Panic Mode On (42) Server problems

Message boards : Number crunching : Panic Mode On (42) Server problems
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · 9 · 10 . . . 11 · Next

AuthorMessage
-BeNt-
Avatar

Send message
Joined: 17 Oct 99
Posts: 1234
Credit: 10,116,112
RAC: 0
United States
Message 1060317 - Posted: 27 Dec 2010, 22:22:01 UTC - in response to Message 1060201.  

I'll give the SETI team at most 5 to 10 years to fix this problem, or I'll leave the project forever....


Don't see the problem here. I've had pages and pages of stalled transfers before and the always get cleared up. If you keep your cache large enough to bridge through these minor inconveniences, then you will eventually get the credit and have plenty of work.

If you can't stand that, then move to another project. If you keep a couple of projects on your list with zero % resource share, then should the unthinkable happen and you run out of SETI units, your BOINC will switch to one of those other projects.

For me, the happy balance is CUDA for SETI and CPU for AQUA, with Einstein on reserve. Plenty of work, plenty of points.

Just take a chill pill. It's all good.


**sniff sniff** I smell someone who doesn't catch the sarcasm.

Traveling through space at ~67,000mph!
ID: 1060317 · Report as offensive
Kevin Olley

Send message
Joined: 3 Aug 99
Posts: 906
Credit: 261,085,289
RAC: 572
United Kingdom
Message 1060331 - Posted: 27 Dec 2010, 23:39:20 UTC - in response to Message 1060201.  

Don't see the problem here. I've had pages and pages of stalled transfers before and the always get cleared up. If you keep your cache large enough to bridge through these minor inconveniences, then you will eventually get the credit and have plenty of work.

If you can't stand that, then move to another project. If you keep a couple of projects on your list with zero % resource share, then should the unthinkable happen and you run out of SETI units, your BOINC will switch to one of those other projects.

For me, the happy balance is CUDA for SETI and CPU for AQUA, with Einstein on reserve. Plenty of work, plenty of points.

Just take a chill pill. It's all good.


I could do with a bucketful of them:-(

Just been shopping, 2 X ASUS ENGTX470/G/2DI/1280MD5 cards and a 1000W PSU to fit into this machine.

I have rescheduled all SETI GPU units to CPU NNT set and minimal cache for when I start testing, If need be I will try to reschedule some GPU units back as Fermi units if Freds Reschedular will allow me to do so.

I am not familiar with multi graphic card setups, so I was hoping for good access to the message boards if it does not go all to plan. If it does then a supply of SETI Fermi GPU units would be handy.

I have been waiting patiently for the time, the parts and a stable SETI service, Oh well two out of three aint bad:-)


Kevin


ID: 1060331 · Report as offensive
Kevin Olley

Send message
Joined: 3 Aug 99
Posts: 906
Credit: 261,085,289
RAC: 572
United Kingdom
Message 1060485 - Posted: 28 Dec 2010, 11:40:51 UTC - in response to Message 1060331.  



Just been shopping, 2 X ASUS ENGTX470/G/2DI/1280MD5 cards and a 1000W PSU to fit into this machine.

I have rescheduled all SETI GPU units to CPU NNT set and minimal cache for when I start testing, If need be I will try to reschedule some GPU units back as Fermi units if Freds Reschedular will allow me to do so.

I am not familiar with multi graphic card setups, so I was hoping for good access to the message boards if it does not go all to plan. If it does then a supply of SETI Fermi GPU units would be handy.



Got them installed, or more correctly have put them into the case and reinstalled latest video drivers.

I was not certain about SLI so I have put little connector that was supplied with motherboard ASUS M3N-HT DELUXE onto top of graphics cards.

Lost a couple of GPU tasks because I did not shut down SETI when re-installing video drivers.

Only seeing a single video card, both are getting warm and both have fans running so presume both are running.

Re-installed Boinc and Lunatics Win64v0.37, let it have some GPU units, rescheduled them back from CPU as Fermi units and left it running overnight. Thanks Fred.

It is processing 1 unit at a time ie: only seeing a single video card, 6 - 12 Min each, and a short while after I updated (NNT still set) I rechecked my average credit for this machine and it is rising so hopefully I am not erroring out too many units. It would be handy if Tasks were turned back on.

Now this is where it gets very complicated for me, My programing skills are non-existent, I have had a quick look at constructing a cc_config.xml file and got totally lost, and I do not know if I should have the cards SLI'd or not so Help is needed, either here or in a new thread in Questions and answers.


Kevin


ID: 1060485 · Report as offensive
Profile Bernie Vine
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 26 May 99
Posts: 9958
Credit: 103,452,613
RAC: 328
United Kingdom
Message 1060488 - Posted: 28 Dec 2010, 11:58:51 UTC - in response to Message 1060485.  
Last modified: 28 Dec 2010, 11:59:40 UTC

I think this is the thread you want.

forum_thread.php?id=5377

Post 3 from Claggy

Bernie
ID: 1060488 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1060490 - Posted: 28 Dec 2010, 12:11:11 UTC

Ohh..

Project communication failed: attempting access to reference site
Scheduler request failed: Couldn't connect to server
Internet access OK - project servers may be temporarily down.


..and this a few hours before the weekly maintenance/outage.

BOINCs have enough WUs, so no panic here.. ;-)

ID: 1060490 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34379
Credit: 79,922,639
RAC: 80
Germany
Message 1060498 - Posted: 28 Dec 2010, 12:47:54 UTC

Maybe only a hicup.

Just reported a few units and got fresh work.



With each crime and every kindness we birth our future.
ID: 1060498 · Report as offensive
Kevin Olley

Send message
Joined: 3 Aug 99
Posts: 906
Credit: 261,085,289
RAC: 572
United Kingdom
Message 1060500 - Posted: 28 Dec 2010, 13:08:58 UTC - in response to Message 1060488.  

I think this is the thread you want.

forum_thread.php?id=5377

Post 3 from Claggy

Bernie


Just tried it, I am seeing

28/12/2010 12:55:50 Config: use all coprocessors

but only seeing a single GPU unit being processed.

I have tried removing the SLI board no joy, checked the MB Bios and activated SLI Memory no luck, and re-fitted SLI board still only 1 GPU showing.


Kevin


ID: 1060500 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1060509 - Posted: 28 Dec 2010, 13:23:16 UTC - in response to Message 1060500.  
Last modified: 28 Dec 2010, 13:27:59 UTC

I think this is the thread you want.

forum_thread.php?id=5377

Post 3 from Claggy

Bernie


Just tried it, I am seeing

28/12/2010 12:55:50 Config: use all coprocessors

but only seeing a single GPU unit being processed.

I have tried removing the SLI board no joy, checked the MB Bios and activated SLI Memory no luck, and re-fitted SLI board still only 1 GPU showing.


You did try restarting Boinc didn't you?

Try disabling SLI in the Nvidia Control Panel (shouldn't need to do this as the latest drivers are supposed to make this redundant)

Claggy
ID: 1060509 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1060515 - Posted: 28 Dec 2010, 13:33:32 UTC - in response to Message 1060498.  

Maybe only a hicup.

Just reported a few units and got fresh work.


Yes, now it looks good.

ID: 1060515 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1060516 - Posted: 28 Dec 2010, 13:36:56 UTC - in response to Message 1060509.  

I think this is the thread you want.

forum_thread.php?id=5377

Post 3 from Claggy

Bernie


Just tried it, I am seeing

28/12/2010 12:55:50 Config: use all coprocessors

but only seeing a single GPU unit being processed.

I have tried removing the SLI board no joy, checked the MB Bios and activated SLI Memory no luck, and re-fitted SLI board still only 1 GPU showing.


You did try restarting Boinc didn't you?

Try disabling SLI in the Nvidia Control Panel (shouldn't need to do this as the latest drivers are supposed to make this redundant)

Claggy


Maybe we could start a new thread about this topic? (for not to be more offtopic here.. ;-)

Maybe post your first messages (~ 35) of your BOINC Manager.

ID: 1060516 · Report as offensive
Kevin Olley

Send message
Joined: 3 Aug 99
Posts: 906
Credit: 261,085,289
RAC: 572
United Kingdom
Message 1060520 - Posted: 28 Dec 2010, 13:48:05 UTC - in response to Message 1060509.  

I think this is the thread you want.

forum_thread.php?id=5377

Post 3 from Claggy

Bernie


Just tried it, I am seeing

28/12/2010 12:55:50 Config: use all coprocessors

but only seeing a single GPU unit being processed.

I have tried removing the SLI board no joy, checked the MB Bios and activated SLI Memory no luck, and re-fitted SLI board still only 1 GPU showing.


You did try restarting Boinc didn't you?

Try disabling SLI in the Nvidia Control Panel (shouldn't need to do this as the latest drivers are supposed to make this redundant)

Claggy


Fixed it:-)

Dug out the MB book, I have 3 video card slots, I knew that much, apparently if you only want to use two you have to pick the right two, I didn't:-(

I shall leave it running on 1 unit per card until tasks are enabled again so that I know that I am not trashing too many tasks, then I shall try to increase tasks per card.


Kevin


ID: 1060520 · Report as offensive
Kevin Olley

Send message
Joined: 3 Aug 99
Posts: 906
Credit: 261,085,289
RAC: 572
United Kingdom
Message 1060521 - Posted: 28 Dec 2010, 13:52:40 UTC - in response to Message 1060516.  


Maybe we could start a new thread about this topic? (for not to be more offtopic here.. ;-)

Maybe post your first messages (~ 35) of your BOINC Manager.


Sorry, will start new thread if needed later.


Kevin


ID: 1060521 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1060522 - Posted: 28 Dec 2010, 13:57:09 UTC - in response to Message 1060521.  
Last modified: 28 Dec 2010, 13:58:04 UTC

Maybe we could start a new thread about this topic? (for not to be more offtopic here.. ;-)

Maybe post your first messages (~ 35) of your BOINC Manager.

Sorry, will start new thread if needed later.


No problem.

It would be easier to follow/to help. ;-)
ID: 1060522 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1060599 - Posted: 28 Dec 2010, 17:00:16 UTC

Outage coming up soon....
Things have been flowing pretty well, so I hope everybody has been able to get a little cache.
If everything goes as Matt has planned, they should only be down today and back up soon.

See ya on the flip side.

Meow meow.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1060599 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1060606 - Posted: 28 Dec 2010, 17:16:08 UTC - in response to Message 1060604.  

Outage coming up soon....
See ya on the flip side.

Meow meow.


Flop side Mark, we're doing flops here, not flips. :-)

/me is ducking, and goes into hiding until AFTER the project is back again.

That's funny, because the term I was gonna use was...See ya on the flip flop.

But I thought that might be one of those local little sayings and didn't know if it would make sense to others...
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1060606 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1060617 - Posted: 28 Dec 2010, 20:44:54 UTC

OMG....
If they are done playing with the buttons, that is about the shortest outage ever......
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1060617 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1060648 - Posted: 28 Dec 2010, 22:16:27 UTC

Only thing that worries me a bit is Jocelyn is losing ground against Carolyn.
Last look was 17,813 seconds behind and still falling back.
I am hoping this is just due to some temporary things going on in the background since the outage or the large influx of work being reported now that we are back up.
Keep an eye on her.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1060648 · Report as offensive
JohnDK Crowdfunding Project Donor*Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 28 May 00
Posts: 1222
Credit: 451,243,443
RAC: 1,127
Denmark
Message 1062751 - Posted: 2 Jan 2011, 15:35:38 UTC

I'm getting mostly no work sent on tasks requests lately. Is that due to the numbers of APs available going down?
ID: 1062751 · Report as offensive
Profile perryjay
Volunteer tester
Avatar

Send message
Joined: 20 Aug 02
Posts: 3377
Credit: 20,676,751
RAC: 0
United States
Message 1062761 - Posted: 2 Jan 2011, 16:00:47 UTC - in response to Message 1062751.  

It's still topping me off with MB and Cuda work. I noticed maybe a half hour ago that the AP splitters were all down. Don't know how long they have been off but that is probably why we aren't getting new AP work.

The boys should be finishing their coffee right about now and should have everything under control soon.


PROUD MEMBER OF Team Starfire World BOINC
ID: 1062761 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1062768 - Posted: 2 Jan 2011, 16:23:04 UTC - in response to Message 1062761.  

It's still topping me off with MB and Cuda work. I noticed maybe a half hour ago that the AP splitters were all down. Don't know how long they have been off but that is probably why we aren't getting new AP work.

The boys should be finishing their coffee right about now and should have everything under control soon.

On a Sunday??!!

No, they left everything on auto and went home (see Matt's midweek post).

As usual, the AP splitters have finished the currently-online disks before the demand for MB work has exhausted them. No doubt the 'fetch-and-blank' daemons will retrieve more from backing storage when it's needed.

A bad sysadmin spends all their time rushing around doing trivial things like loading disks. A good sysadmin (and we have good ones here) writes programs to handle boring, repetitive chores like that, and then spends time on the things only human beings can do.
ID: 1062768 · Report as offensive
Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · 9 · 10 . . . 11 · Next

Message boards : Number crunching : Panic Mode On (42) Server problems


 
©2024 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.