Panic Mode On (96) Server Problems?

Message boards : Number crunching : Panic Mode On (96) Server Problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 19 · 20 · 21 · 22 · 23 · 24 · Next

AuthorMessage
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 24025
Credit: 261,360,520
RAC: 489
Australia
Message 1662773 - Posted: 8 Apr 2015, 23:39:02 UTC - in response to Message 1662770.  

How do you set up a backup project? If I have one set to "no new tasks" in BOINC, but have my S@H project to allow work from other projects when no work for S@H is available, will work for other projects still be downloaded?

I have SETI share set at 100 while backup projects are set to 0 or as close as possible (Rosetta is 0.01 here as 0 wouldn't work).

If you have a project set to "no new tasks" you will not get work from it. ;-)

Cheers.
ID: 1662773 · Report as offensive
Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar

Send message
Joined: 20 Dec 05
Posts: 3187
Credit: 57,163,290
RAC: 0
United States
Message 1662777 - Posted: 9 Apr 2015, 0:00:59 UTC

did someone forget to bring Beta up? I'm getting the ol'

4/8/2015 4:55:59 PM | SETI@home Beta Test | Project is temporarily shut down for maintenance

even tho' everything on the beta server page says it's up...
.

Hello, from Albany, CA!...
ID: 1662777 · Report as offensive
Profile TimeLord04
Volunteer tester
Avatar

Send message
Joined: 9 Mar 06
Posts: 20936
Credit: 33,933,039
RAC: 23
United States
Message 1662833 - Posted: 9 Apr 2015, 4:09:16 UTC

That's why I asked if Beta was borked... Both of my computers keep giving me Communication Deferred with LONG timer countdowns.

Prometheus's log:

4/8/2015 8:23:13 PM | SETI@home Beta Test | Requesting new tasks for NVIDIA GPU
4/8/2015 8:23:14 PM | SETI@home Beta Test | Scheduler request completed: got 0 new tasks
4/8/2015 8:23:14 PM | SETI@home Beta Test | Project is temporarily shut down for maintenance


NOT good... :-(

I'm completely out of SETI work on both Exeter and Prometheus. :-(

No backup work is being pulled on Prometheus, and I don't know why. :-(

Exeter is only working two Einstein WUs at a time; including downloading them. :-(

Still can't Report completed Beta WUs on Exeter. :-(

Seems borked on SETI and Beta...
TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos
ID: 1662833 · Report as offensive
Victor Wedge
Avatar

Send message
Joined: 3 Apr 04
Posts: 28
Credit: 12,569,503
RAC: 0
Message 1662835 - Posted: 9 Apr 2015, 4:11:36 UTC
Last modified: 9 Apr 2015, 4:15:03 UTC

I've seen several spurts of outgoing data on the kiosk over the past half hour. Then, it dies down to 12/minute. Maybe they'll get her running soon.
ID: 1662835 · Report as offensive
Profile Cactus Bob
Avatar

Send message
Joined: 19 May 99
Posts: 209
Credit: 10,924,287
RAC: 29
Canada
Message 1662854 - Posted: 9 Apr 2015, 5:05:03 UTC

OMG = I got 10 CPU Work units left. I have been out of GPU units all day.

I am forced helping MilkyWay find their way out of the dark. What does that matter - They are just looking for dark matter. By the way the credit scale given for WU is hugely different. It has taken me a couple weeks to get my RAC here to 7.5K I went from 500 to just under 10K in 10 hours. I guess there isn't anyway to standardize credit. Not to get too off topic. It looks like the servers will stay down for the rest of night. SIGH . THE DB must have been FUBARed big time.

Bob

-----------
Still working on an awesome sig
Sometimes I wonder, what happened to all the people I gave directions to?
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
SETI@home classic workunits 4,321
SETI@home classic CPU time 22,169 hours
ID: 1662854 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1662865 - Posted: 9 Apr 2015, 5:27:36 UTC

Yep its back up projects for two of my machines. The oldest one still has a lot of work.
[/quote]

Old James
ID: 1662865 · Report as offensive
Profile Lynn Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 20 Nov 00
Posts: 14060
Credit: 79,603,650
RAC: 123
United States
Message 1662882 - Posted: 9 Apr 2015, 6:12:08 UTC - in response to Message 1662865.  

No work on 4 machines. One full day now. :((((
Why donate to SETI????
ID: 1662882 · Report as offensive
Victor Wedge
Avatar

Send message
Joined: 3 Apr 04
Posts: 28
Credit: 12,569,503
RAC: 0
Message 1662893 - Posted: 9 Apr 2015, 6:37:19 UTC
Last modified: 9 Apr 2015, 6:40:06 UTC

Question relevant to server downtime and BOINC.

I have my minimum work buffer and max additional buffer both set to 10 days. I have enough WUs on hand to keep CPU crunching for days. But, I ran out of GPU WUs very quickly.

The question, why didn't these setting set up enough buffer for the GPU WUs? Is there a setting I've overlooked?

Maybe relevant, my GPU out paces my eight CPU cores by about 5x based on RAC.
ID: 1662893 · Report as offensive
Profile Jimbocous Project Donor
Volunteer tester
Avatar

Send message
Joined: 1 Apr 13
Posts: 1849
Credit: 268,616,081
RAC: 1,349
United States
Message 1662902 - Posted: 9 Apr 2015, 6:52:41 UTC - in response to Message 1662893.  
Last modified: 9 Apr 2015, 7:05:26 UTC

Question relevant to server downtime and BOINC.

I have my minimum work buffer and max additional buffer both set to 10 days. I have enough WUs on hand to keep CPU crunching for days. But, I ran out of GPU WUs very quickly.

The question, why didn't these setting set up enough buffer for the GPU WUs? Is there a setting I've overlooked?

Maybe relevant, my GPU out paces my eight CPU cores by about 5x based on RAC.

I think that regardless of those settings, you are limited to a max of 100 WUs per CPU/GPU. So unless the machine processes slowly, that limit will kick in before your cache settings ever come into play.

Case in point, I have one machine with 3 GPUs, and I permit 2 tasks simultaneously per GPU, for a max of 6 tasks in process at once. No matter my other settings, I will never have more than 300 GPU tasks on the machine, 6 working and 294 waiting to run, which is about 8-12 hours work.

Conversely, my CPU is slow enough (different machine) that I will never hit the full 100 tasks, as they would take longer than the 10 days I have selected as the cache. Usually see about 85 tasks on that machine, max. And this machine, with a single GPU running two simultaneous tasks consistently has 100 tasks, 2 working and 98 waiting.
ID: 1662902 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13269
Credit: 208,696,464
RAC: 304
Australia
Message 1662920 - Posted: 9 Apr 2015, 7:44:30 UTC - in response to Message 1662902.  

Out of GPU work on both systems, 3 CPU WUs still running on the i7. Another hour or so & i'll be completely out of work there.
My E6400 is so slow I've still got several days worth of CPU work there.
Grant
Darwin NT
ID: 1662920 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 24025
Credit: 261,360,520
RAC: 489
Australia
Message 1662956 - Posted: 9 Apr 2015, 9:16:53 UTC

15mins more and my 2 rigs will be totally onto backup work. :-(

Cheers.
ID: 1662956 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13269
Credit: 208,696,464
RAC: 304
Australia
Message 1662963 - Posted: 9 Apr 2015, 9:32:16 UTC - in response to Message 1662956.  
Last modified: 9 Apr 2015, 9:32:48 UTC

15mins more and my 2 rigs will be totally onto backup work. :-(

Cheers.

I'm using it as a chance to reduce my power bill. Dropped from 420W or so down to 185W, the old Core 2 Duo still chugging along.
UPS run time has gone from less than 10min to over 30min.
Grant
Darwin NT
ID: 1662963 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 24025
Credit: 261,360,520
RAC: 489
Australia
Message 1662970 - Posted: 9 Apr 2015, 9:44:09 UTC - in response to Message 1662963.  

15mins more and my 2 rigs will be totally onto backup work. :-(

Cheers.

I'm using it as a chance to reduce my power bill. Dropped from 420W or so down to 185W, the old Core 2 Duo still chugging along.
UPS run time has gone from less than 10min to over 30min.

With the temps dropping up here in the highlands Grant I need all the heat that the rigs can produce at this time of year. ;-)

Cheers.
ID: 1662970 · Report as offensive
Rasputin42
Volunteer tester

Send message
Joined: 25 Jul 08
Posts: 412
Credit: 5,834,661
RAC: 0
United States
Message 1662971 - Posted: 9 Apr 2015, 9:47:02 UTC

Maybe Seti is too big. They should split it into a few INDEPENDENT Projects.
That way,there should allways be some work to crunch.
Seti had reliabilaty issues, ever since i knew about it, years ago.
ID: 1662971 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1662974 - Posted: 9 Apr 2015, 10:00:09 UTC - in response to Message 1662971.  

I think one big problem there is with SETI is the size of the MB task files. They are so small and designed for way back when everyone had modems.

Now they are just swamped sending out all those little files and processing them. I seen the 'big boys' bragging that they do a task every 9s, wow talk about server load. Split,send,receive,verify,assimilate ... hell that takes them more time than the big crunchers do!

Sure I could just say "make the task files bigger" but then begins the monumental task of combining new and old data.
ID: 1662974 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6323
Credit: 106,370,077
RAC: 121
Russia
Message 1662985 - Posted: 9 Apr 2015, 10:53:54 UTC - in response to Message 1662974.  


Sure I could just say "make the task files bigger" but then begins the monumental task of combining new and old data.

It was in plans more than year already. Things going slow these days.
AFAIK that task "biggering" now replaced with new project of SETI 8. Not only bigger data chunk size but maybe some new ways to process that data. Too little details known still.
ID: 1662985 · Report as offensive
Grumpy Swede (Democratic Socialist)
Volunteer tester
Avatar

Send message
Joined: 1 Nov 08
Posts: 8745
Credit: 49,849,242
RAC: 65
Sweden
Message 1662998 - Posted: 9 Apr 2015, 11:51:23 UTC

Everything is running perfectly. Except a few minor issues, like no new work units, and some minor DB issues, but that's not so important.

It's Thursday, and I'm reasonably healthy. :-)
ID: 1662998 · Report as offensive
The Jedi Alliance - Ranger
Avatar

Send message
Joined: 27 Dec 00
Posts: 72
Credit: 60,982,863
RAC: 0
United States
Message 1663036 - Posted: 9 Apr 2015, 13:40:02 UTC

It's 6:40 AM Thursday Berkeley time. GPU's have been out of work since Tuesday afternoon. CPU's will be out of work in 4-5 hours. It's cold outside and my computers aren't generating enough heat to keep my office warm.

I actually got 1 Cuda WU last night. 15 minutes later it was gone...
ID: 1663036 · Report as offensive
FeK9

Send message
Joined: 20 May 99
Posts: 40
Credit: 61,229,677
RAC: 26
South Africa
Message 1663043 - Posted: 9 Apr 2015, 13:51:25 UTC

Ping :)
Noli tangere circulos meos...
ID: 1663043 · Report as offensive
Victor Wedge
Avatar

Send message
Joined: 3 Apr 04
Posts: 28
Credit: 12,569,503
RAC: 0
Message 1663049 - Posted: 9 Apr 2015, 14:01:47 UTC
Last modified: 9 Apr 2015, 14:07:01 UTC

Alright!!!! New GPU WUs just arrived! :-)



EDIT: Aw shucks! I only got 4 Cuda WUs. That will last through my 2nd cup of coffee.
ID: 1663049 · Report as offensive
Previous · 1 . . . 19 · 20 · 21 · 22 · 23 · 24 · Next

Message boards : Number crunching : Panic Mode On (96) Server Problems?


 
©2022 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.