Message boards :
ATLAS application :
No tasks available
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
| Author | Message |
|---|---|
|
Send message Joined: 13 May 14 Posts: 387 Credit: 15,314,184 RAC: 0   |
Claiming that Canadian beer is better than German beer is a bold statement... but I don't want to start a flame war over beer :) The reason the unsent WU dropped is because we cleaned up around 7000 WU which were stuck there in a kind of zombie state since July, when there was some manual cleanup of the database. The previous times when there were problems you probably noticed that no tasks were being sent even though there were 7000 unsent. In ATLAS we keep a queue of unsent which is around 20% of the running tasks, which is why you see 2000 now and why it varies with the number of running. |
|
Send message Joined: 13 Apr 18 Posts: 443 Credit: 8,438,885 RAC: 0 
|
Thank you, Dave. We are all looking forward to more reliable task distribution. |
|
Send message Joined: 2 May 07 Posts: 2255 Credit: 174,204,943 RAC: 10,166 
|
Atlas show Zero new tasks for the moment! |
|
Send message Joined: 13 May 14 Posts: 387 Credit: 15,314,184 RAC: 0 |
As you can see on the server status page we have a growing backlog of tasks to assimilate. This means that the total number of tasks hit a limit in our ATLAS submission system which stopped it submitting new tasks. I've increased the limit so there are new tasks now but if the assimilator backlog continues to increase we might have to throttle the submission. The LHC@Home admins are working on how to improve the performance of the assimilator. |
|
Send message Joined: 2 May 07 Posts: 2255 Credit: 174,204,943 RAC: 10,166
|
Thank you David, Atlas-Dashboard is showing 18k Slots in use! |
|
Send message Joined: 2 May 07 Posts: 2255 Credit: 174,204,943 RAC: 10,166
|
Atlas show only 240 unsend Tasks on the Serverpage. Normal are 2k or 3k? |
 Send message Joined: 30 Aug 14 Posts: 145 Credit: 10,847,070 RAC: 0
|
Atlas seems to be out of work. No unsent tasks on serverpage. Unfortunately CERN guys are on holidays till monday... Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us |
|
Send message Joined: 18 Dec 15 Posts: 1838 Credit: 121,595,592 RAC: 88,970
|
Atlas seems to be out of work.the Server Status Page shows about 2700 unsent tasks at this point. |
|
Send message Joined: 30 Aug 14 Posts: 145 Credit: 10,847,070 RAC: 0
|
Yes, new tasks are available now. Seemed to be only a temporary problem between 02.00 and 03.00 CET. Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us |
 Send message Joined: 28 Sep 04 Posts: 739 Credit: 50,532,098 RAC: 33,576 
|
The ready to send task buffer has almost dried out (17 and >26000 in progress). Probably because the number of currently running slots of jobs has risen close to 40000 from the usual 10000...15000. The servers might be having a hard time keeping up with the demand. 
|
|
Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0
|
I have received a "No tasks are available for ATLAS Simulation" message since 03:25 UTC. I will be out of work in six hours (native ATLAS). EDIT: A manual request just picked up two work units. |
|
Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0
|
The servers might be having a hard time keeping up with the demand. They are now down to zero and I am not getting any more. It looks like last year; deja vu all over again. Is there a plan? |
|
Send message Joined: 13 May 14 Posts: 387 Credit: 15,314,184 RAC: 0 |
It's basically what Harri said, the number of ATLAS WU more than doubled overnight and it revealed some weak points in the infrastructure. We're working on fixing these and keeping the queue full. |
|
Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0
|
OK, that is a better problem than no work units at all. I can keep a machine active, more or less. |
|
Send message Joined: 28 Sep 04 Posts: 739 Credit: 50,532,098 RAC: 33,576
|
We seem to have run out of tasks to send. Two of my tasks were cancelled by server during the night. The graphics at the bottom of this page http://lhcathome.web.cern.ch/projects/atlas indicates that tasks are on their way but not yet here. Houston, do we have a problem?
|
|
Send message Joined: 2 May 07 Posts: 2255 Credit: 174,204,943 RAC: 10,166
|
+1 |
|
Send message Joined: 13 Apr 18 Posts: 443 Credit: 8,438,885 RAC: 0
|
In the past 2 days I've had ~10 native ATLAS canceled by the server on 3 different hosts. None were anywhere near expiry date. Most had run for several hours. This is definitely not a case of caching too many tasks. My watchdog script showed 5 were processing events. I didn't check the other 5. |
|
Send message Joined: 13 May 14 Posts: 387 Credit: 15,314,184 RAC: 0 |
We had some problems overnight with the services submitting tasks to LHC@home. More WU should be coming in now, apologies for the disruption. |
|
Send message Joined: 22 Mar 17 Posts: 66 Credit: 20,452,173 RAC: 289,694
|
In the past 2 days I've had ~10 native ATLAS canceled by the server on 3 different hosts. None were anywhere near expiry date. Most had run for several hours. This is definitely not a case of caching too many tasks. My watchdog script showed 5 were processing events. I didn't check the other 5. I had a couple as well. |
|
Send message Joined: 27 Aug 15 Posts: 27 Credit: 15,069,153 RAC: 52,179
|
Atlas seems to be out of work |
©2025 CERN