Message boards : Number crunching : Computer not receiving Work units
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · 4 . . . 6 · Next

AuthorMessage
vlado101

Send message
Joined: 29 Apr 15
Posts: 33
Credit: 19,566,183
RAC: 19
Message 1402 - Posted: 22 Aug 2016, 14:01:10 UTC

Hello,

I recently added a power edge 840 server to a list of the computer that I use to crunch for Universe@home and WCG.

I have no issues getting work units for WCG, however I consistently do not see any comput out of Universe@home.

The following are the details about the server that I have running:

GenuineIntel Intel(R) Pentium(R) D CPU 2.80GHz [Family 15 Model 6 Stepping 4]
Linux 4.2.0-42-generic (lubuntu).

I have WCG set as 100 and Universe@home set as 20 project weight.

Did anyone experience this issue before with any of their other machines?
ID: 1402 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Krzysztof Piszczek - wspieram ...
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 4 Feb 15
Posts: 841
Credit: 144,180,465
RAC: 2
Message 1404 - Posted: 22 Aug 2016, 14:05:19 UTC - in response to Message 1402.  

I have WCG set as 100 and Universe@home set as 20 project weight.

This is probably reason for rare getting work.
Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT
ID: 1404 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Sergey Kovalchuk

Send message
Joined: 16 Oct 15
Posts: 1
Credit: 521,726
RAC: 0
Message 1407 - Posted: 22 Aug 2016, 14:37:03 UTC - in response to Message 1404.  
Last modified: 22 Aug 2016, 14:43:20 UTC

Client ask work but don't get WU

22.08.2016 17:15:56 | Universe@Home | Requesting new tasks for CPU
22.08.2016 17:15:57 | Universe@Home | Scheduler request completed: got 0 new tasks
22.08.2016 17:15:57 | Universe@Home | No tasks sent
22.08.2016 17:15:57 | Universe@Home | Tasks are committed to other platforms

Or

22.08.2016 17:30:25 | Universe@Home | No tasks are available for Universe BHspin v2

(only this selected in pref)

On server status page - 40K tasks ready to send for BHspin v2

PS. Last task i received 21 Aug 2016, 22:11:21 UTC
ID: 1407 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
vlado101

Send message
Joined: 29 Apr 15
Posts: 33
Credit: 19,566,183
RAC: 19
Message 1409 - Posted: 22 Aug 2016, 16:43:05 UTC - in response to Message 1407.  

Client ask work but don't get WU

22.08.2016 17:15:56 | Universe@Home | Requesting new tasks for CPU
22.08.2016 17:15:57 | Universe@Home | Scheduler request completed: got 0 new tasks
22.08.2016 17:15:57 | Universe@Home | No tasks sent
22.08.2016 17:15:57 | Universe@Home | Tasks are committed to other platforms

Or

22.08.2016 17:30:25 | Universe@Home | No tasks are available for Universe BHspin v2

(only this selected in pref)

On server status page - 40K tasks ready to send for BHspin v2

PS. Last task i received 21 Aug 2016, 22:11:21 UTC



That is the same issue that I am getting. I have two other laptops, and both have work units that are waiting to be crunched. So while WCG are getting priority I do have about six or seven BHspin v2 or other tasks waiting.

This machine when I click on update produces that same message as Sergey every time. I would think that even if my project weight is 20 it would at least capture one or two tasks.
ID: 1409 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
lestus

Send message
Joined: 4 Jul 16
Posts: 3
Credit: 2,509,783
RAC: 0
Message 1414 - Posted: 22 Aug 2016, 21:43:50 UTC

Got this same problem, as of about a day ago, new WUs stopped coming through.
ID: 1414 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
vlado101

Send message
Joined: 29 Apr 15
Posts: 33
Credit: 19,566,183
RAC: 19
Message 1415 - Posted: 23 Aug 2016, 12:07:47 UTC

I now have the same issue for my two other laptops. Both are not receiving work units.
ID: 1415 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Krzysztof Piszczek - wspieram ...
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 4 Feb 15
Posts: 841
Credit: 144,180,465
RAC: 2
Message 1416 - Posted: 23 Aug 2016, 12:13:12 UTC - in response to Message 1415.  

After two days my two odroids and one Windows machine have new tasks.
Firstly WU's are coming to "reliable" hosts with delay_bound less than two days.
I had to set this because loads of WU's stuck on some hosts which not send back results in deadline.

We will see if this helps. If not, I will switch it back to previous state.
Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT
ID: 1416 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jim1348

Send message
Joined: 28 Feb 15
Posts: 253
Credit: 200,562,581
RAC: 0
Message 1417 - Posted: 23 Aug 2016, 12:23:24 UTC - in response to Message 1416.  

I received over 40 reionization tasks a couple of hours ago, so it is working again. I keep my buffer at the default size of 0.1 + 0.5 days.
ID: 1417 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
fruehwf

Send message
Joined: 5 Jul 16
Posts: 31
Credit: 18,447,833
RAC: 0
Message 1418 - Posted: 23 Aug 2016, 12:40:44 UTC - in response to Message 1417.  

23.08.2016 14:38:26 | Universe@Home | Requesting new tasks for CPU
23.08.2016 14:38:28 | Universe@Home | Scheduler request completed: got 0 new tasks
23.08.2016 14:38:28 | Universe@Home | No tasks sent
23.08.2016 14:38:28 | Universe@Home | No tasks are available for Universe BHspin
23.08.2016 14:38:28 | Universe@Home | No tasks are available for Universe QuarkStars
23.08.2016 14:38:28 | Universe@Home | No tasks are available for Universe Ultraviolet reionization
23.08.2016 14:38:28 | Universe@Home | Tasks are committed to other platforms


mittlere Taskdauer 1.67 Tage
(delay_bound 1.67 Days)

None of my machines gets new work..

regards Franz
ID: 1418 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
vlado101

Send message
Joined: 29 Apr 15
Posts: 33
Credit: 19,566,183
RAC: 19
Message 1419 - Posted: 23 Aug 2016, 13:56:37 UTC

Just hit update on all of my machines and get the same log as shown by other members here. Still am not getting tasks.
ID: 1419 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
BetelgeuseFive

Send message
Joined: 8 Feb 16
Posts: 19
Credit: 22,093,689
RAC: 84
Message 1420 - Posted: 23 Aug 2016, 15:24:35 UTC

Something is wrong. I am running on two nearly identical Raspberry Pi's (model 2B).
They take 20-24 hours to process a Universe BHspin v2 task.
I did not get any work (like may others) for a while, but now all of a sudden I am getting lots of work. The problem is that the estimated time for these tasks is 1 hour. This means I am getting too many and they may even abort because they take too long to process before they are ready.
Did anyone else notice sudden problems with estimated times ?
ID: 1420 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Krzysztof Piszczek - wspieram ...
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 4 Feb 15
Posts: 841
Credit: 144,180,465
RAC: 2
Message 1421 - Posted: 23 Aug 2016, 15:33:05 UTC - in response to Message 1420.  

No, for BHspin2 it looks good, as time to send it back is 2 weeks.
UV will take probably less than one hour and RPI can process many of them in one day.

Also, I'm right that you get only BHspin2 tasks?
Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT
ID: 1421 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
BetelgeuseFive

Send message
Joined: 8 Feb 16
Posts: 19
Credit: 22,093,689
RAC: 84
Message 1422 - Posted: 23 Aug 2016, 16:20:24 UTC - in response to Message 1421.  

No, for BHspin2 it looks good, as time to send it back is 2 weeks.
UV will take probably less than one hour and RPI can process many of them in one day.

Also, I'm right that you get only BHspin2 tasks?


I have only enabled BHspin2 at the moment, I did not know there were other application for Raspberry Pi (I will enable them as well to check it out). The esitimates for my BHspin2 tasks are a real problem. According to BOINC manager they now have an estimate of only 1 hour, while they will take 20-24 hours to complete. This means I will get way too many tasks (I have now set no new tasks to prevent this) and may be unable to process them all before the deadline.

Tom
ID: 1422 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
fruehwf

Send message
Joined: 5 Jul 16
Posts: 31
Credit: 18,447,833
RAC: 0
Message 1423 - Posted: 23 Aug 2016, 16:36:14 UTC

23.08.2016 18:35:17 | Universe@Home | Sending scheduler request: To fetch work.
23.08.2016 18:35:17 | Universe@Home | Requesting new tasks for CPU
23.08.2016 18:35:18 | Universe@Home | Scheduler request completed: got 0 new tasks
23.08.2016 18:35:18 | Universe@Home | No tasks sent
23.08.2016 18:35:18 | Universe@Home | Tasks are committed to other platforms

Still the same situation....
ID: 1423 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Krzysztof Piszczek - wspieram ...
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 4 Feb 15
Posts: 841
Credit: 144,180,465
RAC: 2
Message 1424 - Posted: 23 Aug 2016, 16:43:51 UTC - in response to Message 1422.  

The main concern is starting form BHspin2 (then Universe-UV) regarding to different results from various architectures we had to start using homogoneus redundancy mechanism which allow server to send tasks to each platform based on previous returned units.
This mechanism in BOINC never works perfectly and adding priorities to it makes server more confused about which computer are reliable and which one are not.

Regarding to BOINC documentation this mechanism should regulate itself and we can see that it isn't still correct.
I'm trying different solutions all the time, but every action made takes hours to see effect.

At the moment, some of machines gets work, some not, but it is progress - yesterday no one from them get work...
Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT
ID: 1424 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ryan

Send message
Joined: 31 Jul 16
Posts: 4
Credit: 321,900
RAC: 0
Message 1425 - Posted: 23 Aug 2016, 17:45:42 UTC

my computers also are getting no work.Server says WUs are alloted to other platforms.Thanks for any help.
ID: 1425 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Fractale2015

Send message
Joined: 13 Aug 15
Posts: 26
Credit: 2,848,895
RAC: 0
Message 1426 - Posted: 23 Aug 2016, 18:58:54 UTC - in response to Message 1402.  

Idem

23/08/2016 20:38:55 | Universe@Home | Sending scheduler request: To fetch work.
23/08/2016 20:38:55 | Universe@Home | Requesting new tasks for CPU
23/08/2016 20:39:07 | Universe@Home | Scheduler request completed: got 0 new tasks
23/08/2016 20:39:07 | Universe@Home | No tasks sent
23/08/2016 20:39:07 | Universe@Home | Tasks are committed to other platforms
ID: 1426 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
fruehwf

Send message
Joined: 5 Jul 16
Posts: 31
Credit: 18,447,833
RAC: 0
Message 1427 - Posted: 23 Aug 2016, 19:38:40 UTC

It seems to me, that only the Cpu with the highest FPU's get work...
:-(
ID: 1427 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
lestus

Send message
Joined: 4 Jul 16
Posts: 3
Credit: 2,509,783
RAC: 0
Message 1428 - Posted: 23 Aug 2016, 20:46:07 UTC

I had 1 task aborted out of 3417, they all typically completed within 3-5 hours and sent back, sometimes completed in 40 mins, something else might be broken, doubt it's due to reliability, i updated, 3 days no work now..
ID: 1428 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Steve Hawker*

Send message
Joined: 10 Mar 15
Posts: 11
Credit: 309,361
RAC: 0
Message 1429 - Posted: 23 Aug 2016, 22:07:03 UTC

Due to the Windows problem, I created a Linux VM.

Guess what....

Tue 23 Aug 2016 03:00:24 PM PDT | Universe@Home | No tasks are available for Universe QuarkStars
Tue 23 Aug 2016 03:00:24 PM PDT | Universe@Home | No tasks are available for Universe Ultraviolet reionization

Meanwhile 110,000 UV tasks on server
ID: 1429 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
1 · 2 · 3 · 4 . . . 6 · Next

Message boards : Number crunching : Computer not receiving Work units




Copyright © 2024 Copernicus Astronomical Centre of the Polish Academy of Sciences
Project server and website managed by Krzysztof 'krzyszp' Piszczek