| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: CEP WUs for LINUX [Resolved] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 9
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Is there a shortage of WUs for the LINUX platform. Windows machines seem to get their allotment but all but one LINUX machine is void of CEP work units. Keep getting:
----------------------------------------eldor 5472 World Community Grid 2/6/2013 8:00:49 PM Sending scheduler request: To fetch work. 5473 World Community Grid 2/6/2013 8:00:49 PM Reporting 1 completed tasks 5474 World Community Grid 2/6/2013 8:00:49 PM Requesting new tasks for CPU 5475 World Community Grid 2/6/2013 8:00:52 PM Scheduler request completed: got 0 new tasks 5476 World Community Grid 2/6/2013 8:00:52 PM No tasks sent 5477 World Community Grid 2/6/2013 8:00:52 PM No tasks are available for The Clean Energy Project - Phase 2 5478 World Community Grid 2/6/2013 8:00:52 PM No tasks are available for the applications you have selected. 5479 World Community Grid 2/6/2013 8:00:52 PM Tasks are committed to other platforms 5480 World Community Grid 2/6/2013 8:00:54 PM Computation for task DSFL_00050-34_0000034_0277_0 finished 5481 World Community Grid 2/6/2013 8:00:54 PM Starting task DSFL_00050-34_0000046_0920_0 using dsfl version 625 in slot 11 5482 World Community Grid 2/6/2013 8:00:56 PM Started upload of DSFL_00050-34_0000034_0277_0_0 5483 World Community Grid 2/6/2013 8:01:01 PM Finished upload of DSFL_00050-34_0000034_0277_0_0 5484 World Community Grid 2/6/2013 8:01:07 PM Sending scheduler request: To fetch work. 5485 World Community Grid 2/6/2013 8:01:07 PM Reporting 1 completed tasks 5486 World Community Grid 2/6/2013 8:01:07 PM Requesting new tasks for CPU 5487 World Community Grid 2/6/2013 8:01:10 PM Scheduler request completed: got 0 new tasks 5488 World Community Grid 2/6/2013 8:01:10 PM No tasks sent 5489 World Community Grid 2/6/2013 8:01:10 PM No tasks are available for The Clean Energy Project - Phase 2 5490 World Community Grid 2/6/2013 8:01:10 PM No tasks are available for the applications you have selected. 5491 World Community Grid 2/6/2013 8:01:10 PM Tasks are committed to other platforms [Edit 2 times, last edit by Doneske at Feb 9, 2013 6:31:04 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Not really as it's all one single queue. Sometimes there's a long repair/no reply queue for a particular platform which clog up the supply as these have the priority in the feeder. Then you get:
5479 World Community Grid 2/6/2013 8:00:52 PM Tasks are committed to other platforms Getting no work available message is sometimes misleading, when in fact there was a temp vacuum for all your selected sciences. At any rate, was concentrating on another science, but briefly flipped on CEP2 as only science which is set to assign max 4, and this came out on my Ubuntu box [with extra log flags, client 7.0.39, which has too 14 more reasons printed why work is not received]. Got 36,61 hours worth on first call, but this is 1.5 days later than your log is dated. 39922 World Community Grid 2/8/2013 10:17:33 AM [sched_op] Starting scheduler request 39923 World Community Grid 2/8/2013 10:17:33 AM Sending scheduler request: To fetch work. 39924 World Community Grid 2/8/2013 10:17:33 AM Requesting new tasks for CPU 39925 World Community Grid 2/8/2013 10:17:33 AM [sched_op] CPU work request: 128403.18 seconds; 0.00 devices 39926 World Community Grid 2/8/2013 10:17:38 AM Scheduler request completed: got 4 new tasks 39927 World Community Grid 2/8/2013 10:17:38 AM [sched_op] Server version 701 39928 World Community Grid 2/8/2013 10:17:38 AM Project requested delay of 11 seconds 39929 World Community Grid 2/8/2013 10:17:38 AM [sched_op] estimated total CPU task duration: 131804 seconds 39930 World Community Grid 2/8/2013 10:17:38 AM [sched_op] Deferring communication for 11 sec 39931 World Community Grid 2/8/2013 10:17:38 AM [sched_op] Reason: requested by project 39932 World Community Grid 2/8/2013 10:17:40 AM Started download of 991c4310da056b4a804e657740778c9f.zip 39933 World Community Grid 2/8/2013 10:17:40 AM Started download of 7ff40f5948f1dacf3029fa330bf335be.zip 39934 World Community Grid 2/8/2013 10:17:47 AM Finished download of 991c4310da056b4a804e657740778c9f.zip 39935 World Community Grid 2/8/2013 10:17:47 AM Started download of 11642ec835988649019fe10bfe627b92.zip 39936 World Community Grid 2/8/2013 10:17:50 AM Finished download of 11642ec835988649019fe10bfe627b92.zip 39937 World Community Grid 2/8/2013 10:17:50 AM Started download of 9374fefc0abbd2cbcbb764abe87372db.zip 39938 World Community Grid 2/8/2013 10:17:51 AM Finished download of 7ff40f5948f1dacf3029fa330bf335be.zip 39939 World Community Grid 2/8/2013 10:17:51 AM Finished download of 9374fefc0abbd2cbcbb764abe87372db.zip |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Just checked - we have plenty of workunits in the feeder.
Best wishes from Your Harvard CEP team |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Updated profile this morning to CEP only, increased cache, hit update and got same message. Have assign max set to 3 and one LINUX box is getting one sometimes two. There are 44 cores waiting to process CEP once I get some. I will poke around on the machines later and try to determine what the problem might be. I'll check the local preferences again in case I missed something. Thanks
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Maybe post a start-up log of one of your suddenly hard-nosed machines. The platform story just is weird... it would apply to all WCG sciences if that were wrong.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Reviewed local preferences and nothing seemed out of the ordinary. I'm including the startup messages as requested.
1 2/8/2013 10:28:34 PM Starting BOINC client version 7.0.31 for x86_64-pc-linux-gnu 2 2/8/2013 10:28:34 PM log flags: file_xfer, sched_ops, task 3 2/8/2013 10:28:34 PM Libraries: libcurl/7.27.0 OpenSSL/1.0.1c zlib/1.2.7 libidn/1.25 librtmp/2.3 4 2/8/2013 10:28:34 PM Running as a daemon 5 2/8/2013 10:28:34 PM Data directory: /home/mike/BOINC 6 2/8/2013 10:28:34 PM Processor: 8 AuthenticAMD AMD FX(tm)-8150 Eight-Core Processor [Family 21 Model 1 Stepping 2] 7 2/8/2013 10:28:34 PM Processor: 2.00 MB cache 8 2/8/2013 10:28:34 PM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 9 2/8/2013 10:28:34 PM OS: Linux: 3.5.0-23-generic 10 2/8/2013 10:28:34 PM Memory: 15.62 GB physical, 256.00 MB virtual 11 2/8/2013 10:28:34 PM Disk: 16.78 GB total, 12.98 GB free 12 2/8/2013 10:28:34 PM Local time is UTC -6 hours 13 2/8/2013 10:28:34 PM No usable GPUs found 14 2/8/2013 10:28:34 PM Config: GUI RPC allowed from: 15 2/8/2013 10:28:34 PM Config: 16 World Community Grid 2/8/2013 10:28:34 PM URL http://www.worldcommunitygrid.org/; Computer ID 2209994; resource share 100 17 World Community Grid 2/8/2013 10:28:34 PM General prefs: from World Community Grid (last modified 26-Jan-2013 17:50:04) 18 World Community Grid 2/8/2013 10:28:34 PM Computer location: work 19 2/8/2013 10:28:34 PM General prefs: using separate prefs for work 20 2/8/2013 10:28:34 PM Reading preferences override file 21 2/8/2013 10:28:34 PM Preferences: 22 2/8/2013 10:28:34 PM max memory usage when active: 15995.18MB 23 2/8/2013 10:28:34 PM max memory usage when idle: 15995.18MB 24 2/8/2013 10:28:34 PM max disk usage: 13.06GB 25 2/8/2013 10:28:34 PM (to change preferences, visit the web site of an attached project, or select Preferences in the Manager) 26 2/8/2013 10:28:34 PM Not using a proxy 27 World Community Grid 2/8/2013 10:28:36 PM Restarting task DSFL_00050-37_0000017_0052_0 using dsfl version 625 in slot 1 28 World Community Grid 2/8/2013 10:28:36 PM Restarting task DSFL_00050-37_0000018_1278_0 using dsfl version 625 in slot 6 29 World Community Grid 2/8/2013 10:28:36 PM Restarting task c4cw_target07_013051813_0 using c4cw version 641 in slot 3 30 World Community Grid 2/8/2013 10:28:36 PM Restarting task DSFL_00050-37_0000021_1118_0 using dsfl version 625 in slot 2 31 World Community Grid 2/8/2013 10:28:36 PM Restarting task DSFL_00050-37_0000022_0763_0 using dsfl version 625 in slot 4 32 World Community Grid 2/8/2013 10:28:36 PM Restarting task DSFL_00050-37_0000022_0950_0 using dsfl version 625 in slot 0 33 World Community Grid 2/8/2013 10:28:36 PM Restarting task c4cw_target07_013590213_0 using c4cw version 641 in slot 7 34 World Community Grid 2/8/2013 10:28:36 PM Restarting task DSFL_00050-37_0000028_0899_0 using dsfl version 625 in slot 5 35 World Community Grid 2/8/2013 10:28:36 PM Sending scheduler request: To fetch work. 36 World Community Grid 2/8/2013 10:28:36 PM Requesting new tasks for CPU 37 World Community Grid 2/8/2013 10:28:39 PM Scheduler request completed: got 0 new tasks 38 World Community Grid 2/8/2013 10:28:39 PM No tasks sent 39 World Community Grid 2/8/2013 10:28:39 PM No tasks are available for The Clean Energy Project - Phase 2 40 World Community Grid 2/8/2013 10:28:39 PM No tasks are available for the applications you have selected. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
hmmm, 64 bits sciences running, when CEP2 is 32 bits, also being one explanation of the "there's work for other platforms" message. If the system is pure 64 bit,s you'd have to install the appropriate 32 bit libs.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
That was it. Went back and looked at the one LINUX system that was getting CEP and the 32-bit libs were there, probably as a dependency to something else that was installed on that machine. Installed 32-bit libs elsewhere and now those systems are downloading CEP units as well. Thanks for the insight..
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Great. Thinking of some message the server could transmit to point to the 32 bit issue [Love message/event logs to read what's the makeup of a "victims" device, and not the censored single line ones :D].
I've just expanded the relevant FAQ on this matter to add this issue ** [Who's reading them?], which when FAAH goes live with the new engine version makes it 6 sciences that are available in 64 bit, and since CEP2 is also due for an Q-Chem engine upgrade, who knows will that one too get that extra... certainly the bigger registers could help performance. ** Only since server 700 do we have an app level check, where before a 64 bit only device could fetch 32 bit sciences and spit out a error message which pointed to the matter of the 32 bit libs. |
||
|
|
|