Message boards :
ATLAS application :
queue is empty
Message board moderation
Previous · 1 . . . 10 · 11 · 12 · 13
| Author | Message |
|---|---|
|
Send message Joined: 14 Jan 10 Posts: 1427 Credit: 9,510,310 RAC: 2,545   |
It was for him an uploadproblem, not a Taskproblem.If I remember correctly: Even upload is not the problem, but is it a network buffer setting on the server. The buffer seems to be smaller than 2,003,558,304 bytes |
 Send message Joined: 15 Jun 08 Posts: 2549 Credit: 255,283,324 RAC: 57,798   |
It was for him an uploadproblem, not a Taskproblem.If I remember correctly: Right. That's exactly the reason why uploads fail and the tasks get lost. Now, before anybody asks for a greater buffer. This would just postpone the basic problem. @ATLAS submitters The only real solution is to keep the task parameters within the ranges that are known to work fine, especially: - keep uploads and downloads as small as possible - do not exceed 400 events to be processed The latter ensures that tasks can be finished within a work day on average computers. This is important since ATLAS (native) can't be suspended/resumed. |
|
Send message Joined: 2 May 07 Posts: 2245 Credit: 174,006,243 RAC: 8,727
|
Had this night the same problem. File was too big. 2 GByte is the limit. Have no restrictions for Boinc or VDSL. Don't know where this limit can be. The only answer fo me is -native with CentOS. <message> upload failure: <file_xfer_error> <file_name>VAnLDm049N6n9Rq4apOajLDm4fhM0noT9bVorHsSDmvT3KDmJZvwqm_0_r122375125_ATLAS_hits</file_name> <error_code>-131 (file size too big)</error_code> </file_xfer_error> </message> https://lhcathome.cern.ch/lhcathome/result.php?resultid=415212983 |
|
Send message Joined: 22 Mar 17 Posts: 66 Credit: 14,580,818 RAC: 489
|
I had 8x Atlas tasks running (4 threads each) and only 2 completed w/o a file size too big. 15-16 hours on 24 threads wasted. :( |
 Send message Joined: 28 Sep 04 Posts: 735 Credit: 49,844,204 RAC: 35,579
|
The queue is dropping down again. 101 ready to send at the moment. 
|
|
Send message Joined: 18 Dec 15 Posts: 1827 Credit: 119,544,484 RAC: 43,177
|
The queue is dropping down again. 101 ready to send at the moment.down to 26 right now |
|
Send message Joined: 28 Sep 04 Posts: 735 Credit: 49,844,204 RAC: 35,579
|
Down to 172 in the queue....
|
|
Send message Joined: 28 Sep 04 Posts: 735 Credit: 49,844,204 RAC: 35,579
|
And the queue is just about empty....
|
|
Send message Joined: 28 Dec 08 Posts: 341 Credit: 4,896,063 RAC: 2,401
|
And the queue is just about empty.... After that round of bad tasks maybe they are taking a pause (except six track) to examine what went wrong. I had a page full of CMS that went bad and a page of ATLAS that went invalid.This was brought up in another page. |
|
Send message Joined: 18 Dec 15 Posts: 1827 Credit: 119,544,484 RAC: 43,177
|
And the queue is just about empty....well, if the figures from the server status page are correct, 367 volunteers have downloaded 6.370 Atlas tasks, obviously to have enough on stock before the queue dries out again; which happened very quickly. Too bad that all my hosts ran out of Atlas tasks during last night already :-( |
|
Send message Joined: 28 Sep 04 Posts: 735 Credit: 49,844,204 RAC: 35,579
|
The queue is draining again. Down to 900 as ready to send. I hope we get new tasks in the queue before it dries out.
|
©2025 CERN