Welcome to MilkyWay@home

Aaargh! Server out of new work!


Advanced search

Message boards : Number crunching : Aaargh! Server out of new work!
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 13 · 14 · 15 · 16 · 17 · 18 · 19 . . . 22 · Next

AuthorMessage
ProfileThe Gas Giant
Avatar

Send message
Joined: 24 Dec 07
Posts: 1947
Credit: 240,884,648
RAC: 0
200 million credit badge10 year member badge
Message 43429 - Posted: 2 Nov 2010, 19:32:38 UTC

..and up.
ID: 43429 · Rating: 0 · rate: Rate + / Rate - Report as offensive
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
50 million credit badge10 year member badge
Message 43430 - Posted: 2 Nov 2010, 20:18:59 UTC

Still no work being distributed though. Running on reducing cache.
Go away, I was asleep


ID: 43430 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
100 million credit badge10 year member badgeextraordinary contributions badge
Message 43431 - Posted: 2 Nov 2010, 20:38:11 UTC - in response to Message 43430.  

I keep getting one machine running dry, and not being fed. The other is ok albeit gets short at times. Its my dual 5970 machine that is not getting fed quick enough, keeps getting "zero sent" whilst there are WUs available. At literally the same instance in time, my other machine with a 5850 gets WUs.....

I have switched the dual 5970 to Collatz for now until supply is sorted out, as I am having to baby sit it, and its running at half capacity at MW at present

Regards
Zy
ID: 43431 · Rating: 0 · rate: Rate + / Rate - Report as offensive
ProfileThe Gas Giant
Avatar

Send message
Joined: 24 Dec 07
Posts: 1947
Credit: 240,884,648
RAC: 0
200 million credit badge10 year member badge
Message 43432 - Posted: 2 Nov 2010, 20:39:33 UTC - in response to Message 43431.  

I keep getting one machine running dry, and not being fed. The other is ok albeit gets short at times. Its my dual 5970 machine that is not getting fed quick enough, keeps getting "zero sent" whilst there are WUs available. At literally the same instance in time, my other machine with a 5850 gets WUs.....

I have switched the dual 5970 to Collatz for now until supply is sorted out, as I am having to baby sit it, and its running at half capacity at MW at present

Regards
Zy

B A C K U P P R O J E C T !
ID: 43432 · Rating: 0 · rate: Rate + / Rate - Report as offensive
TJ

Send message
Joined: 12 Aug 09
Posts: 262
Credit: 92,631,041
RAC: 0
50 million credit badge10 year member badge
Message 43454 - Posted: 3 Nov 2010, 21:14:58 UTC

We are in "code red" at the moment.
Greetings from,
TJ
ID: 43454 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Matthew
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 6 May 09
Posts: 217
Credit: 6,856,375
RAC: 0
5 million credit badge10 year member badge
Message 43456 - Posted: 3 Nov 2010, 22:13:27 UTC

We're working on it.

-Matthew
ID: 43456 · Rating: 0 · rate: Rate + / Rate - Report as offensive
55degrees

Send message
Joined: 8 Sep 09
Posts: 62
Credit: 61,330,584
RAC: 0
50 million credit badge10 year member badge
Message 43457 - Posted: 3 Nov 2010, 22:17:32 UTC - in response to Message 43432.  

B A C K U P P R O J E C T !


this has probably been hammered to pulp, but I am not interested in scouring this whopper thread.

I have a feeling backupproject is, in the current ati environement, not as envisioned when collatz thrives on memory clock and mw does not. changes have to be made if interested in keeping the gpus setup optimally for each project and that requires intervention.


ID: 43457 · Rating: 0 · rate: Rate + / Rate - Report as offensive
BarryAZ

Send message
Joined: 1 Sep 08
Posts: 519
Credit: 283,790,280
RAC: 4,733
200 million credit badge10 year member badgeextraordinary contributions badge
Message 43461 - Posted: 3 Nov 2010, 23:26:40 UTC - in response to Message 43457.  

I agree, I run three ATI GPU projects, Dnetc and Collatz run at least moderately well together. When I add MW into the mix, the only way MW will process work is when I temporarily suspend the other two projects.

Since MW is, frankly, the LEAST reliable of the three projects at this point, what I do is suspend the other two projects only when I can check back often enough to make sure that MW hasn't gone walkabout (like it has this afternoon), so I can pull either or both Collatz and Dnetc off of suspend mode, at which point they take over GPU processing (even if there is MW work).



B A C K U P P R O J E C T !


this has probably been hammered to pulp, but I am not interested in scouring this whopper thread.

I have a feeling backupproject is, in the current ati environement, not as envisioned when collatz thrives on memory clock and mw does not. changes have to be made if interested in keeping the gpus setup optimally for each project and that requires intervention.




ID: 43461 · Rating: 0 · rate: Rate + / Rate - Report as offensive
ProfileSutaru Tsureku

Send message
Joined: 30 Apr 09
Posts: 99
Credit: 25,715,405
RAC: 29
20 million credit badge10 year member badge
Message 43462 - Posted: 4 Nov 2010, 0:06:34 UTC - in response to Message 43456.  

We're working on it.

-Matthew


Thanks for info!

ID: 43462 · Rating: 0 · rate: Rate + / Rate - Report as offensive
BarryAZ

Send message
Joined: 1 Sep 08
Posts: 519
Credit: 283,790,280
RAC: 4,733
200 million credit badge10 year member badgeextraordinary contributions badge
Message 43463 - Posted: 4 Nov 2010, 0:28:47 UTC - in response to Message 43462.  

Though a bit more info might be a good thing (ie outage for hours, days)...

We're working on it.

-Matthew


Thanks for info!


ID: 43463 · Rating: 0 · rate: Rate + / Rate - Report as offensive
ProfileMike
Avatar

Send message
Joined: 31 Oct 10
Posts: 135
Credit: 3,727,511
RAC: 1
3 million credit badge9 year member badge
Message 43561 - Posted: 7 Nov 2010, 9:08:15 UTC
Last modified: 7 Nov 2010, 9:17:55 UTC


ID: 43561 · Rating: 0 · rate: Rate + / Rate - Report as offensive
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
50 million credit badge10 year member badge
Message 43563 - Posted: 7 Nov 2010, 9:26:55 UTC

Mike

Many of the BM Messages tab messages are there to be ignored if you know your monster is meeting it's project targets.
Go away, I was asleep


ID: 43563 · Rating: 0 · rate: Rate + / Rate - Report as offensive
ProfileMike
Avatar

Send message
Joined: 31 Oct 10
Posts: 135
Credit: 3,727,511
RAC: 1
3 million credit badge9 year member badge
Message 43564 - Posted: 7 Nov 2010, 9:51:56 UTC - in response to Message 43563.  


ID: 43564 · Rating: 0 · rate: Rate + / Rate - Report as offensive
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
50 million credit badge10 year member badge
Message 43584 - Posted: 8 Nov 2010, 2:20:49 UTC

Validator up to 50K ATM, so trouble is here.

My MW tasks report but are now not getting new work.

I will leave the good cache and revert to Collatz.
Go away, I was asleep


ID: 43584 · Rating: 0 · rate: Rate + / Rate - Report as offensive
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
50 million credit badge10 year member badge
Message 43753 - Posted: 11 Nov 2010, 20:43:23 UTC

Validator choking on returned work.

Is this another Aaaarrrrgggghhhh moment soon?
Go away, I was asleep


ID: 43753 · Rating: 0 · rate: Rate + / Rate - Report as offensive
55degrees

Send message
Joined: 8 Sep 09
Posts: 62
Credit: 61,330,584
RAC: 0
50 million credit badge10 year member badge
Message 43822 - Posted: 14 Nov 2010, 11:30:31 UTC
Last modified: 14 Nov 2010, 11:31:21 UTC

11/14/2010 6:14:52 AM Milkyway@home Sending scheduler request: To fetch work.
11/14/2010 6:14:52 AM Milkyway@home Requesting new tasks for GPU
11/14/2010 6:14:55 AM Milkyway@home Scheduler request completed: got 0 new tasks
11/14/2010 6:14:55 AM Milkyway@home Message from server: No work sent
11/14/2010 6:14:55 AM Milkyway@home Message from server: Your app_info.xml file doesn't have a version of MilkyWay@Home N-Body Simulation.
11/14/2010 6:17:02 AM Milkyway@home Sending scheduler request: To fetch work.
11/14/2010 6:17:02 AM Milkyway@home Requesting new tasks for GPU
11/14/2010 6:17:05 AM Milkyway@home Scheduler request completed: got 0 new tasks
11/14/2010 6:17:05 AM Milkyway@home Message from server: No work sent
11/14/2010 6:17:05 AM Milkyway@home Message from server: Your app_info.xml file doesn't have a version of MilkyWay@Home N-Body Simulation.
11/14/2010 6:19:35 AM Milkyway@home update requested by user
11/14/2010 6:19:37 AM Milkyway@home Sending scheduler request: Requested by user.
11/14/2010 6:19:37 AM Milkyway@home Requesting new tasks for GPU
11/14/2010 6:19:40 AM Milkyway@home Scheduler request completed: got 0 new tasks
11/14/2010 6:19:40 AM Milkyway@home Message from server: No work sent
11/14/2010 6:19:40 AM Milkyway@home Message from server: Your app_info.xml file doesn't have a version of MilkyWay@Home N-Body Simulation.


server status indicates WUs available, running. I run gpus, not cpu. not getting WUs. I post this because the messages kind of suggest no WUs because I am not running n-body simulation. that has not been a problem until about an hour ago. something change, lately?
ID: 43822 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
100 million credit badge10 year member badgeextraordinary contributions badge
Message 43823 - Posted: 14 Nov 2010, 11:50:31 UTC - in response to Message 43822.  
Last modified: 14 Nov 2010, 11:54:24 UTC

The validator is choking again. An hour ago is was about par, 6,500 waiting for validation. Over the last hour its steadily risen, and its now well over 50,000 and still rising. When it gets like that it trys to balance off WU production V Supply V Validation and you end up with bursts of available WUs etc, but its an inexorable move to falling over. See what happens, lately its survived on over 200,000 waiting validation, but at present the runes are not good, and back up projects need checking and winding up. Would be no surprise if it fell over totally within an hour from now.

EDIT:
Hope springs eternal - the GPU WU assimilator just went offline, could be someone is beavering away up there , lets hope so (and plaudits as its Sunday and a day off as such for them)

Regards
Zy
ID: 43823 · Rating: 0 · rate: Rate + / Rate - Report as offensive
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
50 million credit badge10 year member badge
Message 43825 - Posted: 14 Nov 2010, 13:33:07 UTC

And still rising - now to 72,490

Some admin needs to sort out the servers, they deserve a good kicking.
Go away, I was asleep


ID: 43825 · Rating: 0 · rate: Rate + / Rate - Report as offensive
KWH*
Avatar

Send message
Joined: 24 Aug 10
Posts: 181
Credit: 83,100,546
RAC: 0
50 million credit badge10 year member badge
Message 43827 - Posted: 14 Nov 2010, 14:54:08 UTC

Not gotten squat the past few days. GPU's sitting there doing nothing.
ID: 43827 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
100 million credit badge10 year member badgeextraordinary contributions badge
Message 43828 - Posted: 14 Nov 2010, 14:58:53 UTC - in response to Message 43827.  
Last modified: 14 Nov 2010, 15:03:32 UTC

Supply was fine up to earlier this morning. There must be another reason connected with your setup - what is your cache level, held WUs for different Projects, and BOINC stated run time for each of your GPU Project WUs?

EDIT:
Just had a look at your machines, your 3300 has been performing fine, the others are empty, so there is a difference somewhere in the setups between the main machine and the others which were not getting work. None will right now due to the validator

Regards
Zy
ID: 43828 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Previous · 1 . . . 13 · 14 · 15 · 16 · 17 · 18 · 19 . . . 22 · Next

Message boards : Number crunching : Aaargh! Server out of new work!

©2020 Astroinformatics Group