Hi all,

I bought a new laptop last month for gaming, Dell XPS M1730, but between Lans I have it crunching away. It has Vista busines on it and it's been fine.

The reason I am posting is that for the last couple days I don't seem to be getting any work units. Looking at the messages tab in boinc there aren't any errors. When I do a manual update it says "Scheduler request succeeded: got 0 new tasks"

Can't seem to see anything on the site indicating problems so just wondered if there was anything up that might prevent me getting new units.

MindModeling is the only project I run on the laptop so it has 100% cpu time.

Thanks

Just wondering if anything is wrong ? Still not had any work for about a week now.

Just wondering if anything is wrong ? Still not had any work for about a week now.

Just wondering if anything is wrong ? Still not had any work for about a week now.

OBVIOUSLY something is wrong: I reinitiated processing MM over 40 hours ago and initially downloaded over 700 units to one quadcore machine. I subsequently downloaded only ONE unit of work on one of three other computers with a combined 8 cpus available. I have been unable to download any further work units for some 20 hours and counting. What I find troubling is no one else has filed a notice of deficiency that they too cannot acquire work to perform. I have tried everything I know to do to FORCE the machines into downloading work to process but to no avail. The point is I absolutely do not believe that my computers and the preferences established for them has anything to do with the lack of work to crunch.

Team Boinc Synergy is currently voting for a "project of the month" to begin 1 May and it is evident to me at this point that MM cannot handle the task. Kindly respond so that I can make an input to the team forum as whether we should select another project to support.

What else I find troubling is that a number of messages requesting assistance in successfully processing MM work units appear to never have been answered.

Bill

Hi:

I have a similar problem. I had not written anything in the forum because my English is very bad (sorry for that) and I did not dare.

Approximately 1 weeks ago I decided to devote my time to MM with 5 of my machines. These are:

3 machines with cpu P8400
1 machine with cpu Q6600
1 machine with cpu i7 920

For 2 or 3 days, unloaded all my machines work normally, but after that no longer download any of them work. However, I see that server_status.php indicates that there are several thousands of jobs waiting to be sent.

Does anyone know what the problem is?


Thanks.

Hello Rafa: The problem of being unable to download work units appears to have been corrected by the administrators. Two of my machines downloaded normally about 2 hours ago, but I initiated "RESET" on the other two before downloads occurred; however, I had tried this technique several times earlier without success. You might try Resetting the project on each computer if you are still unable to download work to process.

I hope this helps; and good luck.

Bill from Austin, Texas USA

I

Hi Billy:

It was not necessary to reset the project. Tonight, when re-check the machines, had downloaded all new work. 5 again be operational.

Thank you very much for the information.

Greetings from Cordoba, Spain.

It is under our current belief that there is a bandwidth issue when dispensing workunits. Under some occassions the assimilator or transitioner may be down, but it seems as of late that all our boinc daemons may be running and work is ready to send but users still are not getting the work they needed from the scheduler.

As of right now, if for some reason you are not getting any workunits, please try the following steps:

1) Update the project
2) Reset the project
3) Detach and Re-attach
4) Download the latest version of BOINC software

If workunits are still not being scheduled to your computer, please let me know and I will personally look into it.

I just added 2 of my 5 boxes to crunching MM; been waiting for the WU's to stop "Downloading" for 27 hours now on one box and can't get any work on the 2nd box. Just DLed the latest Ver of BOINC and reattachecd to MM, still no luck!

Change 1: As I was writing this the box corrected itself and has begun DLing WU's. Still have to check the E6700 Box yet.
____________
JoeBnKY; JoeB@KY[/b]

Acording to the website there should be Mega Work in this job. So where is it? The server says "O" "WU's to send. MM is our Project of the Month at BOINC Synergy this month and there is no work all of a sudden.
____________
JoeBnKY; JoeB@KY[/b]

I've got a question regarding work, and also to report something strange going on with the 0 work available problem. On the two machines I have mm on the only major difference is one has an Intel Core2Duo and the other an AMD Athlon FX. Though the AMD has a lot more resources (in terms of available memory and disk space) both processors have 2 cores and clocked at 2800 Mhz (Intel) and 3100 Mhz for the AMD, both have 64 bit OS's and clients installed.

My AMD machine has been able to continuously get work ever since I was able to download those 4 key files. This is from the log just a little bit ago:

5/10/2009 2:24:03 AM|MindModeling@Beta|Sending scheduler request: To fetch work
5/10/2009 2:24:03 AM|MindModeling@Beta|Requesting 147637 seconds of new work, and reporting 9 completed tasks
5/10/2009 2:24:08 AM|MindModeling@Beta|Scheduler RPC succeeded [server version 603]
5/10/2009 2:24:08 AM|MindModeling@Beta|Deferring communication for 7 sec
5/10/2009 2:24:08 AM|MindModeling@Beta|Reason: requested by project
5/10/2009 2:24:10 AM|MindModeling@Beta|[file_xfer] Started download of file SASTNM_11_2-3-1750_1241937617

After which there is a page full of file_xfer's in my messages log. My work queue always has between 100-150 tasks at any given time. Which leads to my question, since one machine is pulling massive amounts of tasks continuously can I copy them over to my other machine to split up the work since the other one doesn't download anything?

Most of the SASTNM jobs are wrapping up, so very few workunits are being created for those jobs. The remaining models, the RPI family, contain workunits which compute a whooping 5,000 parameter sets each (for a grand total of 16 million paramters sets). It takes a lot of overhead to resubmit, parse, and upload each point into data that we can analyze. Future jobs of this type may not be used for our volunteer resources because they incur too much overhead and clog our resources.

Actions are being taken currently to offload some of the work and redistribute WU's to volunteers.

As for your question of copying workunits from one box to another, I am not sure whether that is supported by the current BOINC infastrucure or not. I will post it as a question to other BOINC developers/users via an email list and see if I get any responses.

I attached my PC to MindModeling Beta but after downloading the first two workunits it hasn't downloaded anything else. My task analysis online shows the uploaded results in "outcome" and "client state" as error. Is there a problem with the project at the moment?

I just switched from the incompetent idiots of Artificial Intelligence Realm who shut down the entire project without even posting a formal communication or warning us beforehand. Please don't tell me this is happening here as well.

So... anything to crunch?!

No, the project hasn't mysteriously shut down. We recently switched to a new database server two weeks ago and have been busy configuring all the settings for file managemnet, security, and permissions. We plan to have more work today and at the latest tomorrow night :)

I'll update the news feed when this occurs.

I tried updating, resetting, and attaching/detaching. I am running BOINC 6.6.46 on a 64-bit Vista laptop. I have not gotten work in at least a month.

I have 6.2.28 for 32-bit systems, i see there is a new work on servers, but i do not receive nothing.

We will be submitting much larger tasks to the volunteers in a day or two, so workunits will be spread across a much larger range of users.

Ok, but what does it mean, that on the main page the percents of "SAST_Reduced6" and "RPI_Split10" are increasing, but my PC and my laptop do not receive the tasks?

Here what i have:

26.06.2009 21:12:58 MindModeling@Beta Sending scheduler request: To fetch work.
26.06.2009 21:12:58 MindModeling@Beta Requesting new tasks
26.06.2009 21:13:07 Project communication failed: attempting access to reference site
26.06.2009 21:13:08 MindModeling@Beta Scheduler request failed: Failure when receiving data from the peer
26.06.2009 21:13:19 Internet access OK - project servers may be temporarily down.


But on the main page i see, that the servers are working.