Draining ATLAS tasks ahead of storage migration

Author	Message
maeax Send message Joined: 2 May 07 Posts: 2242 Credit: 173,899,983 RAC: 2,804	Message 34765 - Posted: 27 Mar 2018, 16:33:23 UTC - in response to Message 34763. What I still don't understand though: the number of "unsent" tasks in the Project Status Page is going up (it's 245 right now, in contrast to 198 as 1 hour ago) - and still I cannot download any. BOINC keeps saying "no tasks are available for ATLAS Simulation" - something seems to go wrong :-( native Linux-App got WU. So this is correct from the Server-stats. ID: 34765 · Reply Quote

Erich56 Send message Joined: 18 Dec 15 Posts: 1811 Credit: 118,317,329 RAC: 27,294	Message 34766 - Posted: 27 Mar 2018, 16:54:22 UTC It's really too bad that https://lhcathome.cern.ch/ATLAS/atlas_job.php does no longer work :-((( ID: 34766 · Reply Quote

David Cameron Project administrator Project developer Project scientist Send message Joined: 13 May 14 Posts: 387 Credit: 15,314,184 RAC: 0	Message 34768 - Posted: 27 Mar 2018, 17:29:42 UTC - in response to Message 34766. It's really too bad that https://lhcathome.cern.ch/ATLAS/atlas_job.php does no longer work :-((( Is this the page you are looking for? http://atlasathome.cern.ch/atlas_job.php ID: 34768 · Reply Quote

maeax Send message Joined: 2 May 07 Posts: 2242 Credit: 173,899,983 RAC: 2,804	Message 34769 - Posted: 27 Mar 2018, 17:29:50 UTC - in response to Message 34745. This is the page for now: http://lhcathome.web.cern.ch/projects/atlas ID: 34769 · Reply Quote

Erich56 Send message Joined: 18 Dec 15 Posts: 1811 Credit: 118,317,329 RAC: 27,294	Message 34770 - Posted: 27 Mar 2018, 18:24:43 UTC - in response to Message 34769. Is this the page you are looking for? http://atlasathome.cern.ch/atlas_job.php yes, it is. This is the page for now: http://lhcathome.web.cern.ch/projects/atlas thanks, guys, for the information. However, I notice that the above two pages are showing different results. How come? ID: 34770 · Reply Quote

Erich56 Send message Joined: 18 Dec 15 Posts: 1811 Credit: 118,317,329 RAC: 27,294	Message 34771 - Posted: 27 Mar 2018, 18:26:09 UTC - in response to Message 34765. What I still don't understand though: the number of "unsent" tasks in the Project Status Page is going up (it's 245 right now, in contrast to 198 as 1 hour ago) - and still I cannot download any. BOINC keeps saying "no tasks are available for ATLAS Simulation" - something seems to go wrong :-( native Linux-App got WU. So this is correct from the Server-stats. The server status page now shows 416 "unsent" tasks. Still all of them Linux only? My Windows machine doesn't download any? ID: 34771 · Reply Quote

computezrmle Volunteer moderator Volunteer developer Volunteer tester Help desk expert Send message Joined: 15 Jun 08 Posts: 2530 Credit: 253,722,201 RAC: 51,175	Message 34772 - Posted: 27 Mar 2018, 19:50:48 UTC - in response to Message 34771. The server status page now shows 416 "unsent" tasks. Still all of them Linux only? My Windows machine doesn't download any? There's no "linux only". The principle is more like "first come, first serve". And the status page does only show an infrequently updated snapshot. ID: 34772 · Reply Quote

computezrmle Volunteer moderator Volunteer developer Volunteer tester Help desk expert Send message Joined: 15 Jun 08 Posts: 2530 Credit: 253,722,201 RAC: 51,175	Message 34775 - Posted: 28 Mar 2018, 7:21:55 UTC Got my first upload problem with the new backend this morning. https://lhcathome.cern.ch/lhcathome/result.php?resultid=185903051 Mi 28 MÃ¤r 2018 08:20:24 CEST \| LHC@home \| Computation for task E7sMDmzizKsnyYickojUe11pABFKDmABFKDmK2HQDmABFKDmKb3zzm_0 finished Mi 28 MÃ¤r 2018 08:20:26 CEST \| LHC@home \| Started upload of E7sMDmzizKsnyYickojUe11pABFKDmABFKDmK2HQDmABFKDmKb3zzm_0_r400670470_ATLAS_result 89% (of 0.2 GB) were uploaded when: Mi 28 MÃ¤r 2018 09:03:45 CEST \| LHC@home \| Backing off 00:03:57 on upload of E7sMDmzizKsnyYickojUe11pABFKDmABFKDmK2HQDmABFKDmKb3zzm_0_r400670470_ATLAS_result Finally: Mi 28 MÃ¤r 2018 09:07:43 CEST \| LHC@home \| Started upload of E7sMDmzizKsnyYickojUe11pABFKDmABFKDmK2HQDmABFKDmKb3zzm_0_r400670470_ATLAS_result Mi 28 MÃ¤r 2018 09:08:07 CEST \| LHC@home \| Finished upload of E7sMDmzizKsnyYickojUe11pABFKDmABFKDmK2HQDmABFKDmKb3zzm_0_r400670470_ATLAS_result Mi 28 MÃ¤r 2018 09:08:07 CEST \| LHC@home \| Sending scheduler request: To report completed tasks. Mi 28 MÃ¤r 2018 09:08:07 CEST \| LHC@home \| Reporting 1 completed tasks ID: 34775 · Reply Quote

computezrmle Volunteer moderator Volunteer developer Volunteer tester Help desk expert Send message Joined: 15 Jun 08 Posts: 2530 Credit: 253,722,201 RAC: 51,175	Message 34776 - Posted: 28 Mar 2018, 7:49:08 UTC Can't get more than 1 ATLAS task per host/BOINC instance. Preferences are set to 3. Local buffers can hold several days. Is there a server limit? ID: 34776 · Reply Quote

Erich56 Send message Joined: 18 Dec 15 Posts: 1811 Credit: 118,317,329 RAC: 27,294	Message 34779 - Posted: 28 Mar 2018, 10:23:47 UTC - in response to Message 34776. Can't get more than 1 ATLAS task per host/BOINC instance. ... Is there a server limit? Similar situation here - ID: 34779 · Reply Quote

gyllic Send message Joined: 9 Dec 14 Posts: 202 Credit: 2,533,875 RAC: 0	Message 34782 - Posted: 28 Mar 2018, 15:52:53 UTC - in response to Message 34776. Can't get more than 1 ATLAS task per host/BOINC instance. Preferences are set to 3. Local buffers can hold several days. Is there a server limit? my pc has downloaded 2 tasks without a problem and is now at 1 running with 3 in local buffer (preferences for this pc are 4 tasks). so dont think that there is some kind of server limit. The principle is more like "first come, first serve". Isnt there also a "reliable hosts first" policy ID: 34782 · Reply Quote

Erich56 Send message Joined: 18 Dec 15 Posts: 1811 Credit: 118,317,329 RAC: 27,294	Message 34783 - Posted: 28 Mar 2018, 16:47:32 UTC - in response to Message 34782. my pc has downloaded 2 tasks without a problem and is now at 1 running with 3 in local buffer (preferences for this pc are 4 tasks). so dont think that there is some kind of server limit. I have 2 ATLAS tasks running, but would like to have 2 more in local buffer. Before, this never was a problem. Now the situation is that I CANNOT download and tasks for keeping in buffer - only if a finished tasks was uploaded, a new task downloads right away. So I guess that definitely some server-side settings have been made to result in the behaviour described above. ID: 34783 · Reply Quote

gyllic Send message Joined: 9 Dec 14 Posts: 202 Credit: 2,533,875 RAC: 0	Message 34817 - Posted: 30 Mar 2018, 19:30:25 UTC have you removed the max concurrent running tasks cap or is it still in place? ID: 34817 · Reply Quote

rbpeake Send message Joined: 17 Sep 04 Posts: 105 Credit: 32,824,853 RAC: 705	Message 34824 - Posted: 30 Mar 2018, 23:54:24 UTC - in response to Message 34817. have you removed the max concurrent running tasks cap or is it still in place? Maybe this is the source of my problem? https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4139&postid=34820#34820 Regards, Bob P. ID: 34824 · Reply Quote

Erich56 Send message Joined: 18 Dec 15 Posts: 1811 Credit: 118,317,329 RAC: 27,294	Message 34826 - Posted: 31 Mar 2018, 5:55:44 UTC - in response to Message 34817. Last modified: 31 Mar 2018, 5:55:55 UTC have you removed the max concurrent running tasks cap or is it still in place? the maximum concurrent running tasks is still in place (the number is: 2). But this I had for long time now; and until short time ago, it did NOT prevent more ATLAS tasks from downloading (in order to have them in a local buffer). ID: 34826 · Reply Quote

gyllic Send message Joined: 9 Dec 14 Posts: 202 Credit: 2,533,875 RAC: 0	Message 34827 - Posted: 31 Mar 2018, 6:30:21 UTC thx for your answers, but i meant the cap that was introduced due to the server problems before the migration, i.e. the e.g. project wide 10.000 tasks cap ID: 34827 · Reply Quote

LHC@home