Message boards :
Number crunching :
No tasks sent
Message board moderation
Author | Message |
---|---|
Send message Joined: 17 Mar 15 Posts: 10 Credit: 45,563,435 RAC: 0 ![]() |
Hi Eric The Server Status page shows all green, but I'm out of work and am receiving none. Is it just me or is there a problem on your end? Tks ! |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
That could be an intermittent problem we have with the feeder going down every now and then. It automatically starts up again within 5 minutes, so next time you see that, wait a few minutes and then do an update and see if that works. |
Send message Joined: 17 Mar 15 Posts: 10 Credit: 45,563,435 RAC: 0 ![]() |
That could be an intermittent problem we have with the feeder going down every now and then. It automatically starts up again within 5 minutes, so next time you see that, wait a few minutes and then do an update and see if that works. Hi again, I read that elsewhere and so have been retrying all morning on six machines. None of them is getting any work at all. ???? |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
That is indeed strange. I've been able to download work on my machines. Does the event log give any hints as to the problem? You have a lot of credit, so you must have been connecting at some point in the past. How long has it been since you've been able to download work? |
Send message Joined: 17 Mar 15 Posts: 10 Credit: 45,563,435 RAC: 0 ![]() |
It stopped downloading new work last night sometime. (EDT) The event log really says nothing, just as follows: 3/2/2018 3:06:13 PM | NumberFields@home | [sched_op] Starting scheduler request 3/2/2018 3:06:13 PM | NumberFields@home | Sending scheduler request: To fetch work. 3/2/2018 3:06:13 PM | NumberFields@home | Requesting new tasks for CPU 3/2/2018 3:06:13 PM | NumberFields@home | [sched_op] CPU work request: 13824.00 seconds; 8.00 devices 3/2/2018 3:06:13 PM | NumberFields@home | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 3/2/2018 3:06:19 PM | NumberFields@home | Scheduler request completed: got 0 new tasks 3/2/2018 3:06:19 PM | NumberFields@home | [sched_op] Server version 707 3/2/2018 3:06:19 PM | NumberFields@home | No tasks sent 3/2/2018 3:06:19 PM | NumberFields@home | Project requested delay of 182 seconds 3/2/2018 3:06:19 PM | NumberFields@home | [sched_op] Deferring communication for 00:03:01 3/2/2018 3:06:19 PM | NumberFields@home | [sched_op] Reason: requested by project I'll try to remove the project and re-add it. . . . Let you know. Nope, that didn't work either. I'll have to work on this later, I just lost power for a couple of minutes due to this Nor'Easter the east coast is receiving with very high winds. No damage, my machine and router are on an UPS, but not a good time to work on a problem. "Talk" at you later. |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
And in the scheduler logs I see this: 2018-03-02 06:36:08.6710 [PID=21509] HOST::parse(): unrecognized: n_usable_coprocs 2018-03-02 06:36:08.6711 [PID=21509] HOST::parse(): unrecognized: 1 2018-03-02 06:36:08.6712 [PID=21509] HOST::parse(): unrecognized: /n_usable_coprocs 2018-03-02 06:36:08.6794 [PID=21509] Request: [USER#66257] [HOST#76451] [IP 71.225.7.31] client 7.8.3 2018-03-02 06:36:08.7180 [PID=21509] Sending reply to [HOST#76451]: 0 results, delay req 181.80 2018-03-02 06:36:08.7181 [PID=21509] Scheduler ran 0.047 seconds I see other users with this same error: the scheduler doesn't recognize "n_usable_coprocs". No results get sent and it requests 182 sec delay. What you all have in common is client version 7.8.x. Did you recently upgrade your client? I wonder if I need to upgrade the server version to make it compatible with the more recent clients??? |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
So you can ignore my last question. I went back further in the logs and I see on March 1st you were getting work with client version 7.8.3 and with the same HOST parse error. So it's still a mystery why the scheduler won't give you work. And in the scheduler logs I see this: |
Send message Joined: 17 Mar 15 Posts: 10 Credit: 45,563,435 RAC: 0 ![]() |
Just a quick note. I've been without power for over 24 hours. It just came back on but I'm not going to play with this until it's been back for a day or two, no trust. I shall return. And, thanks for looking into it. |
Send message Joined: 17 Mar 15 Posts: 10 Credit: 45,563,435 RAC: 0 ![]() |
OK, got it. I'm a dummy: The only change I had made to my setup (once I stopped to think of it) was changing the venue from Default to Home. Unfortunately the Home venue had never been setup to choose which type of tasks to ask for, so the server did what it was asked to do and sent nothing. Changed that to (all applications) and am receiving work now. All is well. Best Regards! |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
Ok, good to hear it's working! OK, got it. |
Send message Joined: 8 Apr 17 Posts: 2 Credit: 65,244,792 RAC: 58,169 ![]() ![]() ![]() |
A few days ago, I also had issues receiving new WU. Today I am receiving new WU but 198 out of 200 WU ran for 10 seconds or less. Are there something wrong with the WU? https://numberfields.asu.edu/NumberFields/result.php?resultid=24615079 |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
A few days ago, I also had issues receiving new WU. Today I am receiving new WU but 198 out of 200 WU ran for 10 seconds or less. Are there something wrong with the WU? I reran several of your WUs and verified they are quickies, so no worries. We have seen this before with the highest tiers of previous searches. Last week I tested a batch of 500 of these new 13x271 WUs and they were all between 20000 and 30000 seconds, so I thought this was a good sign there would only be a small number of the quick ones. But it looks like I was wrong from what you've reported. We will just have to wait until the batch finishes to see the final distribution of times. |
![]() Send message Joined: 30 Apr 18 Posts: 4 Credit: 972,300 RAC: 0 ![]() |
Seem to be a problem getting work again .... Getting the same "error" message across all my PC's. Server Status shows all GREEN and there seems to be work to be sent out: Server software version: 24527 / 3 Aug 2018, 2:30:01 UTC Get Decic Fields 41,784 6,464 3.74 (0.00 - 12.23) 192 Minimum Discriminant Septics 52,570 43,935 4.35 (0.00 - 15.20) 292 From the event log: 8/2/2018 21:18:06 | NumberFields@home | Requesting new tasks for CPU and NVIDIA GPU ![]() ![]() |
![]() Send message Joined: 8 Jul 11 Posts: 1355 Credit: 578,526,012 RAC: 795,455 ![]() ![]() ![]() |
So the error message infers that the database was down. I am able to connect to it now and all seems to be fine. Give it another try and see what happens. I have noticed there are several times throughout the day when the database goes down, and it seems to happen at about the same time every day. I looked into this a few days ago and it reminds me of a denial of service attack - a huge number of database queries at the same time which lock up the database. I haven't looked much further into this, but it could be one of the external aggregators collecting stats. I also thought it might be spammers trying to create user accounts, but the creation times of the new spammer accounts seem to be uncorrelated with when the db gets locked up, so I think I can rule that out. Anyways, to make a long story short, whenever this happens try again after 5 to 10 minutes. |
![]() Send message Joined: 30 Apr 18 Posts: 4 Credit: 972,300 RAC: 0 ![]() |
Thank you for the reply and taking a quick look at things. All seems to be OK at the moment - my PC's have been able to download new work since posting yesterday. ![]() ![]() |