Panic Mode On (70) Server problems?

Message boards : Number crunching : Panic Mode On (70) Server problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9

AuthorMessage
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22160
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1203801 - Posted: 8 Mar 2012, 18:42:19 UTC

Read carefully -


Thus if you are requesting "GPU only", and there are no WU suitable you will get the dreaded "no tasks available" message, or one of its variants. That said, if you are asking for "any WU I can crunch" preference will be given to satisfying that particular crunchers "best" processor's needs.


Which is exactly what I've been seeing for a few days now - nothing coming through for the CPU (hex core), because the GPU('460) is crunching its assigned tasks before its tanks are filled...
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1203801 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1203806 - Posted: 8 Mar 2012, 18:49:11 UTC - in response to Message 1203801.  


Whatever the problem was, it looks like they managed to sort it out.
Still get the odd "Project has no tasks available" message, but now it's only every now & then, not most of the time & most requests result in work. No longer getting the Scheduler error messages & the network traffic has picked up & smothed out.
Grant
Darwin NT
ID: 1203806 · Report as offensive
LadyL
Volunteer tester
Avatar

Send message
Joined: 14 Sep 11
Posts: 1679
Credit: 5,230,097
RAC: 0
Message 1203822 - Posted: 8 Mar 2012, 19:35:31 UTC - in response to Message 1203816.  


(but still the 100wu server cache is stupid, always empty)


it's called the feeder. and there is a reason for the limited amount and not too fast turnover - something about the i/o load of acessing the database.

With the new servers maybe it's possible to increase the values.
To be fair when we run without limits you don't see the 'no tasks' messgage too often, usually only after maintenace when all the caches are screaming to be filled again.
I'm not the Pope. I don't speak Ex Cathedra!
ID: 1203822 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 1203978 - Posted: 9 Mar 2012, 1:57:28 UTC - in response to Message 1203744.  

I think that Synergy is just overloaded, or the Apache settings are not quite right yet.....dunno, just a guess. Synergy is trying to do a yeoman's duty here....

As stated on the SSP, the purpose of the feeder is "Fills up the scheduler work queue with workunits ready to be sent. The scheduler is usually too busy handling client transactions to maintain such a queue itself."

So is that effect being negated by the scheduling server, scheduler process, and feeder all being on the same machine?
...

I think scheduler processes and feeder must be on the same machine, the "scheduler work queue" is in shared memory.

The project is using FastCGI so it might be possible for the scheduling server (Apache) to be on a different machine. More LAN traffic that way, though.
                                                                  Joe
ID: 1203978 · Report as offensive
Previous · 1 . . . 6 · 7 · 8 · 9

Message boards : Number crunching : Panic Mode On (70) 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.