Message boards : Number crunching : Project has no tasks available
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · Next

AuthorMessage
Jim1348

Send message
Joined: 28 Feb 15
Posts: 253
Credit: 200,562,581
RAC: 0
Message 3358 - Posted: 13 Feb 2019, 17:17:44 UTC - in response to Message 3357.  
Last modified: 13 Feb 2019, 18:09:45 UTC

No, it isn't normal. My veteran hosts are also being told there are no tasks available.

I am seeing it more frequently on various projects where I shouldn't be seeing it. Maybe there was a change to the BOINC server code?
There was some mention of that on another project, but I don't have it at the moment. I think it straightens itself out eventually, but we will see.

I should point out that I have been changing work around between machines, so it increases the chances of something going wrong. But BOINC should never leave cores free, and it is doing that with some regularity. I think it is temporary though.
ID: 3358 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Brummig
Avatar

Send message
Joined: 23 Mar 16
Posts: 95
Credit: 23,431,842
RAC: 25
Message 3360 - Posted: 14 Feb 2019, 8:49:18 UTC - in response to Message 3358.  
Last modified: 14 Feb 2019, 8:50:28 UTC

In my case BOINC isn't leaving cores idling, it's just getting no tasks for U@H and consequently dedicating the cores to the other project(s) active on the host:

14/02/2019 01:00:00 | Universe@Home | Sending scheduler request: To fetch work.
14/02/2019 01:00:00 | Universe@Home | Requesting new tasks for CPU

14/02/2019 01:00:03 | Einstein@Home | Finished upload of p2030.20170413.G36.05+01.86.N.b6s0g0.00000_1940_1_0
14/02/2019 01:00:03 | Einstein@Home | Finished upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3044_1_0
14/02/2019 01:00:03 | Einstein@Home | Started upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3047_1_0
14/02/2019 01:00:03 | Einstein@Home | Started upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3033_1_0
14/02/2019 01:00:03 | Universe@Home | Scheduler request completed: got 0 new tasks
14/02/2019 01:00:03 | Universe@Home | Project has no tasks available

14/02/2019 01:00:05 | Einstein@Home | Finished upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3047_1_0
14/02/2019 01:00:05 | Einstein@Home | Finished upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3033_1_0
14/02/2019 01:00:05 | Einstein@Home | Started upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3048_1_0
14/02/2019 01:00:07 | Einstein@Home | Finished upload of p2030.20170413.G36.17+01.63.N.b5s0g0.00000_3048_1_0
14/02/2019 01:00:08 | Einstein@Home | Sending scheduler request: To fetch work.
14/02/2019 01:00:08 | Einstein@Home | Reporting 5 completed tasks
14/02/2019 01:00:08 | Einstein@Home | Requesting new tasks for CPU
14/02/2019 01:00:15 | Einstein@Home | Scheduler request completed: got 5 new tasks


Sometimes one of my hosts does manage to get one or two new U@H tasks, but I am now down to three U@H tasks on five hosts.
ID: 3360 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Shadak

Send message
Joined: 19 Sep 18
Posts: 20
Credit: 12,132,575
RAC: 0
Message 3361 - Posted: 14 Feb 2019, 10:32:43 UTC

all my host are getting work, but it should be a bit more. running Asteroids, einstein and universe at 5:5:10 ratio and getting 5:5:5 work...
ID: 3361 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin

Send message
Joined: 2 Jun 16
Posts: 169
Credit: 317,253,046
RAC: 6
Message 3363 - Posted: 14 Feb 2019, 11:29:29 UTC

What is your queue. I have 0.5 days or 0.25 days on my PCs and they are getting work. One thing I have noticed since the server update is that multiple tasks pile up before being reported. Like 6-8 sometimes on a 32c machine. I don't recall that before but no real proof. Maybe a small buffer and a longer reporting time dries up your queue?
ID: 3363 · 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 3364 - Posted: 14 Feb 2019, 15:45:33 UTC

I am now getting the dreaded "Project has no tasks available" again for no reason. Even if I suspend my other project (Einstein) entirely, I get that message.

I think it is a problem at Universe rather than BOINC in general, but who knows.
ID: 3364 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Shadak

Send message
Joined: 19 Sep 18
Posts: 20
Credit: 12,132,575
RAC: 0
Message 3365 - Posted: 14 Feb 2019, 17:21:03 UTC

I have a 0 day cache ^^
ID: 3365 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin

Send message
Joined: 2 Jun 16
Posts: 169
Credit: 317,253,046
RAC: 6
Message 3366 - Posted: 14 Feb 2019, 18:15:44 UTC

Ah, maybe the period where I have several tasks waiting to report is where I might be getting the No tasks available but since I have a queue I never run out. All projects take a bit of time to Upload/download tasks so having no queue you'll end up with periods of no crunching on a thread or two.
ID: 3366 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin

Send message
Joined: 2 Jun 16
Posts: 169
Credit: 317,253,046
RAC: 6
Message 3367 - Posted: 15 Feb 2019, 0:14:38 UTC

I checked once I got home.
Universe@Home 2/14/2019 7:04:24 PM Project has no tasks available

But 9 minutes later I received 5 tasks. Pretty much just need a queue.
ID: 3367 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Brummig
Avatar

Send message
Joined: 23 Mar 16
Posts: 95
Credit: 23,431,842
RAC: 25
Message 3368 - Posted: 15 Feb 2019, 8:49:50 UTC - in response to Message 3367.  

No, that argument about having a queue doesn't stack up. I have a 1 day cache on my Pis. One that has been told for some time now that U@H has no tasks available suddenly received ten tasks last night. Meanwhile, the one that managed to get two tasks the night before last reported them last night, and is now being told the project has no tasks available:

15/02/2019 01:00:27 | Universe@Home | Sending scheduler request: To fetch work.
15/02/2019 01:00:27 | Universe@Home | Reporting 2 completed tasks
15/02/2019 01:00:27 | Universe@Home | Requesting new tasks for CPU
15/02/2019 01:00:29 | Universe@Home | Scheduler request completed: got 0 new tasks
15/02/2019 01:00:29 | Universe@Home | Project has no tasks available
15/02/2019 01:10:39 | Universe@Home | Sending scheduler request: To fetch work.
15/02/2019 01:10:39 | Universe@Home | Requesting new tasks for CPU
15/02/2019 01:10:42 | Universe@Home | Scheduler request completed: got 0 new tasks
15/02/2019 01:10:42 | Universe@Home | Project has no tasks available
ID: 3368 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin

Send message
Joined: 2 Jun 16
Posts: 169
Credit: 317,253,046
RAC: 6
Message 3369 - Posted: 15 Feb 2019, 11:41:47 UTC

You're not following. I have a queue and still get it but it allows a PC to have something to do instead of running dry.
ID: 3369 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Brummig
Avatar

Send message
Joined: 23 Mar 16
Posts: 95
Credit: 23,431,842
RAC: 25
Message 3370 - Posted: 15 Feb 2019, 12:26:56 UTC - in response to Message 3369.  

Well my queue is twice the length of your queue, but since my hosts are told for very long periods of time (days or weeks) by U@H that there are no tasks available, even with that queue they run dry. And then when they do get tasks, it may only be one or two, so the queue doesn't get refilled.
ID: 3370 · 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 3371 - Posted: 15 Feb 2019, 12:38:33 UTC

This happens always when we have loads (more than 1kk) tasks generated. It will server better when amount of tasks available drops to lover level.
Unfortunately I had to do other tasks and created more tasks to not dry server.
Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT
ID: 3371 · 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 3372 - Posted: 15 Feb 2019, 14:03:53 UTC - in response to Message 3371.  

OK, it is good to know the cause. We can live with it and buffer accordingly.
ID: 3372 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Greg_BE

Send message
Joined: 24 Nov 17
Posts: 26
Credit: 18,746,503
RAC: 40
Message 3375 - Posted: 19 Feb 2019, 20:17:23 UTC
Last modified: 19 Feb 2019, 20:19:52 UTC

I don't understand, if you have so many tasks ready to go, then why is your server locking us out?
That's the most conflicted logic I have ever encountered.

If you have an overload of work and your server can't handle it, then you should take some of that work and store it elsewhere until the server has depleted the tasks it has to a sufficient level that you can add more.

Meanwhile a lot of us out here are seeing our credits goto 0 with your project and then when we go get work, your project then hogs all the resources of our system to get caught back up in BOINC with credit.

This is crazy!

Tasks ready to send 2,260,206
Tasks in progress 82,542

That is so lopsided its crazy! You should have in theory about 90,000 or so in progress. I am sitting here with no work for weeks now. How many tasks could I have processed by now?
ID: 3375 · 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 3380 - Posted: 20 Feb 2019, 13:10:47 UTC

I am getting "no tasks available again. It worked for a few hours yesterday, and then I got one task again this morning, and now I am almost out of work.
ID: 3380 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
additude

Send message
Joined: 6 Feb 19
Posts: 41
Credit: 131,325,319
RAC: 0
Message 3382 - Posted: 20 Feb 2019, 15:43:57 UTC

No problems here....
I guess I buffer accordingly...
ID: 3382 · 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 3384 - Posted: 20 Feb 2019, 16:29:33 UTC - in response to Message 3380.  

It is back.
I hope the new funding buys more servers or whatever else is needed.
ID: 3384 · 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 3385 - Posted: 20 Feb 2019, 19:14:19 UTC - in response to Message 3382.  
Last modified: 20 Feb 2019, 19:17:01 UTC

No problems here....
I guess I buffer accordingly...

If it is a server problem at their end, adding more to a buffer would not increase the total output for the project. It would just send more work to one user, and prevent it from going to another,

They need to see where the bottleneck is, and add more capacity as necessary.
ID: 3385 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Greg_BE

Send message
Joined: 24 Nov 17
Posts: 26
Credit: 18,746,503
RAC: 40
Message 3387 - Posted: 20 Feb 2019, 20:37:25 UTC - in response to Message 3382.  

I have 1 day of work and buffer 1 day extra so I can rotate through all the projects I participate in, in an even fashion. But U@H seems to just want to say no work. I could buffer 2 days extra and get nothing.
ID: 3387 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Retnek
Avatar

Send message
Joined: 26 Sep 16
Posts: 3
Credit: 264,690,611
RAC: 24
Message 3389 - Posted: 20 Feb 2019, 21:39:32 UTC

Can confirm this kind problems. My PCs during the last 2 weeks maybe got 20 - 30 % of what would have been possible. I was thinking about a kind of gate limiting the number of jobs given to certain IP-adress, f.e. But during the last 2 days the already thin water-jet became a drippy misery.
ID: 3389 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · Next

Message boards : Number crunching : Project has no tasks available




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