Development for LHC@home

A current task should pick up a new job once its one-hour pause is over. That's the theory, anyway...

Thanks, Laurence.

In practice too!

Returned successful the first job after the restart: jobNumber=385

Another "runaway".
10+ fails from the same IP address and still continuing.

Yesterday's task that started at 5:30pm was reported at 20:26 today, not a problem.

However I now have 7 tasks listed as having been sent to the computer (472) all at 20:26:27...

77329 68196 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77355 68758 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77356 68888 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77365 68749 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77366 68824 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77377 68431 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77277 68012 472 3 Feb 2016, 20:26:27 UTC 10 Feb 2016, 20:26:27 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)

Looking on the computer there is only 1 task listed, it is using 1 full core (on a Helix job), the one it has is the bottom one in the list above (name: CMS_14774_1427806996.975027_1)

Nothing changed my end !

Would it be possible to allow for a second task on one computer or, even better,allow a setting to specify the number of cores to be used?

Yesterday's task that started at 5:30pm was reported at 20:26 today, not a problem.

However I now have 7 tasks listed as having been sent to the computer (472) all at 20:26:27...


Looking on the computer there is only 1 task listed, it is using 1 full core (on a Helix job), the one it has is the bottom one in the list above (name: CMS_14774_1427806996.975027_1)

Nothing changed my end !

I sort of had the same thing
http://boincai05.cern.ch/CMS-dev/results.php?userid=192

I sent in one and got one back to do but when I look here at my *Tasks* it has 2 in progress and one of them is the same number as the one just sent back.

(I am only testing these on one host now)

I sort of had the same thing
http://boincai05.cern.ch/CMS-dev/results.php?userid=192

I sent in one and got one back to do but when I look here at my *Tasks* it has 2 in progress and one of them is the same number as the one just sent back.

(I am only testing these on one host now)

I can see you have two sent out but the numbers don't match the one returned (unless the task name is the same which I can't see, I can I was too lazy to click !)...

77354 68905 3 Feb 2016, 19:08:02 UTC 10 Feb 2016, 19:08:02 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77211 66415 3 Feb 2016, 19:08:02 UTC 10 Feb 2016, 19:08:02 UTC In progress --- --- --- CMS Simulation v46.22 (vbox64)
77171 66820 2 Feb 2016, 10:18:38 UTC 3 Feb 2016, 14:32:27 UTC Completed and validated

At least you have 8 cores for them to run on, my old computer only has 4 HT cores so no idea why it thinks it should have 7 tasks to do !

Well they thing is that those *2* new tasks it says are in progress are NOT even on this host.

Just one of them. (77354)

That other one is not on the host.

I got 1, but within a few minutes 16 assigned, but only task 65665 is on my system.

77583 56057 4 Feb 2016, 13:02:33 UTC 11 Feb 2016, 13:02:33 UTC In progress
77098 65665 4 Feb 2016, 13:02:33 UTC 11 Feb 2016, 13:02:33 UTC In progress
77777 63563 4 Feb 2016, 13:00:20 UTC 11 Feb 2016, 13:00:20 UTC In progress
77778 58580 4 Feb 2016, 13:00:20 UTC 11 Feb 2016, 13:00:20 UTC In progress
77819 68313 4 Feb 2016, 13:00:20 UTC 11 Feb 2016, 13:00:20 UTC In progress
77820 68006 4 Feb 2016, 13:00:20 UTC 11 Feb 2016, 13:00:20 UTC In progress
77802 67138 4 Feb 2016, 13:00:08 UTC 11 Feb 2016, 13:00:08 UTC In progress
77803 66530 4 Feb 2016, 13:00:08 UTC 11 Feb 2016, 13:00:08 UTC In progress
77804 65222 4 Feb 2016, 13:00:08 UTC 11 Feb 2016, 13:00:08 UTC In progress
77805 66706 4 Feb 2016, 13:00:08 UTC 11 Feb 2016, 13:00:08 UTC In progress
77405 69140 4 Feb 2016, 12:59:23 UTC 11 Feb 2016, 12:59:23 UTC In progress
77438 69057 4 Feb 2016, 12:59:23 UTC 11 Feb 2016, 12:59:23 UTC In progress
77440 69783 4 Feb 2016, 12:59:23 UTC 11 Feb 2016, 12:59:23 UTC In progress
77798 64754 4 Feb 2016, 12:57:49 UTC 11 Feb 2016, 12:57:49 UTC In progress
77807 62932 4 Feb 2016, 12:57:49 UTC 11 Feb 2016, 12:57:49 UTC In progress
77808 67308 4 Feb 2016, 12:57:49 UTC 11 Feb 2016, 12:57:49 UTC In progress

Not strange that the project status shows 536 tasks in progress.

Although I had 31GB free diskspace available for BOINC, I got a few times:

04-Feb-2016 13:57:46 [CMS-dev] Sending scheduler request: To fetch work.
04-Feb-2016 13:57:46 [CMS-dev] Requesting new tasks for CPU
04-Feb-2016 13:57:48 [CMS-dev] Scheduler request completed: got 0 new tasks
04-Feb-2016 13:57:48 [CMS-dev] No tasks sent
04-Feb-2016 13:57:48 [CMS-dev] CMS Simulation needs 5895.94MB more disk space. You currently have 3640.80 MB available and it needs 9536.74 MB.

Probably related, but did not get tasks at that moment.
After restarting BOINC client, I got task 65665.

Because of that the "Task ready to send" are falling like a rock.
Please keep an eye on that.

Because of that the "Task ready to send" are falling like a rock.
Please keep an eye on that.

We're just surviving, cause all last ~thousand tasks are resends of workunits original created in March 2015.

1 host with 52 tasks in progress
1 host with 32 tasks in progress

I was going to congratulate you on winning the trophy for most jobs, puts me down to 3rd now :-(

No wonder all the jobs are disappearing from the queue !

Edit: Down to 4th now !

I was going to congratulate you on winning the trophy for most jobs, puts me down to 3rd now :-(

That are not mine hosts, I've 'only' 16 virtual tasks and 1 real task in progress.

That are not mine hosts, I've 'only' 16 virtual tasks and 1 real task in progress.

I know, that's why you are 3rd and I'm now 4th.

THIS has to be an *Issue*

---

Mad Scientist For Life

I know, the invitation code is a problem........

Because of that the "Task ready to send" are falling like a rock.
Please keep an eye on that.

We're just surviving, cause all last ~thousand tasks are resends of workunits original created in March 2015.

March 2015 was probably the last time I created tasks (and first time, Daniele had done them before that)... I'm stumped, Dashboard, as unreliable as it is, is showing recent progress though there is a huge spike in failures on the jobs graph. Let's leave it overnight, I have to dig out the recipe on how to create more tasks, and see what happens to the progress charts and tables.

---

Any ideas what is causing this behaviour?

I have not seen this on my computer,but we need to know:

a)When did it start?
b)Is it also present at vLHC?(with cms-simulation tasks)
c)Do the computers, it happened to, have something in common* that makes them different to the other ones.

Any other suggestions/comments?

* (like os, boinc version, vbox version etc)

I would think the fact that I got 7 tasks all sent out at the exact same time means it started at the server end rather than the client end.

You haven't upgraded anything ?