Application 0.15 (t1) (beta test) Result Not Completing Successfully

Author	Message
rbpeake Send message Joined: 4 Oct 19 Posts: 8 Credit: 3,108,300 RAC: 0	Message 506 - Posted: 6 Feb 2020, 16:25:38 UTC All of my 0.15 (t1) (beta test) results are not doing calculations, but continue running anyway and do not stop. Will need to abort them it appears. ID: 506 · Rating: 0 · rate: / Reply Quote

Aurum Send message Joined: 14 Dec 19 Posts: 68 Credit: 45,744,261 RAC: 0	Message 507 - Posted: 6 Feb 2020, 17:43:39 UTC Last modified: 6 Feb 2020, 17:43:58 UTC I'm seeing a lot of 0.15s that just sit there near completion but little to no CPU usage. Do they need to meditate before ascending??? ID: 507 · Rating: 0 · rate: / Reply Quote

Trotador Send message Joined: 3 Jan 20 Posts: 5 Credit: 31,342,930 RAC: 0	Message 509 - Posted: 6 Feb 2020, 18:13:42 UTC +1 ID: 509 · Rating: 0 · rate: / Reply Quote

damotbe Volunteer moderator Project administrator Project developer Project tester Project scientist Help desk expert Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0	Message 510 - Posted: 6 Feb 2020, 18:37:52 UTC I just test the last Boinc wrapper version because in the 0.14 it is a very old one (available on the boinc website). Seems that the version on github is bogus... I downgrade to 0.14, you can cancel jobs ID: 510 · Rating: 0 · rate: / Reply Quote

rbpeake Send message Joined: 4 Oct 19 Posts: 8 Credit: 3,108,300 RAC: 0	Message 511 - Posted: 6 Feb 2020, 20:42:34 UTC Can't seem to get any work units. ID: 511 · Rating: 0 · rate: / Reply Quote

Jim1348 Send message Joined: 3 Oct 19 Posts: 153 Credit: 32,412,973 RAC: 0	Message 512 - Posted: 6 Feb 2020, 23:02:57 UTC - in response to Message 507. Do they need to meditate before ascending??? I will be running them on a Zen core. It might work. ID: 512 · Rating: 0 · rate: / Reply Quote

Aurum Send message Joined: 14 Dec 19 Posts: 68 Credit: 45,744,261 RAC: 0	Message 518 - Posted: 7 Feb 2020, 16:49:43 UTC Are 0.16s and 0.17s behaving like 0.15s and stalling at the last moment??? ID: 518 · Rating: 0 · rate: / Reply Quote

Jim1348 Send message Joined: 3 Oct 19 Posts: 153 Credit: 32,412,973 RAC: 0	Message 519 - Posted: 7 Feb 2020, 16:50:54 UTC The 0.16 don't work either. I have over 20 of them running on two Ubuntu 18.04 machines, and they are up to about four hours with 0% CPU usage. I am aborting them. ID: 519 · Rating: 0 · rate: / Reply Quote

Aurum Send message Joined: 14 Dec 19 Posts: 68 Credit: 45,744,261 RAC: 0	Message 520 - Posted: 7 Feb 2020, 17:41:47 UTC They're really piling up. Time for a server-side abort signal. ID: 520 · Rating: 0 · rate: / Reply Quote

damotbe Volunteer moderator Project administrator Project developer Project tester Project scientist Help desk expert Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0	Message 521 - Posted: 7 Feb 2020, 18:44:07 UTC - in response to Message 520. Finally, the 0.17 is working well ! I don't know how to send an abort signal from server side with a condition on version. I Look at the documentation. ID: 521 · Rating: 0 · rate: / Reply Quote

damotbe Volunteer moderator Project administrator Project developer Project tester Project scientist Help desk expert Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0	Message 522 - Posted: 7 Feb 2020, 19:30:39 UTC - in response to Message 521. I Look at the documentation. Nothing planed to abort only computation with a given app version id... All these problems, because I tried to use the official boinc wrapper on github... 0.17 is a rollback to the old wrapper plus a sentry process (that I just coded) to patch bogus signal of this version. From the beginning, our application works and we have a lot of problems with the official BOINC tools. ID: 522 · Rating: 0 · rate: / Reply Quote

crashtech Send message Joined: 9 Dec 19 Posts: 11 Credit: 19,162,966 RAC: 0	Message 523 - Posted: 7 Feb 2020, 19:49:05 UTC Hi, how do I tell between good and bad units? On a few machines that were experiencing low CPU utilization or unusually long run times, I aborted them all, but I'd rather not make a habit of that. ID: 523 · Rating: 0 · rate: / Reply Quote

damotbe Volunteer moderator Project administrator Project developer Project tester Project scientist Help desk expert Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0	Message 524 - Posted: 7 Feb 2020, 20:11:37 UTC - in response to Message 523. 0.15 and 0.16 are bogus on Linux Windows and Mac depends of official vbox_wrapper, so they could idle instead of computing... ID: 524 · Rating: 0 · rate: / Reply Quote

Jim1348 Send message Joined: 3 Oct 19 Posts: 153 Credit: 32,412,973 RAC: 0	Message 525 - Posted: 7 Feb 2020, 20:30:03 UTC - in response to Message 524. Yes, 0.17 is working great. We are in business again. ID: 525 · Rating: 0 · rate: / Reply Quote

Aurum Send message Joined: 14 Dec 19 Posts: 68 Credit: 45,744,261 RAC: 0	Message 528 - Posted: 8 Feb 2020, 10:41:57 UTC - in response to Message 524. 0.15 and 0.16 are bogus on Linux Windows and Mac depends of official vbox_wrapper, so they could idle instead of computing... Then a Linux host should not even be sent a windoze WU. Right? ID: 528 · Rating: 0 · rate: / Reply Quote

damotbe Volunteer moderator Project administrator Project developer Project tester Project scientist Help desk expert Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0	Message 529 - Posted: 8 Feb 2020, 10:47:44 UTC - in response to Message 528. 0.15 and 0.16 are bogus on Linux Windows and Mac depends of official vbox_wrapper, so they could idle instead of computing... Then a Linux host should not even be sent a windoze WU. Right? Normally, no. ID: 529 · Rating: 0 · rate: / Reply Quote

Serg Send message Joined: 12 Jan 20 Posts: 12 Credit: 220,914 RAC: 0	Message 530 - Posted: 8 Feb 2020, 14:49:37 UTC - in response to Message 529. As i see, the project sends one WU to linux and Windows in many cases: Example ID: 530 · Rating: 0 · rate: / Reply Quote

damotbe Volunteer moderator Project administrator Project developer Project tester Project scientist Help desk expert Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0	Message 531 - Posted: 8 Feb 2020, 16:39:27 UTC - in response to Message 530. To be clear, one WU is sent to two volunteers, regardless of their operating system. It can be heterogeneous like the example you're showing. On the other hand, a Linux host will only receive calculations with the application for Linux. ID: 531 · Rating: 0 · rate: / Reply Quote

STE\/E Send message Joined: 7 Oct 19 Posts: 8 Credit: 2,016,759 RAC: 0	Message 562 - Posted: 14 Feb 2020, 10:48:24 UTC All errors so far they don't see to run very long ... 1719898 1101058 485 13 Feb 2020, 22:34:27 UTC 13 Feb 2020, 22:35:34 UTC Validate error 6.10 1.68 --- NWChem long v0.18 (t1) x86_64-pc-linux-gnu 1719741 1101036 485 13 Feb 2020, 18:49:07 UTC 13 Feb 2020, 18:51:07 UTC Validate error 5.30 1.67 --- NWChem long v0.18 (t1) x86_64-pc-linux-gnu 1718999 1101014 485 13 Feb 2020, 10:41:25 UTC 13 Feb 2020, 10:42:33 UTC Validate error 4.26 0.00 --- NWChem long v0.18 (t2) x86_64-pc-linux-gnu 1718890 1101013 485 13 Feb 2020, 9:12:08 UTC 13 Feb 2020, 9:15:40 UTC Validate error 3.32 0.00 --- NWChem long v0.18 (t4) x86_64-pc-linux-gnu 1716552 1101042 485 12 Feb 2020, 21:55:03 UTC 13 Feb 2020, 0:10:34 UTC Validate error 4.33 0.00 --- NWChem long v0.18 (t2) x86_64-pc-linux-gnu ID: 562 · Rating: 0 · rate: / Reply Quote

PDW Send message Joined: 3 Oct 19 Posts: 43 Credit: 40,548,179 RAC: 0	Message 563 - Posted: 14 Feb 2020, 10:58:25 UTC - in response to Message 562. All errors so far they don't seem to run very long ... You want to get yourself one of these... https://quchempedia.univ-angers.fr/athome/workunit.php?wuid=1101056 I assume it is a validation problem again but I've let it run as it is for testing and I just got given a new shiny badge for testing :) ID: 563 · Rating: 0 · rate: / Reply Quote