Message from server: (reached daily quota of 100 tasks)

Message boards : Number crunching : Message from server: (reached daily quota of 100 tasks)
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · 4 . . . 27 · Next

AuthorMessage
woodenboatguy

Send message
Joined: 10 Nov 00
Posts: 368
Credit: 3,969,364
RAC: 0
Canada
Message 1002251 - Posted: 9 Jun 2010, 22:13:25 UTC
Last modified: 9 Jun 2010, 22:13:54 UTC

Beg pardon?

I've seen where a machine gets throttled back from downloading when it has reported too many errors.

I've never seen the rig report the above however. Does that mean it's just running "too fast"?

Can you ever be running "too fast"?

And, is "too fast" like "too rich" or "too thin" or "too hot"?

At least I can then qualify for one out of four if that's the case!

Regards,
ID: 1002251 · Report as offensive
W5DMG - Dave

Send message
Joined: 19 May 99
Posts: 155
Credit: 33,162,251
RAC: 0
United States
Message 1002264 - Posted: 9 Jun 2010, 22:30:53 UTC - in response to Message 1002251.  
Last modified: 9 Jun 2010, 22:33:51 UTC

You are not the only one, if you would kindly take a look at the Panic Mode On 32 Server problems link you will see others are having the same issue.
ID: 1002264 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1002275 - Posted: 9 Jun 2010, 22:42:24 UTC

Others have reported this, in all sorts of threads. Since this one has the subject in the title, let's gather here.

There have always been quotas, but since they are 'per core', and each GPU was counted as something like five cores, they made little practical difference.

Worse, quotas are very quickly restored by successful results. That makes no sense when there are multiple applications floating around. We went through a phase one Christmas when a whole clump of hosts started erroring out every Astropulse task. Quota should have stopped that, but as few as 8 successful MB tasks could remove the block - and the download pipe was saturated for days with wasted (and hence resent) AP jobs.

There's been an experiment running at Beta for a while, where you start with a quota of 100 per application version. That's 100 for CUDA, 100 for CPU/MB, 100 for AP and so on. Plus, when one of your returned results validates, you get a bonus task added to your quota so you can replace it.

The idea is to allow hosts which are running properly as much work as they need, but hosts which fail to return valid work - like the Fermis with the pseudo -9 overflows - get throttled back more quickly than before.

It's all a bit experimental, even at Beta: and I have no idea whether they've imported the whole package into the main project (with or without further tweaks). But something like that to restrain the runaway cudas has been sorely needed, and this may just be it.
ID: 1002275 · Report as offensive
Profile SciManStev Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Jun 99
Posts: 6651
Credit: 121,090,076
RAC: 0
United States
Message 1002284 - Posted: 9 Jun 2010, 23:05:48 UTC

For what it's worth, my GPU's just ran dry. The 100 WU quota may be the reason, but they did not replenish after the last outage.

Steve
Warning, addicted to SETI crunching!
Crunching as a member of GPU Users Group.
GPUUG Website
ID: 1002284 · Report as offensive
Profile Geek@Play
Volunteer tester
Avatar

Send message
Joined: 31 Jul 01
Posts: 2467
Credit: 86,146,931
RAC: 0
United States
Message 1002285 - Posted: 9 Jun 2010, 23:09:24 UTC
Last modified: 9 Jun 2010, 23:15:44 UTC

Something about this does not look right. Have we changed quota amounts?

And my GPU does NOT do AP, that is done on my CPU's.

6/9/2010 5:59:36 PM SETI@home Requesting new tasks for GPU
6/9/2010 5:59:43 PM SETI@home Scheduler request completed: got 0 new tasks
6/9/2010 5:59:43 PM SETI@home Message from server: No work sent
6/9/2010 5:59:43 PM SETI@home Message from server: No work is available for Astropulse v5
6/9/2010 5:59:43 PM SETI@home Message from server: (reached daily quota of 100 tasks)


I thought a Q6600 (quad) should get 400 wu per day.
I have NOT received 100 AP wu today, in fact I have received 0.
Boinc....Boinc....Boinc....Boinc....
ID: 1002285 · Report as offensive
woodenboatguy

Send message
Joined: 10 Nov 00
Posts: 368
Credit: 3,969,364
RAC: 0
Canada
Message 1002289 - Posted: 9 Jun 2010, 23:15:51 UTC - in response to Message 1002285.  

Something about this does not look right. Have we changed quota amounts?

And my GPU does NOT do AP, that is done on my CPU's.

6/9/2010 5:59:36 PM SETI@home Requesting new tasks for GPU
6/9/2010 5:59:43 PM SETI@home Scheduler request completed: got 0 new tasks
6/9/2010 5:59:43 PM SETI@home Message from server: No work sent
6/9/2010 5:59:43 PM SETI@home Message from server: No work is available for Astropulse v5
6/9/2010 5:59:43 PM SETI@home Message from server: (reached daily quota of 100 tasks)



That's the messaging that I got exactly.

BTW, thanks Dave. Hadn't made my rounds yet of all the threads. Just smiled actually when I saw this but it actually seems to be something on the go presently I take it.

Regards,
ID: 1002289 · Report as offensive
Profile Geek@Play
Volunteer tester
Avatar

Send message
Joined: 31 Jul 01
Posts: 2467
Credit: 86,146,931
RAC: 0
United States
Message 1002307 - Posted: 9 Jun 2010, 23:59:07 UTC

New line is being displayed..............

6/9/2010 6:54:30 PM SETI@home Requesting new tasks for GPU
6/9/2010 6:54:41 PM SETI@home Scheduler request completed: got 0 new tasks
6/9/2010 6:54:41 PM SETI@home Message from server: No work sent
6/9/2010 6:54:41 PM SETI@home Message from server: No work is available for Astropulse v5
6/9/2010 6:54:41 PM SETI@home Message from server: No work available for the applications you have selected. Please check your preferences on the web site.
6/9/2010 6:54:41 PM SETI@home Message from server: (reached daily quota of 100 tasks)


Boinc....Boinc....Boinc....Boinc....
ID: 1002307 · Report as offensive
woodenboatguy

Send message
Joined: 10 Nov 00
Posts: 368
Credit: 3,969,364
RAC: 0
Canada
Message 1002324 - Posted: 10 Jun 2010, 0:33:47 UTC - in response to Message 1002307.  

Interesting.

I wonder if they've reasserted some of the quota programming and thereby made the distribution of units more equitable (assuming the slower crunchers were getting out muscled by the big ones - I could be completely in error about that).

Regards,
ID: 1002324 · Report as offensive
Profile Sutaru Tsureku
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1002340 - Posted: 10 Jun 2010, 1:28:57 UTC - in response to Message 1002275.  

Others have reported this, in all sorts of threads. Since this one has the subject in the title, let's gather here.

There have always been quotas, but since they are 'per core', and each GPU was counted as something like five cores, they made little practical difference.

Worse, quotas are very quickly restored by successful results. That makes no sense when there are multiple applications floating around. We went through a phase one Christmas when a whole clump of hosts started erroring out every Astropulse task. Quota should have stopped that, but as few as 8 successful MB tasks could remove the block - and the download pipe was saturated for days with wasted (and hence resent) AP jobs.

There's been an experiment running at Beta for a while, where you start with a quota of 100 per application version. That's 100 for CUDA, 100 for CPU/MB, 100 for AP and so on. Plus, when one of your returned results validates, you get a bonus task added to your quota so you can replace it.

The idea is to allow hosts which are running properly as much work as they need, but hosts which fail to return valid work - like the Fermis with the pseudo -9 overflows - get throttled back more quickly than before.

It's all a bit experimental, even at Beta: and I have no idea whether they've imported the whole package into the main project (with or without further tweaks). But something like that to restrain the runaway cudas has been sorely needed, and this may just be it.


I don't know if this work like you said.

I DLed 100 BETA WUs, BOINC calculated a few WUs, but this 100 WU limit stay (BOINC need to wait 24 hours for again 100 new WUs).

I DLed 100 MAIN WUs (for ~ 22 hours the last one), BOINC calculated a lot of WUs, but don't get new because of a 100 WU limit.

ID: 1002340 · Report as offensive
Profile Sutaru Tsureku
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1002376 - Posted: 10 Jun 2010, 4:27:07 UTC - in response to Message 1002340.  

Only for to understand well..


BOINC DLed at 9 Jun 2010 3:19:30 UTC the last WU. Marked as 'Not in DB'.

Since then BOINC got:
Message from server: No work sent
Message from server: (reached daily quota of 100 tasks)


Normally at 08:00 UTC a new day start for the SETI@home server (so it was in past).
So BOINC could DL (now) 100 new WUs.
..but BOINC get since the last DLed WU only the upper mentioned message.

Now it's over 25 hours ago that BOINC could DL.

Or today at 08:00 UTC BOINC can DL again WUs?


Or it's like this, if BOINC sent a result back of a WU which is marked as 'Not in DB', then this 100 WUs/day will increase?

ID: 1002376 · Report as offensive
Profile Hellsheep
Volunteer tester

Send message
Joined: 12 Sep 08
Posts: 428
Credit: 784,780
RAC: 0
Australia
Message 1002378 - Posted: 10 Jun 2010, 4:31:05 UTC

I'm getting the AP error too, even though i've gotten no AP work, i haven't gotten 400 WU's today for my Q9550 (unless it doesn't count per core anymore).

It seems whenever it checks for new WU's for the CPU and requests them, i get the astropulse v5 error, otherwise GPU is fine. I can download normal CPU tasks fine every so often though?
ID: 1002378 · 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 1002386 - Posted: 10 Jun 2010, 4:57:54 UTC - in response to Message 1002376.  

Only for to understand well..

BOINC DLed at 9 Jun 2010 3:19:30 UTC the last WU. Marked as 'Not in DB'.

Since then BOINC got:
Message from server: No work sent
Message from server: (reached daily quota of 100 tasks)


Normally at 08:00 UTC a new day start for the SETI@home server (so it was in past).
So BOINC could DL (now) 100 new WUs.
..but BOINC get since the last DLed WU only the upper mentioned message.

Now it's over 25 hours ago that BOINC could DL.

Or today at 08:00 UTC BOINC can DL again WUs?


Or it's like this, if BOINC sent a result back of a WU which is marked as 'Not in DB', then this 100 WUs/day will increase?

We really don't know what the servers are going to do, both Richard Haselgrove and I have noted that it appears some of the features being tested at SETI Beta have been adopted here in part. That kind of patching isn't easy and often produces unexpected side effects.

What we can possibly do to help is to observe carefully how the system is operating. Then we could pass information back to the project if there's a clear flaw. One day of observation is not enough, there are always temporary effects when a major change is first implemented. Be as patient as you can, and watch carefully.
                                                                Joe
ID: 1002386 · Report as offensive
Profile hiamps
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 4292
Credit: 72,971,319
RAC: 0
United States
Message 1002389 - Posted: 10 Jun 2010, 5:06:53 UTC

This one will surely push everyone to a 10 day cache...Glad I got mine.
Official Abuser of Boinc Buttons...
And no good credit hound!
ID: 1002389 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 1002396 - Posted: 10 Jun 2010, 5:23:01 UTC

Even I'm getting one of these 100 tasks messages, Plus one other saying this:


6/9/2010 9:24:54 PM SETI@home Message from server: Your app_info.xml file doesn't have a usable version of SETI@home Enhanced.


But I haven't seen that or any message in red in almost half an hour(9:24pm PDT).
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1002396 · Report as offensive
Profile Geek@Play
Volunteer tester
Avatar

Send message
Joined: 31 Jul 01
Posts: 2467
Credit: 86,146,931
RAC: 0
United States
Message 1002400 - Posted: 10 Jun 2010, 5:28:18 UTC


Looking at my 4 computers I have thousands of wu's marked "Not in DB" in the Application column.

I fear that something is seriously wrong at Berkeley.
Boinc....Boinc....Boinc....Boinc....
ID: 1002400 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 1002426 - Posted: 10 Jun 2010, 6:16:26 UTC - in response to Message 1002396.  

Even I'm getting one of these 100 tasks messages, Plus one other saying this:


6/9/2010 9:24:54 PM SETI@home Message from server: Your app_info.xml file doesn't have a usable version of SETI@home Enhanced.


But I haven't seen that or any message in red in almost half an hour(9:24pm PDT).

Oopsie, I spoke too soon, It's back. :o
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1002426 · Report as offensive
Profile Gatekeeper
Avatar

Send message
Joined: 14 Jul 04
Posts: 887
Credit: 176,479,616
RAC: 0
United States
Message 1002431 - Posted: 10 Jun 2010, 6:54:39 UTC - in response to Message 1002389.  

This one will surely push everyone to a 10 day cache...Glad I got mine.


Ummm

Pardon my AARP moment here...

If we're going to be blocked out at 100 units per day..how are we ever going to stockpile a 10 day cache??

On the plus side..I'm finally making a dent in my VLAR backlog..down to under 700 of 'em, from a high of over 1350.
ID: 1002431 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1002433 - Posted: 10 Jun 2010, 7:11:12 UTC

Got one thing to say......

THIS SUX.

I have been known to be one of the more patient types when it come to Seti server foibles.

But I got a GPU that is about to be starved outta work because the servers think it should have a quota of 100 WU....

If this is due to the Boinc devs experimenting......

THIS SUX EVEN MORE.

I hope it gets sorted very soon.

Now back to patient kitty mode.

Meow meow.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1002433 · Report as offensive
Profile Hellsheep
Volunteer tester

Send message
Joined: 12 Sep 08
Posts: 428
Credit: 784,780
RAC: 0
Australia
Message 1002435 - Posted: 10 Jun 2010, 7:33:38 UTC - in response to Message 1002389.  

This one will surely push everyone to a 10 day cache...Glad I got mine.


I pushed 10 day cache already, funny thing is. As soon as some WU's started to complete (CUDA one's so they were ~10 minutes each) my quota limit of 100 seems to have disappeared (well the error message anyway).
ID: 1002435 · Report as offensive
Profile Gatekeeper
Avatar

Send message
Joined: 14 Jul 04
Posts: 887
Credit: 176,479,616
RAC: 0
United States
Message 1002439 - Posted: 10 Jun 2010, 7:42:12 UTC

FWIW...

It just rolled over to Thursday morning here in the Pacific Time Zone, and sure enough, BOINC started DL'ing work on my 920 rig for the GPU only, since I have plenty of CPU work.

After DL'ing 117 WU's (of which 23 were VLAR's) I got the dreaded 100 unit limit message.

So, it appears to my (non-techie) way of thinking that the "limit" isn't even 100 per GPU, as I have 2 GPU's on this box.

Like Mr. Segur said several posts above, clearly this needs to be watched and reported to the powers that be.
ID: 1002439 · Report as offensive
1 · 2 · 3 · 4 . . . 27 · Next

Message boards : Number crunching : Message from server: (reached daily quota of 100 tasks)


 
©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.