 AM Requesting new tasks Anakin GPUGRID 27/04/2009 1:00:47 AM [sched_op_debug] CPU work request: 22844.74 seconds; 0 idle CPUs Anakin GPUGRID 27/04/2009 1:00:47 AM [sched_op_debug] CUDA work request: 0.00 seconds; 0 idle GPUs Anakin GPUGRID 27/04/2009 1:00:53 AM Scheduler request completed: got 0 new tasks Anakin GPUGRID 27/04/2009 1:00:53 AM [sched_op_debug] Server version 607 Anakin GPUGRID 27/04/2009 1:00:53 AM Message from server: No work sent Anakin GPUGRID 27/04/2009 1:00:53 AM Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer. Anakin GPUGRID 27/04/2009 1:00:53 AM Message from server: CUDA app exists for Full-atom molecular dynamics but no CUDA work requested Anakin GPUGRID 27/04/2009 1:00:53 AM Project requested delay of 31 seconds Anakin GPUGRID 27/04/2009 1:00:53 AM [sched_op_debug] Deferring communication for 31 sec Anakin GPUGRID 27/04/2009 1:00:53 AM [sched_op_debug] Reason: requested by project ____________ BOINC blog
	ID: 8979 \| Rating: 0 \| rate: / Reply Quote

Paul D. Buck Send message Joined: 9 Jun 08 Posts: 1050 Credit: 37,321,185 RAC: 0 Level Scientific publications	Message 8981 - Posted: 27 Apr 2009 \| 11:07:43 UTC
	Ah, I mis-read the note ... ok... i gotta play with this flag for a bit ...
	ID: 8981 \| Rating: 0 \| rate: / Reply Quote

(_KoDAk_) Send message Joined: 18 Oct 08 Posts: 43 Credit: 6,924,807 RAC: 0 Level Scientific publications	Message 8985 - Posted: 27 Apr 2009 \| 12:06:01 UTC
	27.04.2009 15:04:57 GPUGRID [error] File m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2 has wrong size: expected 915316, got 0 27.04.2009 15:04:57 GPUGRID Started download of m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2 27.04.2009 15:04:59 GPUGRID Temporarily failed download of m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2: HTTP error 27.04.2009 15:04:59 GPUGRID Backing off 39 min 59 sec on download of m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2
	ID: 8985 \| Rating: 0 \| rate: / Reply Quote

computerguy09 Send message Joined: 20 Aug 08 Posts: 10 Credit: 15,789,768 RAC: 0 Level Scientific publications	Message 8990 - Posted: 27 Apr 2009 \| 13:06:38 UTC - in response to Message 8985.
	27.04.2009 15:04:57 GPUGRID [error] File m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2 has wrong size: expected 915316, got 0 27.04.2009 15:04:57 GPUGRID Started download of m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2 27.04.2009 15:04:59 GPUGRID Temporarily failed download of m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2: HTTP error 27.04.2009 15:04:59 GPUGRID Backing off 39 min 59 sec on download of m200000-GIANNI_pYIpYVk12204-6-m200000-GIANNI_pYIpYVk12204-5-10-RND3565_2 If this problem continues for more than 2-3 hours, and your Internet access is OK, you might want to abort the WU and the transfer and get another WU. I had one of my hosts get stuck with one of these types of errors. Even though I was getting WU on other hosts, there seem to be some stuck ones lately. Mark
	ID: 8990 \| Rating: 0 \| rate: / Reply Quote

Post to thread

Message boards : Graphics cards (GPUs) : Can't get WUs for NVidia

	About	Science	Volunteers	Performance	Forum	Join us	Donate