Message boards : Number crunching : All running ATLAS and CMS tasks "aborted by project" - why so?
Message board moderation

To post messages, you must log in.

AuthorMessage
Erich56

Send message
Joined: 18 Dec 15
Posts: 1346
Credit: 25,252,756
RAC: 24,077
Message 31566 - Posted: 23 Jul 2017, 18:32:27 UTC

Short time ago, on one of my PCs on which I had running 4 ATLAS and 4 CMS tasks, all these were "aborted by Project".
On the other two PCs, this was not the case.
None of the aborted tasks are being shown in my tasks list on the Webpage.

New CMS tasks were downloaded and got started, ATLAS tasks show up in the BOINC Manager as being downloaded, but the download is extremely slow and comes to a complete halt most of the time.
I checked my Internet connection, it works perfectly.

Anyone making the same experience?
ID: 31566 · Report as offensive     Reply Quote
ivan
Volunteer moderator
Project tester
Volunteer developer
Volunteer tester
Project scientist
Avatar

Send message
Joined: 29 Aug 05
Posts: 750
Credit: 5,688,626
RAC: 0
Message 31572 - Posted: 23 Jul 2017, 19:06:16 UTC - in response to Message 31566.  

ID: 31572 · Report as offensive     Reply Quote
Erich56

Send message
Joined: 18 Dec 15
Posts: 1346
Credit: 25,252,756
RAC: 24,077
Message 31573 - Posted: 23 Jul 2017, 19:12:31 UTC - in response to Message 31572.  

but there's something funny going on

okay, I see. Thanks.
So I can rest assured that nothing is wrong with my PC :-)
ID: 31573 · Report as offensive     Reply Quote
Profile Olivier Fehr
Avatar

Send message
Joined: 1 Jun 17
Posts: 9
Credit: 964,242
RAC: 0
Message 31574 - Posted: 23 Jul 2017, 19:22:07 UTC - in response to Message 31566.  

Same experience here - on several different platforms. As an temporary 'emergency' measure, I've stopped accepting new tasks from LCH@home until I know what's going on...
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4362&postid=31564#31564
ID: 31574 · Report as offensive     Reply Quote
Erich56

Send message
Joined: 18 Dec 15
Posts: 1346
Credit: 25,252,756
RAC: 24,077
Message 31575 - Posted: 23 Jul 2017, 19:41:23 UTC

What's interesting is that this happened only on one of my 3 PCs with which I crunch LHC tasks.
ID: 31575 · Report as offensive     Reply Quote
Erich56

Send message
Joined: 18 Dec 15
Posts: 1346
Credit: 25,252,756
RAC: 24,077
Message 31577 - Posted: 23 Jul 2017, 19:55:03 UTC

The following was shown in the BOINC event log for the 8 abortet Tasks:

23/07/2017 20:16:51 | LHC@home | [error] Got ack for task CMS_20797_1500746391.189425_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task CMS_20276_1500746091.040548_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task CMS_17026_1500712758.273755_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task CMS_3419_1500753597.693533_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task aMQNDm2FOtqnSu7Ccp2YYBZmABFKDmABFKDmXNGKDmwtGKDmsXBe3n_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task NsiNDmtXStqnSu7Ccp2YYBZmABFKDmABFKDmXNGKDmU2GKDm0MrXKo_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task 60lKDmm6TtqnSu7Ccp2YYBZmABFKDmABFKDmXNGKDms5GKDm0zNFSm_0, but can't find it
23/07/2017 20:16:51 | LHC@home | [error] Got ack for task GleMDm0hTtqnSu7Ccp2YYBZmABFKDmABFKDmXNGKDmN5GKDmNstWLn_0, but can't find it
ID: 31577 · Report as offensive     Reply Quote
Erich56

Send message
Joined: 18 Dec 15
Posts: 1346
Credit: 25,252,756
RAC: 24,077
Message 31578 - Posted: 23 Jul 2017, 20:21:23 UTC

short time ago, on one of my other PCs the two running CMS tasks were abortet.

The BOINC log shows:

23.07.2017 21:52:14 | LHC@home | [error] garbage_collect(); still have active task for acked result CMS_22808_1500731676.693265_0; state 5
23.07.2017 21:52:19 | LHC@home | [error] garbage_collect(); still have active task for acked result CMS_16476_1500744589.556904_0; state 5
23.07.2017 21:52:24 | LHC@home | [error] garbage_collect(); still have active task for acked result CMS_22808_1500731676.693265_0; state 6
23.07.2017 21:52:29 | LHC@home | [error] garbage_collect(); still have active task for acked result CMS_16476_1500744589.556904_0; state 6
23.07.2017 21:52:34 | LHC@home | Computation for task CMS_22808_1500731676.693265_0 finished
23.07.2017 21:52:34 | LHC@home | Computation for task CMS_16476_1500744589.556904_0 finished

what does this now mean?
ID: 31578 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 853
Credit: 1,619,050
RAC: 0
Message 31601 - Posted: 24 Jul 2017, 12:31:44 UTC - in response to Message 31566.  

I think the aborted tasks were a side effect of deleting ancient results.
(I should have restricted this to SixTrack, but there many many others from
other sub-projects cluttering the database/upload/download directories
which are shared by all.) Full report to follow.
Eric.
P.S. They "might" have been deleted anyway if redundant and too late, but I
rather think not.

Short time ago, on one of my PCs on which I had running 4 ATLAS and 4 CMS tasks, all these were "aborted by Project".
On the other two PCs, this was not the case.
None of the aborted tasks are being shown in my tasks list on the Webpage.

New CMS tasks were downloaded and got started, ATLAS tasks show up in the BOINC Manager as being downloaded, but the download is extremely slow and comes to a complete halt most of the time.
I checked my Internet connection, it works perfectly.

Anyone making the same experience?

ID: 31601 · Report as offensive     Reply Quote

Message boards : Number crunching : All running ATLAS and CMS tasks "aborted by project" - why so?


©2021 CERN