Welcome to MilkyWay@home

Server Trouble

Message boards : News : Server Trouble
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 22 · Next

AuthorMessage
Profile Wrend
Avatar

Send message
Joined: 4 Nov 12
Posts: 96
Credit: 251,528,484
RAC: 0
Message 72627 - Posted: 9 Apr 2022, 4:35:31 UTC - in response to Message 72625.  
Last modified: 9 Apr 2022, 5:27:53 UTC

Thanks for the gpu longevity answers, I'll stick with more worry free CPU's only. That's one thing I like about World Community Grid, it's CPU only.

Sorry, I wasn't meaning to scare you off. I thought you were maybe wanting to optimize their total work potential. I'm running 4 MW@H tasks per GPU and could run up to 5 or 6 if I wanted to fully load them up, but fewer tasks per GPU shouldn't load them up too much to worry about either way beyond needing to clean dust off a bit more often, if you have your computer on and are using it anyway. I can generally run 2 tasks per GPU and play games at the same time without issue, since I use V-sync or frame rate limits in the video games for 60 FPS. 1 task per GPU wouldn't be much above idle for me, expect for keeping the clock rates up.

As Peter said above, GPUs are often much more powerful at crunching than CPUs for the work units that can make use of them. You can pretty easily tell from my computer's BOINC stats in my signature which projects I've used the GPUs or CPU on. Currently I'm running 8 MW@H tasks total on the GPUs, and 8 Einstein task on the CPU. https://i.imgur.com/YgGl6Zs.png
ID: 72627 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Septimus

Send message
Joined: 8 Nov 11
Posts: 205
Credit: 2,882,881
RAC: 267
Message 72628 - Posted: 9 Apr 2022, 6:13:38 UTC
Last modified: 9 Apr 2022, 6:18:29 UTC

Still no Nbody Simulations validated, all in Validation Inconclusive. All completed on my machine but status showing as unsent ?
ID: 72628 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72629 - Posted: 9 Apr 2022, 8:54:53 UTC

Uh oh, very slow server responses again. [sending adrenaline patch]
ID: 72629 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
macgeyer

Send message
Joined: 2 Mar 18
Posts: 9
Credit: 457,043,383
RAC: 0
Message 72630 - Posted: 9 Apr 2022, 9:24:04 UTC

Impossible to get any WUs for severall days now ! when will this project run again as it should ?

09/04/2022 11:22:48 | Milkyway@Home | update requested by user
09/04/2022 11:22:50 | Milkyway@Home | Sending scheduler request: Requested by user.
09/04/2022 11:22:50 | Milkyway@Home | Requesting new tasks for AMD/ATI GPU
09/04/2022 11:22:51 | Milkyway@Home | Scheduler request completed: got 0 new tasks
09/04/2022 11:22:51 | Milkyway@Home | Project requested delay of 91 seconds
ID: 72630 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72632 - Posted: 9 Apr 2022, 9:43:24 UTC - in response to Message 72630.  

Impossible to get any WUs for severall days now ! when will this project run again as it should ?

09/04/2022 11:22:48 | Milkyway@Home | update requested by user
09/04/2022 11:22:50 | Milkyway@Home | Sending scheduler request: Requested by user.
09/04/2022 11:22:50 | Milkyway@Home | Requesting new tasks for AMD/ATI GPU
09/04/2022 11:22:51 | Milkyway@Home | Scheduler request completed: got 0 new tasks
09/04/2022 11:22:51 | Milkyway@Home | Project requested delay of 91 seconds
When Tom buys SSDs.

In the meantime I set Windows Task scheduler to ask for work every 2 minutes, I have my GPUs running 80% of the time.
ID: 72632 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Septimus

Send message
Joined: 8 Nov 11
Posts: 205
Credit: 2,882,881
RAC: 267
Message 72636 - Posted: 9 Apr 2022, 11:54:41 UTC

Back to Collatz and Einstein. Will watch and see if things get back to normal.
ID: 72636 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
arcturus

Send message
Joined: 20 Nov 07
Posts: 54
Credit: 2,663,789
RAC: 0
Message 72639 - Posted: 9 Apr 2022, 14:38:35 UTC - in response to Message 72627.  

As Peter said above, GPUs are often much more powerful at crunching than CPUs for the work units that can make use of them. You can pretty easily tell from my computer's BOINC stats in my signature which projects I've used the GPUs or CPU on. Currently I'm running 8 MW@H tasks total on the GPUs, and 8 Einstein task on the CPU. https://i.imgur.com/YgGl6Zs.png

I'm not deterred, however once WCG is back up I'm outta here anyway so the issue is moot, plus I have a buyer for 5 decent vid cards gathering dust. The market is just too hot not to sell.
ID: 72639 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Robert Coplin

Send message
Joined: 23 Sep 13
Posts: 19
Credit: 36,217,133
RAC: 0
Message 72640 - Posted: 9 Apr 2022, 14:57:57 UTC - in response to Message 72639.  

When World Community Grid has new work units i will be going back to World Community Grid at least for CPU work units and i may stay here for GPU work units.World Community Grid in the past didnt have enough GPU work units for my liking except when they had the stress test
ID: 72640 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Skillz

Send message
Joined: 28 May 17
Posts: 76
Credit: 4,386,301,288
RAC: 0
Message 72641 - Posted: 9 Apr 2022, 15:23:57 UTC

PrimeGrid, GPUGrid, Einstein@home, Amicable Numbers, MLC@home all can utilize GPUs. So anyone with idling GPUs I would recommend joining one of those projects and setting the resource share to 0 on it. So when MW@Home comes back your systems will automatically switch to it.
ID: 72641 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72644 - Posted: 10 Apr 2022, 0:24:54 UTC - in response to Message 72639.  
Last modified: 10 Apr 2022, 0:25:33 UTC

As Peter said above, GPUs are often much more powerful at crunching than CPUs for the work units that can make use of them. You can pretty easily tell from my computer's BOINC stats in my signature which projects I've used the GPUs or CPU on. Currently I'm running 8 MW@H tasks total on the GPUs, and 8 Einstein task on the CPU. https://i.imgur.com/YgGl6Zs.png

I'm not deterred, however once WCG is back up I'm outta here anyway so the issue is moot, plus I have a buyer for 5 decent vid cards gathering dust. The market is just too hot not to sell.
Odd, I've been getting very cheap GPUs. They've dropped in price here in the UK.

P.S. why do only Americans use the word moot?
ID: 72644 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72645 - Posted: 10 Apr 2022, 0:26:33 UTC - in response to Message 72640.  

When World Community Grid has new work units i will be going back to World Community Grid at least for CPU work units and i may stay here for GPU work units.World Community Grid in the past didnt have enough GPU work units for my liking except when they had the stress test
They might have a burst because the scientists will be currently starved of results.
ID: 72645 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile alk44
Avatar

Send message
Joined: 2 Mar 20
Posts: 131
Credit: 315,530,379
RAC: 31,968
Message 72715 - Posted: 12 Apr 2022, 3:14:50 UTC

.No Separation tasks available
ID: 72715 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Keith Myers
Avatar

Send message
Joined: 24 Jan 11
Posts: 696
Credit: 540,058,660
RAC: 86,805
Message 72731 - Posted: 12 Apr 2022, 19:09:50 UTC - in response to Message 72715.  

.No Separation tasks available

There are plenty of Separation tasks. 21K currently.

You need to script a project update to get tasks every scheduler connection.

If BOINC receives a no task reply for a scheduler connection, it immediately backs off for 1 hour, then progressively adds more backoff hours to get to 24 hours.

If you don't intercede, no wonder you aren't getting work.
ID: 72731 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Spatzthecat

Send message
Joined: 1 Dec 10
Posts: 82
Credit: 15,452,009,012
RAC: 2
Message 72735 - Posted: 13 Apr 2022, 0:37:00 UTC - in response to Message 72731.  

.No Separation tasks available

There are plenty of Separation tasks. 21K currently.

You need to script a project update to get tasks every scheduler connection.

If BOINC receives a no task reply for a scheduler connection, it immediately backs off for 1 hour, then progressively adds more backoff hours to get to 24 hours.

If you don't intercede, no wonder you aren't getting work.

?????
ID: 72735 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Keith Myers
Avatar

Send message
Joined: 24 Jan 11
Posts: 696
Credit: 540,058,660
RAC: 86,805
Message 72736 - Posted: 13 Apr 2022, 1:05:05 UTC

watch -n 720 ./boinccmd --project http://milkyway.cs.rpi.edu/milkyway/ update or the equivalent in Windows Task Scheduler
ID: 72736 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72739 - Posted: 13 Apr 2022, 3:58:03 UTC - in response to Message 72735.  

.No Separation tasks available

There are plenty of Separation tasks. 21K currently.

You need to script a project update to get tasks every scheduler connection.

If BOINC receives a no task reply for a scheduler connection, it immediately backs off for 1 hour, then progressively adds more backoff hours to get to 24 hours.

If you don't intercede, no wonder you aren't getting work.

?????
The server is broken, nobody knows why, sometimes it can't be bothered giving you tasks even though they're there in the queue. For some reason you hid your computers so I can't tell what OS to advise you on.
ID: 72739 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Ralph Little

Send message
Joined: 30 Jul 16
Posts: 6
Credit: 58,669,504
RAC: 0
Message 72741 - Posted: 13 Apr 2022, 6:44:19 UTC
Last modified: 13 Apr 2022, 6:45:22 UTC

Same with me. Gets GPU work from time to time but mostly the machine is idle, despite the server status showing plenty of jobs available.
I only do separation on my GPU at the moment.

Tue 12 Apr 2022 11:42:16 PM PDT | Milkyway@Home | update requested by user
Tue 12 Apr 2022 11:42:20 PM PDT | Milkyway@Home | Sending scheduler request: Requested by user.
Tue 12 Apr 2022 11:42:20 PM PDT | Milkyway@Home | Requesting new tasks for NVIDIA GPU
Tue 12 Apr 2022 11:42:22 PM PDT | Milkyway@Home | Scheduler request completed: got 0 new tasks

This is on a Linux machine with a single GPU card.
ID: 72741 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72742 - Posted: 13 Apr 2022, 6:54:32 UTC - in response to Message 72741.  
Last modified: 13 Apr 2022, 6:55:34 UTC

Same with me. Gets GPU work from time to time but mostly the machine is idle, despite the server status showing plenty of jobs available.
I only do separation on my GPU at the moment.

Tue 12 Apr 2022 11:42:16 PM PDT | Milkyway@Home | update requested by user
Tue 12 Apr 2022 11:42:20 PM PDT | Milkyway@Home | Sending scheduler request: Requested by user.
Tue 12 Apr 2022 11:42:20 PM PDT | Milkyway@Home | Requesting new tasks for NVIDIA GPU
Tue 12 Apr 2022 11:42:22 PM PDT | Milkyway@Home | Scheduler request completed: got 0 new tasks

This is on a Linux machine with a single GPU card.
I use Windows so I'm not sure what you need to do, but I've set up a scheduled task to ask for work every 2 minutes (since 1.5 minutes is a server limit), and I have my GPUs running most of the time. But you could try Keith's suggestion of:

watch -n 720 ./boinccmd --project http://milkyway.cs.rpi.edu/milkyway/ update

But overall this is probably just taking tasks that would have gone to other people. Tom needs to get this server working harder somehow.
ID: 72742 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
alanb1951

Send message
Joined: 16 Mar 10
Posts: 208
Credit: 105,453,763
RAC: 36,229
Message 72743 - Posted: 13 Apr 2022, 7:32:04 UTC - in response to Message 72742.  

But overall this is probably just taking tasks that would have gone to other people. Tom needs to get this server working harder somehow.
Valid point, but I suspect that as long as there are issues with that huge number of NBody tasks waiting there will continue to be difficulties. Lots of NBody users are clocking up huge numbers of "Validation inconclusive" results, and the retries for those will go on the end of the queue, I fear... I wonder how long it will take to get 13 million tasks out and returned?...

I don't know if there is an official mechanism that could be used to alter the order in which the feeder sends stuff out, or whether it would be possible to cancel the bulk of those excess tasks. (Of course, cancelling them would not be a good idea if they can't be recreated somehow at a later point in time.) NBody is Eric's project, not Tom's, so I guess they'll have to sort it out between them somehow.

On a related topic, I would suggest that someone writes a script that checks the number of tasks waiting to go out for each project and turns off the work generator for a project if it exceeds some pre-chosen limit, turning it back on again when the number of tasks falls below a lower limit (chosen to avoid constant stop-start) - that could then be a cron job (or equivalent) running (say) every 15 or 30 minutes. It won't solve the current problem but it would mean a recurrence could be avoided without the need for constant supervision! (I presume there isn't such a capability built into the generators already...)

Cheers - Al.
ID: 72743 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72744 - Posted: 13 Apr 2022, 7:38:20 UTC - in response to Message 72743.  

But overall this is probably just taking tasks that would have gone to other people. Tom needs to get this server working harder somehow.
Valid point, but I suspect that as long as there are issues with that huge number of NBody tasks waiting there will continue to be difficulties. Lots of NBody users are clocking up huge numbers of "Validation inconclusive" results, and the retries for those will go on the end of the queue, I fear... I wonder how long it will take to get 13 million tasks out and returned?...

I don't know if there is an official mechanism that could be used to alter the order in which the feeder sends stuff out, or whether it would be possible to cancel the bulk of those excess tasks. (Of course, cancelling them would not be a good idea if they can't be recreated somehow at a later point in time.).
The Boinc server seems to be a cobbled together piece of crap that nobody knows how to use. We all know the client sux, but the server I think is even worse.

NBody is Eric's project, not Tom's, so I guess they'll have to sort it out between them somehow..
I thought Tom was Eric's replacement?

On a related topic, I would suggest that someone writes a script that checks the number of tasks waiting to go out for each project and turns off the work generator for a project if it exceeds some pre-chosen limit, turning it back on again when the number of tasks falls below a lower limit (chosen to avoid constant stop-start) - that could then be a cron job (or equivalent) running (say) every 15 or 30 minutes. It won't solve the current problem but it would mean a recurrence could be avoided without the need for constant supervision! (I presume there isn't such a capability built into the generators already...)
That should be a main part of the boinc server software. Surely there's a setting to say what size each queue should be? Who writes this crap?
ID: 72744 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 22 · Next

Message boards : News : Server Trouble

©2024 Astroinformatics Group