I keep running out of work.

Sending scheduler request: To fetch work.
Requesting new tasks for GPU
Scheduler request completed: got 0 new tasks
No tasks sent Tasks for CPU are available, but your preferences are set to not accept them

I have this on both my machines and can see no reason for no work being sent.

Hi,

Ah, looking at scheduler log, it seems scheduler doesn't find anything suitable for your host. We seem to have too many small work units so that could mean there are huge work units in the memory area where scheduler looks for work.

I've at least temporarily increased that area so you should hopefully get more work next time.

-w

Have the same issue. First time for moo wrapper! too

same problem here,also not receiving work for any other project im attached to as well

same problem here,also not receiving work for any other project im attached to as well

John I think your problem is all the errors you are returning, look here this is your pc: http://moowrap.net/results.php?hostid=2751 It says you are returning 576 and 8 invalid units and that is just what is knows about recently. For every unit you return that is not 'valid' you get one less unit the next days allotment of units. Eventually you get down to the point of not being able to get any units at all, which is the point because your pc needs tuning to return 'valid' work. Are you overclocking by chance? Did you recently increase the amount of memory or even swap sticks around? It also could be as simple as you 'resetting' the Project to ensure you get the proper files as they could be corrupted. Resetting WILL dump everything you have so only do that if you HAVE TO!

Hi,

The original problem of not having anything to send by the scheduler should be fixed by the work generation fixes I did to handle fragmentation (huge packet counts in our work).

Is anybody still having problems with fetching work?

-w

I have 7 systems hooked up to Moo and I'm not getting anymore work. Mix of ATI/Nvidia GPU's.

I went from 1.5 mil/day to almost nothing. Did I hit some max WUs or something?

I have 7 systems hooked up to Moo and I'm not getting anymore work. Mix of ATI/Nvidia GPU's.

I went from 1.5 mil/day to almost nothing. Did I hit some max WUs or something?

Do you crunch for other gpu projects too? If so you could be full of them and to now get some Moo units would put you into the 'high priority' crunching mode and Boinc does not like to do that. Some of your pc's do have work, but some do not.

1/8/2012 3:40:20 AM | Moo! Wrapper | Sending scheduler request: To fetch work.
1/8/2012 3:40:20 AM | Moo! Wrapper | Requesting new tasks for ATI
1/8/2012 3:40:25 AM | Moo! Wrapper | Scheduler request completed: got 0 new tasks
1/8/2012 3:40:25 AM | Moo! Wrapper | No tasks sent


Same issue for this PC. Just upgraded it to a 6990 as well. I removed all other projects and only have Moo! running. I've tried completely uninstalling and still won't pull work.

Just got 2 WU's. Sitting in Ready to Start.

1/8/2012 3:51:46 AM | Moo! Wrapper | [cpu_sched_debug] insufficient ATI for dnetc_r72_1325970819_362_788_0
1/8/2012 3:51:46 AM | Moo! Wrapper | [cpu_sched_debug] insufficient ATI for dnetc_r72_1325970769_456_768_0



So on a whim, instead of fully uninstalling, I just installed 6.12.34 over top of 7.0.7:

O yeah and added a 5870 to the rig.

1/8/2012 4:26:00 AM | | Version change (7.0.7 -> 6.12.34)
1/8/2012 4:26:00 AM | Moo! Wrapper | URL http://moowrap.net/; Computer ID 8036; resource share 500
1/8/2012 4:26:00 AM | Moo! Wrapper | General prefs: from Moo! Wrapper (last modified 31-Dec-2011 12:33:37)
1/8/2012 4:26:00 AM | Moo! Wrapper | Host location: none
1/8/2012 4:26:00 AM | Moo! Wrapper | General prefs: using your defaults
1/8/2012 4:26:00 AM | | Reading preferences override file
1/8/2012 4:26:00 AM | | Preferences:
1/8/2012 4:26:00 AM | | max memory usage when active: 6141.91MB
1/8/2012 4:26:00 AM | | max memory usage when idle: 7370.29MB
1/8/2012 4:26:00 AM | | max disk usage: 10.00GB
1/8/2012 4:26:00 AM | | max CPUs used: 4
1/8/2012 4:26:00 AM | | (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
1/8/2012 4:26:00 AM | | [cpu_sched] Request CPU reschedule: Prefs update
1/8/2012 4:26:00 AM | | [cpu_sched] Request CPU reschedule: Startup
1/8/2012 4:26:00 AM | | Not using a proxy
1/8/2012 4:26:00 AM | | Running CPU benchmarks
1/8/2012 4:26:00 AM | | [cpu_sched] Request CPU reschedule: Idle state change
1/8/2012 4:26:00 AM | | Suspending computation - CPU benchmarks in progress
1/8/2012 4:26:31 AM | | Benchmark results:
1/8/2012 4:26:31 AM | | Number of CPUs: 4
1/8/2012 4:26:31 AM | | 2739 floating point MIPS (Whetstone) per CPU
1/8/2012 4:26:31 AM | | 9411 integer MIPS (Dhrystone) per CPU
1/8/2012 4:26:35 AM | Moo! Wrapper | Sending scheduler request: To fetch work.
1/8/2012 4:26:35 AM | Moo! Wrapper | Requesting new tasks for ATI GPU
1/8/2012 4:26:37 AM | Moo! Wrapper | Scheduler request completed: got 1 new tasks
1/8/2012 4:26:37 AM | Moo! Wrapper | Resent lost task dnetc_r72_1325970769_456_768_0
1/8/2012 4:26:37 AM | Moo! Wrapper | Didn't resend lost task dnetc_r72_1325970819_362_788_0 (expired)
1/8/2012 4:26:37 AM | Moo! Wrapper | Didn't resend lost task dnetc_r72_1325383261_9_9_1 (expired)
1/8/2012 4:26:37 AM | Moo! Wrapper | Didn't resend lost task dnetc_r72_1325383119_9_9_1 (expired)
1/8/2012 4:26:37 AM | | [cpu_sched] Request CPU reschedule: periodic CPU scheduling
1/8/2012 4:26:37 AM | | [cpu_sched] schedule_cpus(): start
1/8/2012 4:26:37 AM | | [cpu_sched] enforce_schedule(): start
1/8/2012 4:26:37 AM | | [cpu_sched] preliminary job list:
1/8/2012 4:26:37 AM | | [cpu_sched] final job list:
1/8/2012 4:26:37 AM | | [cpu_sched] using 0.00 out of 4 CPUs
1/8/2012 4:26:37 AM | | [cpu_sched] enforce_schedule: end
1/8/2012 4:26:39 AM | Moo! Wrapper | File dnetc_1.02_windows_intelx86__ati14.exe exists already, skipping download
1/8/2012 4:26:39 AM | Moo! Wrapper | File dnetc518-win32-x86-stream.exe exists already, skipping download
1/8/2012 4:26:39 AM | Moo! Wrapper | File dnetc-1.00.ini exists already, skipping download
1/8/2012 4:26:39 AM | Moo! Wrapper | File job-ati14-1.00.xml exists already, skipping download
1/8/2012 4:26:39 AM | Moo! Wrapper | File dnetc_r72_1325970769_456_768 exists already, skipping download
1/8/2012 4:26:39 AM | | [cpu_sched] Request CPU reschedule: files downloaded
1/8/2012 4:26:39 AM | | [cpu_sched] schedule_cpus(): start
1/8/2012 4:26:39 AM | Moo! Wrapper | [cpu_sched] scheduling dnetc_r72_1325970769_456_768_0 (coprocessor job, FIFO)
1/8/2012 4:26:39 AM | | [cpu_sched_debug] reserving 3.000000 of coproc ATI
1/8/2012 4:26:39 AM | | [cpu_sched] enforce_schedule(): start
1/8/2012 4:26:39 AM | | [cpu_sched] preliminary job list:
1/8/2012 4:26:39 AM | Moo! Wrapper | [cpu_sched] 0: dnetc_r72_1325970769_456_768_0 (MD: no; UTS: no)
1/8/2012 4:26:39 AM | | [cpu_sched] final job list:
1/8/2012 4:26:39 AM | Moo! Wrapper | [cpu_sched] 0: dnetc_r72_1325970769_456_768_0 (MD: no; UTS: no)
1/8/2012 4:26:39 AM | Moo! Wrapper | [coproc] Assigning ATI instance 0 to dnetc_r72_1325970769_456_768_0
1/8/2012 4:26:39 AM | Moo! Wrapper | [coproc] Assigning ATI instance 1 to dnetc_r72_1325970769_456_768_0
1/8/2012 4:26:39 AM | Moo! Wrapper | [coproc] Assigning ATI instance 2 to dnetc_r72_1325970769_456_768_0
1/8/2012 4:26:39 AM | Moo! Wrapper | [cpu_sched] scheduling dnetc_r72_1325970769_456_768_0
1/8/2012 4:26:39 AM | | [cpu_sched] using 0.05 out of 4 CPUs
1/8/2012 4:26:39 AM | Moo! Wrapper | [cpu_sched] dnetc_r72_1325970769_456_768_0 sched state 0 next 2 task state 0
1/8/2012 4:26:39 AM | Moo! Wrapper | Starting task dnetc_r72_1325970769_456_768_0 using dnetc version 102
1/8/2012 4:26:39 AM | | [cpu_sched] enforce_schedule: end

1/8/2012 3:51:46 AM | Moo! Wrapper | [cpu_sched_debug] insufficient ATI for dnetc_r72_1325970819_362_788_0
...
O yeah and added a 5870 to the rig.

Hmm, this inssufficient ATI could happen if you remove a card from your system when you still have work left that are marked to require the previous number of cards. Any new (or resent) work should get the new number of cards as requirement so the BOINC Client can start it. It's our scheduler that tells this and it bases it's decision on the number of cards reported to it by the BOINC Client.

So if somebody removes (or adds, for that matter) GPUs to their system, I'd suggest crunching all work you have before changes or a project reset afterwards to reset the card requirements. Note that you will loose any partially crunched or non-reported work in a reset so a forced reporting is highly recommended before it.

-w

yes im over clocked on both the cpu and gpu.cpu 3.3 from 3.0,and the gpu930,from 850.not having heating issues,cpu runs about 115 f,gpu runs 57 c.think it may have something to do with bm im running,7.0.12.been runnig that for a week ,i think,havent had any problems till today,not getting any work for any projects again.if i run bm 6.12.34,dosent have any problems getting work,for anything.

yes im over clocked on both the cpu and gpu.cpu 3.3 from 3.0,and the gpu930,from 850.not having heating issues,cpu runs about 115 f,gpu runs 57 c.think it may have something to do with bm im running,7.0.12.been runnig that for a week ,i think,havent had any problems till today,not getting any work for any projects again.if i run bm 6.12.34,dosent have any problems getting work,for anything.

The version 7 series of Boinc handles work requests MUCH different that the older version of Boinc, they are almost totally different!! Under Network Usage there are two setting...one for "connect about every" and one for "additional work buffer". They are NOT treated as they say, they treated as hard count numbers now!! So if you put 1.25 in the 2nd one, "additional work buffer" you are telling Boinc to NOT connect to a project for 1.25 days! It is a learning curve and you MUST work with it to get it working like you want it to. Remember programmers think they KNOW what you want and give it to you whether you really want it or not! They also have stuff they are working towards that the rest of us have no clue about.

I was looking for, and couldn't find until just now, the thread detailing the problems, it is here: http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,30185_offset,240

Ignore the WCG specific stuff. SekeRob is the poster and he posted it at Jan 30, 2012 7:13:59 AM.

A key thing to note re BOINC V7.XX, it works fundamentally differently, it really is a Major change in BOINCLand, driven (in the main), by VM Box and OpenCL needs. It deals with the files and related information in such a way that it is not compatible with 6.XX. If you go 7.XX, you stay ... you cant go back without trashing all held data on BOINC on all projects - including WUs - there is no "I'll just roll back" and keep the contents on the ProgramData/BOINC directory - not happening - if you roll back you have to be prepared to dump the lot, and start again :)

So those reading the thread, have a think .... because there is no going back if you end up hating the current state of 7.XX after trying it - it is at present an Alpha Developer version of BOINC.

Regards
Zy

well got work again.seems it just ran me down to nothing,then slammed me with a couple hundred huge wu.feast or famine i guess.

well got work again.seems it just ran me down to nothing,then slammed me with a couple hundred huge wu.feast or famine i guess.

Yes that is Boinc 7 at its best! Play with the numbers and it WILL get better. Small cache's are best too.

i belive i have it set at 3 days,more than enough to keep me busy.oddly though,there all due the same day within 3 seconds of each other.crunch,crunch,crunch.

i belive i have it set at 3 days,more than enough to keep me busy.oddly though,there all due the same day within 3 seconds of each other.crunch,crunch,crunch.

As I said some of those are HARD numbers now, meaning you say 3 days, it DOES NOT connect for 3 days, even if you are out of work!! So keep an eye on it, things ARE changing!!!