Welcome to MilkyWay@home

Server Update

Message boards : News : Server Update
Message board moderation

To post messages, you must log in.

AuthorMessage
Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 25 Feb 13
Posts: 580
Credit: 94,200,158
RAC: 0
Message 65974 - Posted: 30 Nov 2016, 5:47:04 UTC

Hey Everyone,

I pushed a quick server update overnight tonight. This will probably result in the remaining:

de_modfit_fast_19_3s_136_bundle5_ModfitConstraints3_5
de_modfit_fast_19_3s_136_bundle5_ModfitConstraints3_6
de_modfit_fast_19_3s_136_bundle5_ModfitConstraints3_7
de_modfit_fast_19_3s_136_bundle5_ModfitConstraints3_8

results being crunched to return invalid results.

The new runs I put up:

de_modfit_fast_19_3s_140_bundle5_ModfitConstraints3_1
de_modfit_fast_19_3s_140_bundle5_ModfitConstraints3_2
de_modfit_fast_19_3s_140_bundle5_ModfitConstraints3_3
de_modfit_fast_19_3s_140_bundle5_ModfitConstraints3_4

should run and validate correctly.

As a status update for the Mac client and Linux client, I am still working on them, they are just slow going. I am working on finishing up some class work before the semester ends while also working on those applications so I will keep everyone posted on their progress.

Jake
ID: 65974 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Arivald Ha'gel

Send message
Joined: 30 Apr 14
Posts: 67
Credit: 160,674,488
RAC: 0
Message 65975 - Posted: 30 Nov 2016, 8:32:16 UTC

Jake,

shouldn't then those WU be cancelled by the server? It would be much better?

I also see that the amount of "Workunits waiting for validation" is climbing. Currently up to: 26,841. It seems bundle of 5 will not be enough to quench thirst for WUs.
Any update on "Max tasks per day" not working correctly?
ID: 65975 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 25 Feb 13
Posts: 580
Credit: 94,200,158
RAC: 0
Message 65976 - Posted: 30 Nov 2016, 16:28:14 UTC

Arivald,

I think I fixed it so they should validate correctly, so no need. Was just taking the cautious approach of warning people.

Jake
ID: 65976 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Arivald Ha'gel

Send message
Joined: 30 Apr 14
Posts: 67
Credit: 160,674,488
RAC: 0
Message 65977 - Posted: 30 Nov 2016, 17:53:32 UTC - in response to Message 65976.  

Jake,

Sweet.

I see that "Workunits waiting for validation" came down also.
However I still see my favorite Hosts churning through a lot of WUs... all of them "validate errors"...

http://milkyway.cs.rpi.edu/milkyway/host_app_versions.php?hostid=606779
ID: 65977 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 25 Feb 13
Posts: 580
Credit: 94,200,158
RAC: 0
Message 65978 - Posted: 30 Nov 2016, 18:53:31 UTC

Arivald,

It looks like you a running a homebrew application on that host. I would recommend recompiling to be the latest version on the github code or running it with the application provided from the server. If there are still issues, let me know.

Jake
ID: 65978 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Arivald Ha'gel

Send message
Joined: 30 Apr 14
Posts: 67
Credit: 160,674,488
RAC: 0
Message 65980 - Posted: 30 Nov 2016, 23:31:56 UTC - in response to Message 65978.  

Arivald,

It looks like you a running a homebrew application on that host. I would recommend recompiling to be the latest version on the github code or running it with the application provided from the server. If there are still issues, let me know.

Jake


What? I'm running?!?

It's not me. It's "Ingmar Hensler" (http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=606779)
He is running homebrew application (Anonymous Platform). But this is not the real problem. Problem is that "Max tasks per day" should be lower than 10 000, and should decrease by 1 (or more) for each "invalid" task. Or at least should not allow MORE than this amount of tasks per day.
But it is 10k, it's not being decreased, and it does allow more than 10k tasks for this host per day (around 20k+ as I remember). This causes many invalid WU, and also serves as a DoS attack on the server (and on MW@H processing).

I have already shown this problem here:
https://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=4052&postid=65875
(16 Nov, 2016 - around 2 weeks ago)
and here:
http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3990
(3 Aug 2016 - 4 MONTHS ago)
a Topic that was ignored by the whole community, and MW@H team also.

Problem is that NOT ANY HOST can process 10k WU per day! Barely any can really. My single R280X processes ~2750 per day. So even with 4 of those, it's barely above 10k. So it should simply start at 1000 (still a high but reasonable number), and go up by 1 for each good task, and 1 down for each bad task.
ID: 65980 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
bluestang

Send message
Joined: 13 Oct 16
Posts: 112
Credit: 1,174,293,644
RAC: 0
Message 65981 - Posted: 1 Dec 2016, 14:57:45 UTC - in response to Message 65980.  
Last modified: 1 Dec 2016, 14:58:34 UTC

I agree on this. These Hosts should be banned somehow or at least limited in the tasks they can request. They don't do a single valid WU to this project! It screws up the rest of us, both in efficiency and points (yes, i can be a point whore!).
ID: 65981 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 25 Feb 13
Posts: 580
Credit: 94,200,158
RAC: 0
Message 65983 - Posted: 2 Dec 2016, 15:02:32 UTC

Arivald,

I didn't catch the sarcasm about you "favorite" host in your post. (Sarcasm is hard to read in text sorry.) I agree there should be a way to limit these host that are constantly failing. I have a few ideas, but I do not have the time to implement them right now. Maybe this is something I will revisit once we are on break for the holidays.

Jake
ID: 65983 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
bluestang

Send message
Joined: 13 Oct 16
Posts: 112
Credit: 1,174,293,644
RAC: 0
Message 65984 - Posted: 2 Dec 2016, 16:29:42 UTC - in response to Message 65983.  

Sounds good...Keep up the good work!
ID: 65984 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
paris
Avatar

Send message
Joined: 26 Apr 08
Posts: 87
Credit: 64,801,496
RAC: 0
Message 65991 - Posted: 8 Dec 2016, 15:24:46 UTC

Do you expect the Mac compatible version to be ready soon? I have back-up projects keeping my machines busy but I am looking forward to helping with MilkyWay again. Thanks for all your hard work and the helpful replies I have seen in the past. It must take away from time spent on the project. I understand that the OS's that have the greatest returns should get the highest priority but I expect that there are a number of Mac users in the same boat as I am. Keep an keeping on. :-)


Plus SETI Classic = 21,082 WUs
ID: 65991 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jean-Pierre HARLE

Send message
Joined: 25 Sep 08
Posts: 15
Credit: 145,544,797
RAC: 0
Message 66011 - Posted: 13 Dec 2016, 10:11:28 UTC - in response to Message 65991.  

Do you expect the Mac compatible version to be ready soon? I have back-up projects keeping my machines busy but I am looking forward to helping with MilkyWay again. Thanks for all your hard work and the helpful replies I have seen in the past. It must take away from time spent on the project. I understand that the OS's that have the greatest returns should get the highest priority but I expect that there are a number of Mac users in the same boat as I am. Keep an keeping on. :-)


Hi Jake,

Same question concerning Mac compatible version. Still no task available today.

Mar 13 déc 11:02:55 2016 | Milkyway@Home | Requesting new tasks for CPU and AMD/ATI GPU
Mar 13 déc 11:02:56 2016 | Milkyway@Home | Scheduler request completed: got 0 new tasks
Mar 13 déc 11:02:56 2016 | Milkyway@Home | No tasks sent
Mar 13 déc 11:02:56 2016 | Milkyway@Home | No tasks are available for MilkyWay@Home

I loaded my last Mac MW@H task on Nov 11... :-((

Jean-Pierre
ID: 66011 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
James Lee*

Send message
Joined: 14 Oct 13
Posts: 6
Credit: 21,573,789
RAC: 0
Message 66016 - Posted: 14 Dec 2016, 7:30:46 UTC
Last modified: 14 Dec 2016, 7:38:12 UTC

I just checked in to check my results, as I was going to add 7 more machines to crunch here... when I noticed TOO MANY are getting "Completed, Validation inconclusive". Seems more are failing, especially Nvidia GPUs, which has a 3 to 1 ratio of failure to the CPUs. If there is some problem that I should know about, tell me, as I will not run 12 machines with this fail rate, and will pull out the 3 or 4 that are running this project.
ID: 66016 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
James Lee*

Send message
Joined: 14 Oct 13
Posts: 6
Credit: 21,573,789
RAC: 0
Message 66018 - Posted: 14 Dec 2016, 15:09:47 UTC

I do carry a 2 day buffer, and so you may have fixed something that will soon show up. In the meantime, I added the new machines, so now have 10 crunching, for now, but removed all GPU processing. Let me know if I have to do something as simple as resetting the project to get things to work properly, as I would like to add 8 GPUs.
ID: 66018 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
bluestang

Send message
Joined: 13 Oct 16
Posts: 112
Credit: 1,174,293,644
RAC: 0
Message 66019 - Posted: 14 Dec 2016, 19:00:41 UTC - in response to Message 66016.  

I just checked in to check my results, as I was going to add 7 more machines to crunch here... when I noticed TOO MANY are getting "Completed, Validation inconclusive". Seems more are failing, especially Nvidia GPUs, which has a 3 to 1 ratio of failure to the CPUs. If there is some problem that I should know about, tell me, as I will not run 12 machines with this fail rate, and will pull out the 3 or 4 that are running this project.


"Validation Inconclusive" is no biggie, it's just waiting for a wingman to run and validate it also. Then it goes to "Validated".
ID: 66019 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Rymorea

Send message
Joined: 6 Oct 14
Posts: 46
Credit: 20,017,425
RAC: 0
Message 66020 - Posted: 14 Dec 2016, 22:48:59 UTC - in response to Message 66018.  

I do carry a 2 day buffer, and so you may have fixed something that will soon show up. In the meantime, I added the new machines, so now have 10 crunching, for now, but removed all GPU processing. Let me know if I have to do something as simple as resetting the project to get things to work properly, as I would like to add 8 GPUs.


Don't worry about "Completed, validation inconclusive" but look "Completed, can't validate" = Invalid results. If invalids goes high something wrong. Sometimes I have %50 Completed, validation inconclusive tasks but in 2-3 days all gone to validated.
ID: 66020 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
James Lee*

Send message
Joined: 14 Oct 13
Posts: 6
Credit: 21,573,789
RAC: 0
Message 66021 - Posted: 15 Dec 2016, 0:35:27 UTC

Thank you for the responses.. maybe "Waiting for Validation" would be a better/clearer message (at least, for me, lol). OK, since everyone made me feel so much better, I'll add some GPUs into the mix and should be doing some serious crunching here as soon as my other projects run out from their buffering. Again, Thanks All!
ID: 66021 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Richard

Send message
Joined: 18 Apr 11
Posts: 1
Credit: 73,859
RAC: 0
Message 66030 - Posted: 19 Dec 2016, 11:39:24 UTC - in response to Message 65974.  

Jake

Most of work units from 17 Dec to 19 Dec have compute error with some units saying 100 percent done but have 14 seconds remaining. Just wondering what happened.

Richard
ID: 66030 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
wb8ili

Send message
Joined: 18 Jul 10
Posts: 76
Credit: 635,998,708
RAC: 0
Message 66031 - Posted: 20 Dec 2016, 15:59:25 UTC

Richard - The error you are getting is that your GPU doesn't support double precision.

I don't know if that is a driver issue or a change in the GPU requirements of the workunits you are being sent.
ID: 66031 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Kailee71

Send message
Joined: 22 Nov 16
Posts: 4
Credit: 46,737,032
RAC: 0
Message 66118 - Posted: 15 Jan 2017, 10:44:55 UTC - in response to Message 66011.  

Do you expect the Mac compatible version to be ready soon? I have back-up projects keeping my machines busy but I am looking forward to helping with MilkyWay again. Thanks for all your hard work and the helpful replies I have seen in the past. It must take away from time spent on the project. I understand that the OS's that have the greatest returns should get the highest priority but I expect that there are a number of Mac users in the same boat as I am. Keep an keeping on. :-)


Hi Jake,

Same question concerning Mac compatible version. Still no task available today.

Mar 13 déc 11:02:55 2016 | Milkyway@Home | Requesting new tasks for CPU and AMD/ATI GPU
Mar 13 déc 11:02:56 2016 | Milkyway@Home | Scheduler request completed: got 0 new tasks
Mar 13 déc 11:02:56 2016 | Milkyway@Home | No tasks sent
Mar 13 déc 11:02:56 2016 | Milkyway@Home | No tasks are available for MilkyWay@Home

I loaded my last Mac MW@H task on Nov 11... :-((

Jean-Pierre


*bump*

Any news on a mac gpu app?

TIA,

Kailee.
ID: 66118 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jozef J

Send message
Joined: 4 Mar 10
Posts: 65
Credit: 639,958,626
RAC: 0
Message 66311 - Posted: 1 May 2017, 12:23:29 UTC

some "glitch" in server today ..? now it look better
ID: 66311 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : News : Server Update

©2024 Astroinformatics Group