Advanced search

Message boards : News : Suspending project

Author Message
Toni
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 9 Dec 08
Posts: 1006
Credit: 5,068,599
RAC: 0
Level
Ser
Scientific publications
watwatwatwat
Message 54451 - Posted: 27 Apr 2020 | 14:51:58 UTC

I'm temporarily suspending the project to give time to empty the disk.

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 566
Credit: 5,936,427,024
RAC: 10,946,881
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54452 - Posted: 27 Apr 2020 | 15:03:21 UTC

Don't worry, We'll be waiting to fill it again...
Thank you for warning.

Kaddaman
Send message
Joined: 22 Mar 20
Posts: 7
Credit: 104,244,847
RAC: 0
Level
Cys
Scientific publications
wat
Message 54459 - Posted: 27 Apr 2020 | 20:04:01 UTC - in response to Message 54451.

Ah, should have checked earlier here. Was wondering the whole day why my finished tasks don't get reported and no new tasks are incoming. Happy to start again as soon as the project is up again!

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 252
Credit: 458,054,251
RAC: 0
Level
Gln
Scientific publications
watwat
Message 54462 - Posted: 27 Apr 2020 | 22:47:21 UTC - in response to Message 54459.

In your BOINC mgr, under tools you'll find Event log.
You can get troubleshooting help there. 😃

Mine currently says
4/27/2020 4:11:47 PM | GPUGRID | Project is temporarily shut down for maintenance
4/27/2020 4:11:47 PM | GPUGRID | Project requested delay of 3600 seconds

Mark
Send message
Joined: 27 Apr 20
Posts: 2
Credit: 5,795,103
RAC: 0
Level
Ser
Scientific publications
wat
Message 54466 - Posted: 28 Apr 2020 | 7:33:07 UTC

My BOINC-Manager can't communicate with the GPUGrid-Server, in the logs i do have the same Message "Project is temporarily shut down for maintenance" since yesterday (as i wanted to start with this project).
Please have a look at this topic, otherwise i have no chance to get workunits.

Greets,
Mark

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,599,461,851
RAC: 8,787,164
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54467 - Posted: 28 Apr 2020 | 8:13:54 UTC - in response to Message 54466.
Last modified: 28 Apr 2020 | 8:14:22 UTC

My BOINC-Manager can't communicate with the GPUGrid-Server...

Nobody can communicate with the server at the moment - it's not personal to you. Toni will re-open it when the disk space has been cleared, and not before. Keep your eye on this thread for announcements, but do be aware that there will be quite a rush when the flood-gates are opened.

Pyro
Send message
Joined: 31 Mar 20
Posts: 3
Credit: 1,972,834
RAC: 0
Level
Ala
Scientific publications
wat
Message 54468 - Posted: 28 Apr 2020 | 8:17:57 UTC

Ah, I also wondered, in the future I'll check here first. This morning the BOINC client told me there were connection issues so I thought this might have been the impact on GPUGRID although I couldn't verify any connection issues.

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 566
Credit: 5,936,427,024
RAC: 10,946,881
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54469 - Posted: 28 Apr 2020 | 8:34:45 UTC

Also, there is an useful page to real-time check Server status.

Zirma
Send message
Joined: 21 Apr 20
Posts: 13
Credit: 4,411,884
RAC: 0
Level
Ala
Scientific publications
wat
Message 54471 - Posted: 28 Apr 2020 | 9:24:36 UTC
Last modified: 28 Apr 2020 | 9:26:55 UTC

i dont know how things work but if (Toni) send a mess to us (12) houer befor .. so we can load (some) wu befor he exit so we have work untill he finish the servic ? or is it all wrong? (for next time)
____________

Toni
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 9 Dec 08
Posts: 1006
Credit: 5,068,599
RAC: 0
Level
Ser
Scientific publications
watwatwatwat
Message 54472 - Posted: 28 Apr 2020 | 10:11:24 UTC - in response to Message 54471.

Stopping the server suspends assignment of WUs, but not (as far as I know) uploads.

Zirma
Send message
Joined: 21 Apr 20
Posts: 13
Credit: 4,411,884
RAC: 0
Level
Ala
Scientific publications
wat
Message 54474 - Posted: 28 Apr 2020 | 10:28:27 UTC - in response to Message 54472.

Stopping the server suspends assignment of WUs, but not (as far as I know) uploads.

I usually save 0.2-05 wu on the computer. (In order not to get any job old) But with a warning I had set it to 1-2 instead. And had a job for 1-2 days extra .. ///google translater//

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 566
Credit: 5,936,427,024
RAC: 10,946,881
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54476 - Posted: 28 Apr 2020 | 11:01:22 UTC - in response to Message 54472.

Stopping the server suspends assignment of WUs, but not (as far as I know) uploads.

Result files upload is not suspended itself, but BOINC Manager can't inform to server that tasks are finished, and they stay indefinitely at "Ready to report" state...

Toni
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 9 Dec 08
Posts: 1006
Credit: 5,068,599
RAC: 0
Level
Ser
Scientific publications
watwatwatwat
Message 54536 - Posted: 2 May 2020 | 16:11:57 UTC - in response to Message 54476.

Suspending again.

Geralt
Send message
Joined: 14 Feb 16
Posts: 5
Credit: 17,756,170
RAC: 6,324
Level
Pro
Scientific publications
wat
Message 54538 - Posted: 2 May 2020 | 17:30:38 UTC

Actually just posted about it but my work units are failing to upload due to the disk being full.

https://i.imgur.com/WCMUGXc.png

Pyro
Send message
Joined: 31 Mar 20
Posts: 3
Credit: 1,972,834
RAC: 0
Level
Ala
Scientific publications
wat
Message 54547 - Posted: 3 May 2020 | 11:01:11 UTC

I had gotten quite a huge task again that needed over a day to complete. Then, I needed 1,5 days to upload it as the server wouldn't accept it earlier. Due to a similar task in the past it'll most likely be approx. 90k points outstanding. Expiration date is tomorrow afternoon. If it can't be validated and all the time/current was wasted I'll be pissed not happy.

oemuser
Send message
Joined: 18 Sep 16
Posts: 10
Credit: 1,291,979
RAC: 0
Level
Ala
Scientific publications
wat
Message 54548 - Posted: 3 May 2020 | 11:23:36 UTC - in response to Message 54451.

Maybe think about server hardware enhancements. You may get more users now as Corona-Virus attracted many people for folding@home and if their servers are too busy then some will join GPUgrid as fallback.

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,599,461,851
RAC: 8,787,164
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54549 - Posted: 3 May 2020 | 12:05:14 UTC

Please be aware that this weekend is a major public holiday in the country where GPUGrid is based.

Allow a little extra slack for the people still working in these uncertain times.

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 566
Credit: 5,936,427,024
RAC: 10,946,881
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54550 - Posted: 3 May 2020 | 12:35:41 UTC - in response to Message 54547.

I had gotten quite a huge task again that needed over a day to complete...

You are probably referring to this PABLO WU.

...Expiration date is tomorrow afternoon. If it can't be validated and all the time/current was wasted...

Don't be afraid about this.
This WU has been assigned to you only, and there is no risk to be assigned to anybody more after deadline, because Scheduler is currently down.
As soon as server is working again, your WorkUnit will be informed as finished, and your kind time/current won't be wasted.
Well...
For this not being so, the following circumstances have to be met:
-1) WU deadline is reached.
-2) Server is brought operative again while your system is switched off or not connected to Internet.
-3) The same WU is assigned to other user, then finished and reported before your system does :-O

Pyro
Send message
Joined: 31 Mar 20
Posts: 3
Credit: 1,972,834
RAC: 0
Level
Ala
Scientific publications
wat
Message 54559 - Posted: 3 May 2020 | 16:30:53 UTC - in response to Message 54550.


You are probably referring to this PABLO WU.


Exactly.

Don't be afraid about this.


It was validated 60 seconds ago, 97k! I don't like waste, that's all, esp. not such a big one. My card working hard for 24 hours in vain is different from losing, let's say, a 5 hour or 7 hour task.

Erich56
Send message
Joined: 1 Jan 15
Posts: 1090
Credit: 6,603,906,926
RAC: 21,893,126
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwat
Message 54616 - Posted: 8 May 2020 | 19:35:07 UTC

a look at the project status page shows that the project seems to be suspended again?

Also, an uploaded task on one of my hosts sticks with "ready to report" in the status line of the BOINC manager.

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,599,461,851
RAC: 8,787,164
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54617 - Posted: 8 May 2020 | 20:07:01 UTC - in response to Message 54616.

Also, an uploaded task on one of my hosts sticks with "ready to report" in the status line of the BOINC manager.

And the reason why can be seen on BOINC's Event Log:

08/05/2020 21:04:51 | GPUGRID | Project is temporarily shut down for maintenance

- though unfortunately we don't get any deeper explanation than that.

Erich56
Send message
Joined: 1 Jan 15
Posts: 1090
Credit: 6,603,906,926
RAC: 21,893,126
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwat
Message 54619 - Posted: 8 May 2020 | 20:13:01 UTC - in response to Message 54617.

- though unfortunately we don't get any deeper explanation than that.

I guess the explanation is the same as Toni gave us in the first posting of this thread:

" I'm temporarily suspending the project to give time to empty the disk. "

Freewill
Send message
Joined: 18 Mar 10
Posts: 13
Credit: 6,070,562,394
RAC: 43,622,225
Level
Tyr
Scientific publications
watwatwatwatwat
Message 54620 - Posted: 8 May 2020 | 20:23:47 UTC - in response to Message 54619.

We're gonna need a bigger disk!
____________

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 727,920,933
RAC: 155,858
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54623 - Posted: 9 May 2020 | 3:45:42 UTC - in response to Message 54620.
Last modified: 9 May 2020 | 3:47:07 UTC

We're gonna need a bigger disk!

I doubt if bigger disks are available that are also fast enough and reliable enough. We probably need another disk server, with all the high speed disks it can handle.

Zirma
Send message
Joined: 21 Apr 20
Posts: 13
Credit: 4,411,884
RAC: 0
Level
Ala
Scientific publications
wat
Message 54625 - Posted: 9 May 2020 | 4:15:47 UTC - in response to Message 54623.

We get the 50% of volunters from seti:-) Now we need 50% of seti's servers to :-)

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 1284
Credit: 4,919,806,959
RAC: 6,293,008
Level
Arg
Scientific publications
watwatwatwatwat
Message 54626 - Posted: 9 May 2020 | 5:27:40 UTC

Yes, Seti should donate the disks we financed on the last pledge drive that they never will use to GPUGrid.net

Erich56
Send message
Joined: 1 Jan 15
Posts: 1090
Credit: 6,603,906,926
RAC: 21,893,126
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwat
Message 54627 - Posted: 9 May 2020 | 6:17:47 UTC

the bad thing is, from a technical point of view, that with all these outages which now seem to occur on a rather regular basis, the GPUs are going through this so-called "thermal cycle" (cooling off - heating up) everytime this happens. Which is not good for the GPUs at all, it certainly shortens their lifetime.

As a consequence, I am thinking about abandoning GPUGRID until these problems are solved. I simply don't want to ruin my GPUs :-(

Speedy
Send message
Joined: 19 Aug 07
Posts: 42
Credit: 28,391,082
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 54628 - Posted: 9 May 2020 | 7:58:58 UTC

@Keith I have to disagree with you. Even though I did not donate the money can go to supporting the small team that worked to keep volunteers supplied with work.

@Erich56 I hear what you are saying in regards to your GPU's calling off/heating up. I wish you all the best with your decision whether you stay or go. In the event you do go (which I hope you don't) :) it would mean the discs won't fill up quite as quickly.

I have not seen on a note from admin since 2nd of May (New Zealand time) saying the server's have been suspended. It looks very much like they have been just no comment about it

Zirma
Send message
Joined: 21 Apr 20
Posts: 13
Credit: 4,411,884
RAC: 0
Level
Ala
Scientific publications
wat
Message 54629 - Posted: 9 May 2020 | 8:15:46 UTC - in response to Message 54628.
Last modified: 9 May 2020 | 8:21:47 UTC

@ Erich56 i run primegrid as backup on gpu(nvidia) in 0 mode. = if gpugrid dont have work/working primgrid star running. // stoptime under 15 sec to change from gpugrid to primegrid .. Runing fast work 3-40 minuts work as backup .. have them even for cpu if rosettta goes down. Set work to 0 (Zero) and they onely start if the cpu or gpu dont have work.
Warning some workt are to hot for my gpu .. over 95 celsius. So it take some time to test it . And ther are work from 3 minuts gpu to over 100 houers.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 7,520
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54630 - Posted: 9 May 2020 | 9:02:17 UTC - in response to Message 54628.

I have not seen on a note from admin since 2nd of May (New Zealand time) saying the server's have been suspended. It looks very much like they have been just no comment about it
They don't have to comment it every time the disk gets full, as we are aware of it. You can see it on the server status (click on the middle text on the upper right corner) page in the form of many server components showing "Not running" status marked with a red background. Also you can see it in the log of your BOINC manager when it tries to communicate with GPUGrid:
09/05/2020 10:58:08 | GPUGRID | Project is temporarily shut down for maintenance 09/05/2020 10:58:08 | GPUGRID | Project requested delay of 3600 seconds


Keith Myers
Send message
Joined: 13 Dec 17
Posts: 1284
Credit: 4,919,806,959
RAC: 6,293,008
Level
Arg
Scientific publications
watwatwatwatwat
Message 54633 - Posted: 9 May 2020 | 16:01:47 UTC

@Speedy, the money was already spent on the new storage array. But Seti has closed down and has no need for any more storage. Their existing storage is sufficient with no work going out and just the last dribs and drabs being returned. They never installed the new array. It is just sitting in boxes somewhere. Put it to good use here where all the Seti orphans ended up.

Speedy
Send message
Joined: 19 Aug 07
Posts: 42
Credit: 28,391,082
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 54634 - Posted: 9 May 2020 | 22:33:07 UTC

@Keith, I hear what you are saying, I was not aware the new aray had been brought. My suggestion would be to use it when Seti is back up and running if it is usable or salad and use the proceeds for the project, or maybe they could be donated to the other project that was/is helping Seti out
@Retvari Zoltan yes .fair enough, I just thought they would have made a note in this thread when it was happening that was all

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 503
Credit: 727,920,933
RAC: 155,858
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54635 - Posted: 10 May 2020 | 3:30:13 UTC - in response to Message 54634.

Speedy, what other project was/is helping Seti out?

Speedy
Send message
Joined: 19 Aug 07
Posts: 42
Credit: 28,391,082
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 54636 - Posted: 10 May 2020 | 4:29:53 UTC - in response to Message 54635.

Speedy, what other project was/is helping Seti out?

Einstein@Home IRC Bruce Allen director of the project was/is going to allow lots of data to be processed on one of their servers. Unfortunately I cannot find the information relating to this, I am sure it on the Seti message boards somewhere

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 1284
Credit: 4,919,806,959
RAC: 6,293,008
Level
Arg
Scientific publications
watwatwatwatwat
Message 54637 - Posted: 10 May 2020 | 5:34:24 UTC

Yes, the focus of analyzing the past twenty years of crunched data was being processed as the "Nebula" project by David Anderson.

https://setiathome.berkeley.edu/nebula/web/

He is analyzing with the Atlas Cluster hosted at Einstein@home.

https://setiathome.berkeley.edu/nebula/web/architecture.php

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,599,461,851
RAC: 8,787,164
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54638 - Posted: 10 May 2020 | 7:18:05 UTC - in response to Message 54633.

@Speedy, the money was already spent on the new storage array. But Seti has closed down and has no need for any more storage. Their existing storage is sufficient with no work going out and just the last dribs and drabs being returned. They never installed the new array. It is just sitting in boxes somewhere. Put it to good use here where all the Seti orphans ended up.

That's not what Eric Korpela said in his blog post:

I know some of you are upset that you donated only to see the public participation part of the project end. If you donated for the disk array, we'll still need that. Jeff is in the process of assembling it. If you gave money in the annual drive, we'll have bills to pay all the way to the end (including page charges for the papers).

(3 March 2020)

Speedy
Send message
Joined: 19 Aug 07
Posts: 42
Credit: 28,391,082
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwat
Message 54639 - Posted: 10 May 2020 | 7:51:54 UTC - in response to Message 54638.

Thank you Richard and Keith for your responses. I am pleased the array did not go to waste after all, until now I hadn't seen Eric's response thanks for posting

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 252
Credit: 458,054,251
RAC: 0
Level
Gln
Scientific publications
watwat
Message 54659 - Posted: 11 May 2020 | 15:55:54 UTC

To those who wish to keep their GPU temp steady, I installed Folding@home alongside BOINC during the outage and tested running multiple apps on my old Optiplex 980 w/2 GTX 1650s to see if it might ensue a train wreck.

To my delight, it ran most politely alongside of BOINC (running 6 CPU threads of Rosetta and 2 threads of Asteroids on the GPUs) and set up 1 CPU slot and 2 GPU slots running COVID-19 protease models and very efficiently, apparently timeslicing itself in so that I saw steady usage and temperature with a varying power consumption between 60-110%.

When Asteroids ran out of work units F@H took over and the GPUs never even noticed.

Then I got brave and ran the F@H app on 'full comuting' while I ran some Milky Way GPU tasks in BOINC. Test result positve.

When I saw that GPUGRID had been relaunched and had taken priority over Milky Way, I was relieved to see that my machine was functioning normally running 2 GPUGRIDs and 2 F@H GPU models, albeit at about half speed. I set F@H back to 'medium' and had it finish the current work and pause, so that I can run BOINC projects full speed.

The obvious caveat is that it will take much longer to finish GPUGRID WUs, as they use much more of the GPU capacity than Milky way or Asteroids tasks. I also slowed Rosetta tasks to share my CPU with F@H but the server adjusted subsequent tasks to run the set duration.

marsinph
Send message
Joined: 11 Feb 18
Posts: 41
Credit: 579,891,424
RAC: 0
Level
Lys
Scientific publications
wat
Message 54737 - Posted: 19 May 2020 | 14:21:24 UTC

Hello everyone and admin,
Any idea when you will produce new WU's ?
All queue are as good as empty !
Take care on health.
Best regards

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1112
Credit: 6,162,416,256
RAC: 0
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54740 - Posted: 19 May 2020 | 20:20:36 UTC

It may help to purge the database of the many thousands of "Error" tasks going back to at least 2014. Should be easy to do.

Post to thread

Message boards : News : Suspending project

//