Servers are back on line after Monday's outage

Message boards : News : Servers are back on line after Monday's outage
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

AuthorMessage
Grasnek

Send message
Joined: 26 Feb 00
Posts: 7
Credit: 365,990
RAC: 0
Netherlands
Message 1324517 - Posted: 4 Jan 2013, 9:28:41 UTC
Last modified: 4 Jan 2013, 9:32:53 UTC

I recently came back to SETI@Home, so I'm a complete noob in this matter and in need of help/advice.

I upgraded my video card from an nVidia EN9800GTX+ to an nVidia 670GTX+ yesterday. For another project (GPUGRID) everything is working well with the new video card; blocks I submitted since the upgrade have been checked and are fine. With the old card I had no problems processing cuda blocks for SETI either, but with the new card something seems to have gone wrong.

I'm receiving cuda_fermi blocks for my kepler card. Does this have something to do with the scheduler trouble of the moment, or is something else wrong? The website does seem to report my computer correctly as having:

GenuineIntel
Intel(R) Core(TM) i5-3570K CPU @ 3.40GHz [Family 6 Model 58 Stepping 9]
(4 processoren)

NVIDIA GeForce GTX 670 (2048MB) driver: 310.70

Any info or help is much appreciated, but keep in mind I'm not a pro at this yet.

Thanks all!

[edit]
Forgot to mention:
Every downloaded block for cuda_fermi has processed for between 17 and 19 seconds, after which a calculation error was reported. There are 63 blocks with errors because of this.
[/edit]
ID: 1324517 · Report as offensive
Profile ivan
Volunteer tester
Avatar

Send message
Joined: 5 Mar 01
Posts: 783
Credit: 348,560,338
RAC: 223
United Kingdom
Message 1324524 - Posted: 4 Jan 2013, 9:43:45 UTC - in response to Message 1324517.  

[edit]
Forgot to mention:
Every downloaded block for cuda_fermi has processed for between 17 and 19 seconds, after which a calculation error was reported. There are 63 blocks with errors because of this.
[/edit]

Try this: http://setiathome.berkeley.edu/forum_thread.php?id=69735
ID: 1324524 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1324530 - Posted: 4 Jan 2013, 10:33:07 UTC - in response to Message 1324524.  

[edit]
Forgot to mention:
Every downloaded block for cuda_fermi has processed for between 17 and 19 seconds, after which a calculation error was reported. There are 63 blocks with errors because of this.
[/edit]

Try this: http://setiathome.berkeley.edu/forum_thread.php?id=69735


Made the link active.

Claggy
ID: 1324530 · Report as offensive
Grasnek

Send message
Joined: 26 Feb 00
Posts: 7
Credit: 365,990
RAC: 0
Netherlands
Message 1324548 - Posted: 4 Jan 2013, 11:41:52 UTC
Last modified: 4 Jan 2013, 11:48:37 UTC

Thanks Ivan & Claggy

It's not a scheduler problem then. I'll try one of the solutions in the list.


PS. is there a way to force my client to start over on the failed jobs while the scheduler is down? Or can I only wait for the scheduler to come back up later (report failure and ask for new jobs)

Would be good to keep it working on the jobs it already downloaded while we wait.
ID: 1324548 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1324556 - Posted: 4 Jan 2013, 12:10:15 UTC - in response to Message 1324548.  
Last modified: 4 Jan 2013, 12:19:00 UTC

Thanks Ivan & Claggy

It's not a scheduler problem then. I'll try one of the solutions in the list.


PS. is there a way to force my client to start over on the failed jobs while the scheduler is down? Or can I only wait for the scheduler to come back up later (report failure and ask for new jobs)

Would be good to keep it working on the jobs it already downloaded while we wait.

You only need to do the second solution,

You could reset the project and then get your failed work resent (but you'd loose any CPU results that you're completed), then hit update until all your tasks are resent, (you don't need to wait 5 minutes between attempts at resending)
But since you're only got a small cache you'd probably get them all in one attempt, upping to about three days cache might get you some fresh work before the servers go down.

Claggy
ID: 1324556 · Report as offensive
Grasnek

Send message
Joined: 26 Feb 00
Posts: 7
Credit: 365,990
RAC: 0
Netherlands
Message 1324564 - Posted: 4 Jan 2013, 12:53:16 UTC - in response to Message 1324556.  

You only need to do the second solution,

You could reset the project and then get your failed work resent (but you'd loose any CPU results that you're completed), then hit update until all your tasks are resent, (you don't need to wait 5 minutes between attempts at resending)
But since you're only got a small cache you'd probably get them all in one attempt, upping to about three days cache might get you some fresh work before the servers go down.

Claggy


I added the "CUDA_GRID_SIZE_COMPAT = 1" value as shown in the screenshots. Will that suffice? It's not clear to me whether I'm supposed to do all A B and C steps listed under the Kepler compatibility section, or just pick one of them to make it work. I'd rather not downgrade my drivers. Shame that topic is closed for comments/questions, else I'd have posted it there by now.

I decided to just report the CUDA errors and CPU work I had done. I thought the servers were down already, but apparently they weren't, so I just got some new jobs instead.

I increased the cache to 3 days.

I have some other questions about configuring projects/BOINC, but as I'm offtopic already here, I guess I should start finding the right forum posts to ask away. After all I already found out my problem wasn't scheduler related.
ID: 1324564 · Report as offensive
Tom*

Send message
Joined: 12 Aug 11
Posts: 127
Credit: 20,769,223
RAC: 9
United States
Message 1324656 - Posted: 4 Jan 2013, 18:32:49 UTC
Last modified: 4 Jan 2013, 18:33:52 UTC

UPDATE: the outage scheduled for the weekend of the 4th-6th has been cancelled so normal operations shall continue. We still plan to have our air conditioner repaired and will be completely off line from January 14th to January 15th during that time.

For those who havn't started at the beginning of this thread

Thank You Matt
ID: 1324656 · Report as offensive
Profile jerbialdo
Volunteer tester
Avatar

Send message
Joined: 17 Dec 12
Posts: 1
Credit: 18,141
RAC: 0
Brazil
Message 1325161 - Posted: 6 Jan 2013, 4:16:04 UTC - in response to Message 1320226.  
Last modified: 6 Jan 2013, 4:17:43 UTC

Em que data (date) o seti@home volta ao normal?
[B]JERBIALDO [/ b]
ID: 1325161 · Report as offensive
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1325178 - Posted: 6 Jan 2013, 6:06:53 UTC - in response to Message 1325161.  

Em que data (date) o seti@home volta ao normal?

Teoricamente já esta normal, pelo menos esta funcionando com o limite de 100WU por host.

ID: 1325178 · Report as offensive
The Mom

Send message
Joined: 16 Aug 10
Posts: 4
Credit: 10,886,625
RAC: 20
United States
Message 1327217 - Posted: 12 Jan 2013, 23:41:14 UTC - in response to Message 1327204.  

Ditto here. My completed units are stacking up and not getting uploaded even if I click "update" When I click on "your results" it says I received 20 work units yesterday. I know I had completed units yesterday and those weren't uploaded.
ID: 1327217 · Report as offensive
Profile DeD
Avatar

Send message
Joined: 7 Jan 13
Posts: 6
Credit: 53,853
RAC: 0
United States
Message 1327231 - Posted: 13 Jan 2013, 0:12:08 UTC

Having the same issue, I cannot upload or get new tasks. Is SETI dying?
ID: 1327231 · Report as offensive
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 28396
Credit: 53,134,872
RAC: 32
United States
Message 1327259 - Posted: 13 Jan 2013, 1:45:36 UTC

crickets are maxed out. you have to wait your turn.

ID: 1327259 · Report as offensive
elvis_666

Send message
Joined: 11 Dec 00
Posts: 1
Credit: 170,851
RAC: 0
Netherlands
Message 1327467 - Posted: 13 Jan 2013, 22:12:05 UTC - in response to Message 1327457.  

Have been waiting for work all day now....
ID: 1327467 · Report as offensive
Profile vmfa531
Avatar

Send message
Joined: 10 Nov 01
Posts: 1
Credit: 17,379,180
RAC: 0
United States
Message 1327478 - Posted: 13 Jan 2013, 22:23:20 UTC

With all of these power outages I have read about recently cursing the eastern limits of the Berkeley campus I can't help but wonder if Pacific Graft & Extortion has a personal grudge against the university.



B.O.H.I.C.A.
ID: 1327478 · Report as offensive
Profile Jeroen Zijlstra
Avatar

Send message
Joined: 28 Jan 00
Posts: 12
Credit: 1,297,570
RAC: 2
Netherlands
Message 1327494 - Posted: 13 Jan 2013, 23:48:50 UTC - in response to Message 1327457.  

Same for me , Please.
With a RAC of 1000 I do more then 100 WU a day.
CPU work continues, but the GPU is getting bored.

ID: 1327494 · Report as offensive
William McCausland

Send message
Joined: 11 Aug 11
Posts: 1
Credit: 66,872
RAC: 0
United States
Message 1327579 - Posted: 14 Jan 2013, 4:08:40 UTC - in response to Message 1320226.  

Am I safe in saying that the Seti@Home servers are now down for the next few days? Not able to get any work.

- William McCausland
Kelso, WA
January 13, 2013 -- 8:10 pm PST (GMT -0800)
ID: 1327579 · Report as offensive
Profile DeD
Avatar

Send message
Joined: 7 Jan 13
Posts: 6
Credit: 53,853
RAC: 0
United States
Message 1327606 - Posted: 14 Jan 2013, 5:57:56 UTC - in response to Message 1327579.  

I'm trying to figure it out too, no news on this.. maybe they found a signal ;)

I also heard that funding might have been cut... not sure rumor for now tho
ID: 1327606 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1327669 - Posted: 14 Jan 2013, 13:09:57 UTC - in response to Message 1327579.  

Am I safe in saying that the Seti@Home servers are now down for the next few days? Not able to get any work.

- William McCausland
Kelso, WA
January 13, 2013 -- 8:10 pm PST (GMT -0800)

You would be correct. Today they will be doing repairs to the server closet air conditioner. It will take two days ( hopefully ) And probally Wednesday they will do the weekly outage.

When they do get back online expect the pipes to be plugged up for a day or two.
[/quote]

Old James
ID: 1327669 · Report as offensive
David S
Volunteer tester
Avatar

Send message
Joined: 4 Oct 99
Posts: 18352
Credit: 27,761,924
RAC: 12
United States
Message 1327677 - Posted: 14 Jan 2013, 14:03:02 UTC - in response to Message 1327669.  

You would be correct. Today they will be doing repairs to the server closet air conditioner. It will take two days ( hopefully ) And probally Wednesday they will do the weekly outage.

When they do get back online expect the pipes to be plugged up for a day or two.

I would think that, depending on just when the AC repairs are finished, they will probably just do the weekly maintenance when they bring everything up but before going live with it. With any luck, that will still be Tuesday.

(Or, the way luck sometimes runs, the AC repairs will take longer than two days and they'll be down all week or longer.)

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

ID: 1327677 · Report as offensive
Jeff Cobb Project Donor
Volunteer moderator
Project administrator
Project developer
Project scientist

Send message
Joined: 1 Mar 99
Posts: 122
Credit: 40,367
RAC: 0
United States
Message 1327709 - Posted: 15 Jan 2013, 22:53:08 UTC

The AC repairs were successful. We then performed our weekly outage tasks and are coming back on line.
ID: 1327709 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

Message boards : News : Servers are back on line after Monday's outage


 
©2021 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.