| Author |
Message |
|---|
|
|
WinterKnight,
Let me guess, it's the quad-core (with the GTX240) that's not getting work i.e. you are getting the "Message from server: This computer has reached a limit on tasks in progress' message...
I have a BAD feeling about this...
Oz
You're right again, the lucky Pent M always gets work. I cannot remember when it has never had work. On time during an outage it was at 97% of the way through it's last task when the project came back up and it got work on it's first request.
Yeah, my Celeron 1.6Ghz did get new work, my other two much faster computers only got the error message.
____________
/The grumpy old Swede.
"I'm so old, that 98% of all trees in the forest, are younger than I am" |
|
|
|
|
|
If Dr A hadn't got his coding so miserably wrong and the estimates were correct, my quad with gpu actually needs gpu tasks also.
It has not asked or received any gpu tasks today, but the true amount of work 'in progress' for the gpu is less than 1.5 days, half my expected cache. |
|
|
|
|
|
All this makes me wonder if it makes any sense to still keep buying new hardware to my computers.
Really, I was about to buy 3 new cuda cards. I won´t if my current computers can´t have any cache to work with during 3/4 day servers outage. 3 more cards to do.. nothing?
I was expecting to fill my caches after all this long outage.
Not good.
____________
 |
|
|
|
|
|
I just uploaded 6 WUs and got 3 more in return. At this rate this machine will be out of work by this afternoon. Didn't get all that much after uploading a boatload of WUs when the servers came back up this morning. If we're going to see 3 day outages and then stingy downloads I may have to find another project to keep this one cruncher busy. It doesn't seem to work well with Rosetta (lots of idle process time for some reason).
____________
 |
|
|
|
|
All this makes me wonder if it makes any sense to still keep buying new hardware to my computers.
Really, I was about to buy 3 new cuda cards. I won´t if my current computers can´t have any cache to work with during 3/4 day servers outage. 3 more cards to do.. nothing?
I was expecting to fill my caches after all this long outage.
Not good.
I am gonna wait and let things shake out a bit and wait for some official explanation. Then I will make some decisions.
If either Boinc or Seti is trying to tell some of us that we are 'contributing too much' to the project, that can be changed....user-side.
____________
******
"Ask not, what your kitty can do for you. Ask what you can do for your kitty."
As it is kitten, so shall it be done.
 |
|
|
|
|
|
Yes, this seems like limiting college entrance to those who rode to high school on the "short bus" (from a which machines get work units standpoint)
Oz
____________
|
|
|
|
|
|
Psst...
Put this in your cc_config in the options part and exit and restart BOINC...
<zero_debts>1</zero_debts>
It might be a debt issue...
____________
Clk2HlpSetiCty:::PayIt4ward
 |
|
|
|
|
Psst...
Put this in your cc_config in the options part and exit and restart BOINC...
<zero_debts>1</zero_debts>
It might be a debt issue...
Hiya John.....
But, NOT.
I only run Seti, so I have no debt to fill on any other project.
____________
******
"Ask not, what your kitty can do for you. Ask what you can do for your kitty."
As it is kitten, so shall it be done.
|
|
|
 rebest Volunteer tester
Send message
Joined: 16 Apr 00
Posts: 1291
Credit: 25,474,029
RAC: 30,349
|
WinterKnight,
Let me guess, it's the quad-core (with the GTX240) that's not getting work i.e. you are getting the "Message from server: This computer has reached a limit on tasks in progress' message...
I have a BAD feeling about this...
Oz
You're right again, the lucky Pent M always gets work. I cannot remember when it has never had work. On time during an outage it was at 97% of the way through it's last task when the project came back up and it got work on it's first request.
Yeah, my Celeron 1.6Ghz did get new work, my other two much faster computers only got the error message.
It's not just heavy duty crunchers being held up. My Core2Duo with no GPU is getting the message, too.
____________
Join the PACK! |
|
|
|
|
Psst...
Put this in your cc_config in the options part and exit and restart BOINC...
<zero_debts>1</zero_debts>
It might be a debt issue...
Why not put <work_fetch_debug> in there while you're at it, and find out what it really is? |
|
|
|
|
Psst...
Put this in your cc_config in the options part and exit and restart BOINC...
<zero_debts>1</zero_debts>
It might be a debt issue...
Hiya John.....
But, NOT.
I only run Seti, so I have no debt to fill on any other project.
But debt will still build...can't hurt, ya know....
'ere's lookin' at you, kid....
____________
Clk2HlpSetiCty:::PayIt4ward
|
|
|
|
|
All this makes me wonder if it makes any sense to still keep buying new hardware to my computers.
Really, I was about to buy 3 new cuda cards. I won´t if my current computers can´t have any cache to work with during 3/4 day servers outage. 3 more cards to do.. nothing?
I was expecting to fill my caches after all this long outage.
Not good.
<conspiracy mode>
Well, I just figured out what it's all about.
DA wants of course as a bragging thing, as many computers as possible from all over the world using BOINC. If everybody has the right to a 10 days cache, bearing in mind the GPU introduction as well as more powerful CPU's into crunching, there wouldn't be any need for so many computers at all.
So, the only way to increase the number of hosts using BOINC, is to limit how much work each computer can do.
</conspiracy mode>
____________
/The grumpy old Swede.
"I'm so old, that 98% of all trees in the forest, are younger than I am" |
|
|
RaistmerVolunteer developer
Volunteer tester
Send message
Joined: 16 Jun 01
Posts: 2547
Credit: 25,761,883
RAC: 38,757
|
|
Well, what is going on :
02/07/2010 21:52:15 SETI@home update requested by user
02/07/2010 21:52:15 [wfd] Request work fetch: project updated by user
02/07/2010 21:52:16 [wfd]: work fetch start
02/07/2010 21:52:16 SETI@home [wfd] FS: 1.000000 = 10000.000000/10000.000000
02/07/2010 21:52:16 SETI@home chosen: major shortfall ATI GPU: 1.00 inst, 98684.98 sec
02/07/2010 21:52:16 [wfd] ------- start work fetch state -------
02/07/2010 21:52:16 [wfd] target work buffer: 17280.00 + 86400.00 sec
02/07/2010 21:52:16 [wfd] CPU: shortfall 0.00 nidle 0.00 saturated 168486.60 busy 0.00 RS fetchable 10000.00 runnable 10000.00
02/07/2010 21:52:16 SETI@home [wfd] CPU: fetch share 1.00 LTD 0.00 backoff dt 0.00 int 0.00
02/07/2010 21:52:16 [wfd] ATI GPU: shortfall 98684.98 nidle 0.00 saturated 4995.02 busy 0.00 RS fetchable 10000.00 runnable 1000.00
02/07/2010 21:52:16 SETI@home [wfd] ATI GPU: fetch share 1.00 LTD 0.00 backoff dt 0.00 int 0.00
02/07/2010 21:52:16 SETI@home [wfd] overall LTD -171776.44
02/07/2010 21:52:16 [wfd] ------- end work fetch state -------
02/07/2010 21:52:16 SETI@home [wfd] request: 98684.98 sec CPU (0.00 sec, 0.00) ATI GPU (98684.98 sec, 1.00)
02/07/2010 21:52:16 SETI@home Sending scheduler request: Requested by user.
02/07/2010 21:52:16 SETI@home Requesting new tasks for GPU
02/07/2010 21:52:18 [wfd] Request work fetch: Backoff ended for Milkyway@home
02/07/2010 21:52:24 [wfd] Request work fetch: application exited
02/07/2010 21:52:24 SETI@home [wfd] FS: 1.000000 = 10000.000000/10000.000000
02/07/2010 21:52:47 SETI@home Scheduler request completed: got 0 new tasks
02/07/2010 21:52:47 SETI@home Message from server: Project has no tasks available
02/07/2010 21:52:47 SETI@home [wfd] backing off ATI GPU 24 sec
02/07/2010 21:52:47 [wfd] Request work fetch: RPC complete
02/07/2010 21:52:52 [wfd]: work fetch start
02/07/2010 21:52:52 [wfd] ------- start work fetch state -------
02/07/2010 21:52:52 [wfd] target work buffer: 17280.00 + 86400.00 sec
02/07/2010 21:52:52 [wfd] CPU: shortfall 0.00 nidle 0.00 saturated 168460.61 busy 0.00 RS fetchable 0.00 runnable 10000.00
02/07/2010 21:52:52 SETI@home [wfd] CPU: fetch share 0.00 LTD 0.00 backoff dt 0.00 int 0.00 (comm deferred)
02/07/2010 21:52:52 [wfd] ATI GPU: shortfall 98713.58 nidle 0.00 saturated 4966.42 busy 0.00 RS fetchable 1000.00 runnable 1000.00
02/07/2010 21:52:52 SETI@home [wfd] ATI GPU: fetch share 0.00 LTD 0.00 backoff dt 18.58 int 60.00 (comm deferred)
02/07/2010 21:52:52 SETI@home [wfd] overall LTD -171745.68
02/07/2010 21:52:52 [wfd] ------- end work fetch state -------
02/07/2010 21:52:57 [wfd] Request work fetch: RPC complete
02/07/2010 21:52:58 [wfd] Request work fetch: Backoff ended for SETI@home
02/07/2010 21:53:00 SETI@home [wfd] FS: 1.000000 = 10000.000000/10000.000000
02/07/2010 21:53:03 [wfd]: work fetch start
02/07/2010 21:53:03 SETI@home [wfd] FS: 1.000000 = 10000.000000/10000.000000
02/07/2010 21:53:03 [wfd] ------- start work fetch state -------
02/07/2010 21:53:03 [wfd] target work buffer: 17280.00 + 86400.00 sec
02/07/2010 21:53:03 [wfd] CPU: shortfall 0.00 nidle 0.00 saturated 168455.45 busy 0.00 RS fetchable 10000.00 runnable 10000.00
02/07/2010 21:53:03 SETI@home [wfd] CPU: fetch share 1.00 LTD 0.00 backoff dt 0.00 int 0.00
02/07/2010 21:53:03 [wfd] ATI GPU: shortfall 98506.22 nidle 0.00 saturated 5173.78 busy 0.00 RS fetchable 0.00 runnable 1000.00
02/07/2010 21:53:03 SETI@home [wfd] ATI GPU: fetch share 0.00 LTD 0.00 backoff dt 8.19 int 60.00
02/07/2010 21:53:03 SETI@home [wfd] overall LTD -171736.44
02/07/2010 21:53:03 [wfd] ------- end work fetch state -------
02/07/2010 21:53:03 [wfd] No project chosen for work fetch
Last string just wonderful :/ |
|
|
|
|
All this makes me wonder if it makes any sense to still keep buying new hardware to my computers.
Really, I was about to buy 3 new cuda cards. I won´t if my current computers can´t have any cache to work with during 3/4 day servers outage. 3 more cards to do.. nothing?
I was expecting to fill my caches after all this long outage.
Not good.
I feel your pain. I got a new rig and a new fermi and wanted to use it to crunch. Since I got it months ago, it has maybe 20 hours of compute time on SETI.
____________
|
|
|
RaistmerVolunteer developer
Volunteer tester
Send message
Joined: 16 Jun 01
Posts: 2547
Credit: 25,761,883
RAC: 38,757
|
|
And more:
02/07/2010 21:54:33 SETI@home [wfd] request: 98603.21 sec CPU (0.00 sec, 0.00) ATI GPU (98603.21 sec, 1.00)
02/07/2010 21:54:34 SETI@home Sending scheduler request: To fetch work.
02/07/2010 21:54:34 SETI@home Requesting new tasks for GPU
02/07/2010 21:54:39 SETI@home Scheduler request completed: got 0 new tasks
02/07/2010 21:54:39 SETI@home Message from server: No work sent
02/07/2010 21:54:39 SETI@home Message from server: This computer has reached a limit on tasks in progress
02/07/2010 21:54:39 SETI@home [wfd] backing off ATI GPU 160 sec
02/07/2010 21:54:39 [wfd] Request work fetch: RPC complete
Did debug messages say more to someone? ...
[EDIT: Maybe MW should donate SETI@home some money too cause they use my GPU in much bigger share now than I would like ;D ] |
|
|
|
|
02/07/2010 21:53:03 [wfd] No project chosen for work fetch
Last string just wonderful :/
Yes, it proves it's nothing to do with the servers.....
They're never going to give you anything if you don't ask. |
|
|
RaistmerVolunteer developer
Volunteer tester
Send message
Joined: 16 Jun 01
Posts: 2547
Credit: 25,761,883
RAC: 38,757
|
02/07/2010 21:53:03 [wfd] No project chosen for work fetch
Last string just wonderful :/
Yes, it proves it's nothing to do with the servers.....
They're never going to give you anything if you don't ask.
IMO it's just ATI backoff, look next post. |
|
|
|
|
|
When seeing the message of task limit, how many tasks are currently in the queue? What is the threshold?
I'm thinking 20, because I downloaded 20 on an idle PC and then when I requested more, I got the message.
On another one, I have 20 either ready to start or running and get the same message. Once this completes the next runnin WU, I'll request more and see if it changes the message.
____________
|
|
|
|
|
Psst...
Put this in your cc_config in the options part and exit and restart BOINC...
<zero_debts>1</zero_debts>
It might be a debt issue...
At the risk of a stupid question, for some reason when I search for the cc_config file on this machine I find nothing. A search for "seti" comes up empty. And BOINC and SETI@Home are running on this machine. ???
____________
|
|
|
|
|
02/07/2010 21:53:03 [wfd] No project chosen for work fetch
Last string just wonderful :/
Yes, it proves it's nothing to do with the servers.....
They're never going to give you anything if you don't ask.
And where is this coming from?
I am getting it on one of my rigs, which was running Cuda Seti until early this morning when it finally ran out.
____________
******
"Ask not, what your kitty can do for you. Ask what you can do for your kitty."
As it is kitten, so shall it be done.
|
|
|
