Message boards :
Number crunching :
Got *much* more work than asked for
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
Author | Message |
---|---|
Juha Send message Joined: 7 Mar 04 Posts: 388 Credit: 1,857,738 RAC: 0 |
It seems to be happening more and more. Just did a 'report only' update on 3751792: I have two 'report only's that worked, with third I got 6 tasks in return. As a workaround I just set my preferences to Run only the selected applications SETI@home Enhanced: no Astropulse: yes If no work for selected applications is available, accept work from other applications? no Use Graphics Processing Unit (GPU) if available no I don't have Astropulse in app_info.xml so I shouldn't get any AP work. We'll see what breaks with these settings. -Juha |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
.. from Messages: The funny thing is, that in most cases your system will not ask for zero seconds of work by itself. All you who managed to get this work due to this flaw have probably pressed Update to see what it did. So is it then really unwanted work? ;-) To say you did it to report the task, that's also not necessary as BOINC will do that eventually all by itself, even when set to NNT. I can't say I can reproduce this, but I don't have a CUDA card. 23-Jan-09 14:27:02|SETI@home|[sched_op_debug] Starting scheduler request 23-Jan-09 14:27:02|SETI@home|Sending scheduler request: Requested by user. 23-Jan-09 14:27:02|SETI@home|CPU work request: 0.00 seconds, 0 instances 23-Jan-09 14:27:07|SETI@home|Scheduler request completed: got 0 new tasks |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
The funny thing is, that in most cases your system will not ask for zero seconds of work by itself. All you who managed to get this work due to this flaw have probably pressed Update to see what it did. So is it then really unwanted work? ;-) No, that's a red herring. Don't let the developers off the hook so easily. For example, this host of mine is perfectly capable of requesting work when it needs it: 22/01/2009 19:13:15|SETI@home|Sending scheduler request: To fetch work However, the server is also capable of sending work to the same host when neither I nor BOINC requested it: 23/01/2009 03:50:45|SETI@home|Sending scheduler request: To report completed tasks No, I wasn't (ab)using the update button at 3:50 AM local time - I was asleep in bed! |
-Bert- Send message Joined: 23 Mar 02 Posts: 152 Credit: 412,754 RAC: 0 |
I suspended the project and pressed No New Work. Then I wanted to report one tiny little WU manually. Wish I never had done that. Good G*d, I now have 10 AP's and almost 40 MB's on my computer. While my cache is set to 1 day. Is already known what is causing this problem? |
Rick A. Sponholz Send message Joined: 13 Jun 99 Posts: 43 Credit: 23,179,807 RAC: 0 |
In my case the HUGE number of work unit downloads was NOT cause by clicking UPDATE. I got 1.10GB worth of work units so far, just by upgrading to BOINC 6.6.2 today. |
Dorphas Send message Joined: 16 May 99 Posts: 118 Credit: 8,007,247 RAC: 0 |
i had seti suspended and won't get new work selected on one box while i was finishing up another project. i hit the update button to get my latest seti totals and suddenly the seti server sent me wu's that were not requested. even said no work requested from my end. but the server sent work anyway. |
W-K 666 Send message Joined: 18 May 99 Posts: 19401 Credit: 40,757,560 RAC: 67 |
.. from Messages: Not true Jord, all my unexpected downloads have been automatic, and I don't even have a CUDA compatible graphics card. |
Rick A. Sponholz Send message Joined: 13 Jun 99 Posts: 43 Credit: 23,179,807 RAC: 0 |
Here's an update of my experience. I upgraded 3 computers from BOINC 6.6.0 to 6.6.2 this morning. I got Mega work unit downloads (1.1GB, & 1.4GB) on two of the three. The third 6.6.2 computer is behaving normally. Makes me wonder if I clicked "Update" on those two machines? On my other eight computers (six on BOINC 6.4.5, two on 6.6.0) no MEGA download of work units, so far. The download problem seams to have been not only SETI work units, but Einstein, and World Community Grid, but NOT GPU GRID. To stop the downloads, I clicked on "No New Tasks", and so far that has worked. |
skildude Send message Joined: 4 Oct 00 Posts: 9541 Credit: 50,759,529 RAC: 60 |
Here's an update of my experience. I upgraded 3 computers from BOINC 6.6.0 to 6.6.2 this morning. I got Mega work unit downloads (1.1GB, & 1.4GB) on two of the three. The third 6.6.2 computer is behaving normally. Makes me wonder if I clicked "Update" on those two machines? On my other eight computers (six on BOINC 6.4.5, two on 6.6.0) no MEGA download of work units, so far. The download problem seams to have been not only SETI work units, but Einstein, and World Community Grid, but NOT GPU GRID. To stop the downloads, I clicked on "No New Tasks", and so far that has worked.So you installed it one at a time and everytime you got the same result...hmmmm Do remember that 6.6.2 is still beta. It is to be used with caution. I realize that some betas are just like a regular release. the 6.6.2 is clearly not one of those. In a rich man's house there is no place to spit but his face. Diogenes Of Sinope |
Dave Stegner Send message Joined: 20 Oct 04 Posts: 540 Credit: 65,583,328 RAC: 27 |
Running 5.10.45 Looks like I got bit by the TOO MUCH WORK bug. Requested 0 seconds, got 11 tasks. 1/23/2009 10:00:01 AM|SETI@home|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 6 completed tasks 1/23/2009 10:00:06 AM|SETI@home|Scheduler request succeeded: got 11 new tasks 1/23/2009 10:00:08 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.160 1/23/2009 10:00:08 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.134 1/23/2009 10:00:10 AM|SETI@home|Temporarily failed download of 15dc08ah.2743.16432.3.8.160: http error 1/23/2009 10:00:10 AM|SETI@home|Backing off 1 min 0 sec on download of 15dc08ah.2743.16432.3.8.160 1/23/2009 10:00:10 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.109 1/23/2009 10:00:12 AM|SETI@home|Finished download of 15dc08ah.2743.16432.3.8.134 1/23/2009 10:00:12 AM|SETI@home|Started download of 15dc08af.31122.62316.9.8.83 1/23/2009 10:00:13 AM|SETI@home|Finished download of 15dc08ah.2743.16432.3.8.109 1/23/2009 10:00:13 AM|SETI@home|Started download of 15dc08ag.2823.16023.4.8.89 1/23/2009 10:00:16 AM|SETI@home|Finished download of 15dc08af.31122.62316.9.8.83 1/23/2009 10:00:16 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.142 1/23/2009 10:00:18 AM|SETI@home|Finished download of 15dc08ag.2823.16023.4.8.89 1/23/2009 10:00:18 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.117 1/23/2009 10:00:29 AM|SETI@home|Finished download of 15dc08ah.2743.16432.3.8.117 1/23/2009 10:00:29 AM|SETI@home|Started download of 15dc08ag.2823.16023.4.8.117 1/23/2009 10:00:32 AM|SETI@home|Finished download of 15dc08ah.2743.16432.3.8.142 1/23/2009 10:00:32 AM|SETI@home|Started download of 15dc08af.31122.62316.9.8.32 1/23/2009 10:00:44 AM|SETI@home|Finished download of 15dc08ag.2823.16023.4.8.117 1/23/2009 10:00:44 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.146 1/23/2009 10:00:47 AM|SETI@home|Finished download of 15dc08af.31122.62316.9.8.32 1/23/2009 10:00:47 AM|SETI@home|Started download of 15dc08ag.2823.16023.4.8.94 1/23/2009 10:00:51 AM|SETI@home|Finished download of 15dc08ah.2743.16432.3.8.146 1/23/2009 10:00:57 AM|SETI@home|Finished download of 15dc08ag.2823.16023.4.8.94 1/23/2009 10:01:11 AM|SETI@home|Started download of 15dc08ah.2743.16432.3.8.160 1/23/2009 10:01:23 AM|SETI@home|Finished download of 15dc08ah.2743.16432.3.8.160 Dave |
skildude Send message Joined: 4 Oct 00 Posts: 9541 Credit: 50,759,529 RAC: 60 |
Running 5.10.45 Ok not sure but it looks like you clicked on the update button. this will send a request for work even though it said requesting 0 seconds work. Not all WU's take the same amount of time. some take an hour some take several Also look at the Report Date. you have 1 week to return them. this means they are short WU's. Most likely you can finish these in less than a day on a 2.4 ghz system. It's understandable that it looks like a lot of work. the reality is its not all that much. In a rich man's house there is no place to spit but his face. Diogenes Of Sinope |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
Ok not sure but it looks like you clicked on the update button. this will send a request for work even though it said requesting 0 seconds work. Yes, skildude, we know that we can cope with the effects of this bug - work will be returned on time, well before deadline, etc.etc. But "Requesting 0 seconds of work" means exactly that: requesting no work at all. It's written that way, it's designed to work that way, and it's worked that way for the last three years in my personal experience. This new behaviour is different, and wrong. It's a bug. Not a fatal one, not the worst bug in the world, but still a bug. It needs fixing, not glossing over. |
Rick A. Sponholz Send message Joined: 13 Jun 99 Posts: 43 Credit: 23,179,807 RAC: 0 |
Don't dismiss this issue. I'm experienced downloading 100's of AP work units (1.5GB) which would be several WEEKS worth of work, not even counting all the other work units that were downloaded. If I hadn't pulled my internet connection to stop the download I'd have had even MORE! |
skildude Send message Joined: 4 Oct 00 Posts: 9541 Credit: 50,759,529 RAC: 60 |
I understand there is a problem but every person that hits the update button is questioning why they got 10-15 WU's. CLearly there is a glitch that downloads 100+ WU's. Yet someone getting a Days worth of work isn't a problem. getting 2-3 weeks worth when you only want 1 day is. I'm not discounting the glitch at all. Just looking at the persons Computer(cpu speed) and the work received doing a bit of math from the work they've completed. I do think some folks are panicing about normal operations when the problem you've mentioned is real and should be looked into. In a rich man's house there is no place to spit but his face. Diogenes Of Sinope |
[B^S] madmac Send message Joined: 9 Feb 04 Posts: 1175 Credit: 4,754,897 RAC: 0 |
|
Kinguni Send message Joined: 15 Feb 00 Posts: 239 Credit: 9,043,007 RAC: 0 |
Just looked at my computers and my AthlonXP 1800 has 11 days worth of work with my cache set at 2 days. My quad has 5 days worth and my D has 7. BOINC versions 6.2.18 and 19. Bug on the Berkeley end for sure. Doesn't matter to me. They'll crunch if needed, if not they'll be aborted I guess. Join Team Starfire BOINC Chat |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
Looking at Scarecrow's 30 day Astropulse graphs, the number of AP tasks in progress has almost doubled over the last month. I can't help but think that's related to what we're seeing here. |
archae86 Send message Joined: 31 Aug 99 Posts: 909 Credit: 1,582,816 RAC: 0 |
Here is my entry into the "overfulfillment of plan" derby. The host is a Q9550 running stock, with SETI at 4% work allocation. This morning it asked SETI for 1 second of SETI work, and was rewarded with 20 WUs, including 4 Astropulse. BoincView thinks my SETI queue on that host is currently sufficient for 13 days (requested is about 2 days). stoll5 SETI@home 1/23/2009 5:22 Sending scheduler request: To fetch work. Requesting 1 seconds of work, reporting 0 completed tasks stoll5 SETI@home 1/23/2009 5:22 Scheduler request succeeded: got 20 new tasks stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.206 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.189 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.206 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.210 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.189 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.210 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.214 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.211 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.214 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.211 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.212 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.194 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.212 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.194 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.190 stoll5 SETI@home 1/23/2009 5:22 Started download of ap_20dc08ad_B5_P0_00028_20090123_17965.wu stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.190 stoll5 SETI@home 1/23/2009 5:22 Started download of 15dc08af.1756.50864.7.8.156 stoll5 SETI@home 1/23/2009 5:22 Finished download of 15dc08af.1756.50864.7.8.156 stoll5 SETI@home 1/23/2009 5:22 Started download of ap_20dc08ad_B4_P1_00356_20090123_15293.wu stoll5 SETI@home 1/23/2009 5:23 Finished download of ap_20dc08ad_B5_P0_00028_20090123_17965.wu stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.193 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.193 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.197 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.197 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.192 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.192 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.199 stoll5 SETI@home 1/23/2009 5:23 Finished download of ap_20dc08ad_B4_P1_00356_20090123_15293.wu stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.196 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.199 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.177 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.196 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.203 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.177 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.205 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.203 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.205 stoll5 SETI@home 1/23/2009 5:23 Started download of 15dc08af.1756.50864.7.8.183 stoll5 SETI@home 1/23/2009 5:23 Finished download of 15dc08af.1756.50864.7.8.183 Something is broken, it causes harm, it would be good to fix it. |
skildude Send message Joined: 4 Oct 00 Posts: 9541 Credit: 50,759,529 RAC: 60 |
Here is my entry into the "overfulfillment of plan" derby.You appear to have repeatedly abort work on this PC. Any reason? BOINC has a nasty habit of filling your queue with any project available. So when your request was sent in it got enough to fill the queue. You seem to finish the short WU's in 15 minutes and moderately large ones in 40. I don't know if the 1 second request is a glitch. Since your boinc is trying to fill your queue. The problem being discussed pertains to excessive WU's received that are far beyond requested. I'd think you'd be able to finish the 20 non AP WU's in a few hours which shouldn't be that much more than 4%. I can't tell you how many times I get 100 WU's from primegrid or 20+ from Milkyway if I let my overall cache of seti get low. right now I have a stockpile of PG, AIS, and SHA-1 because I screwed up my app_info. It will take a day or so before I get back on my seti's. So no harm In a rich man's house there is no place to spit but his face. Diogenes Of Sinope |
archae86 Send message Joined: 31 Aug 99 Posts: 909 Credit: 1,582,816 RAC: 0 |
You appear to have repeatedly abort work on this PC. Any reason? BOINC has a nasty habit of filling your queue with any project available. So when your request was sent in it got enough to fill the queue. Backwards--I aborted some of the excess after this overfetch, not before. Also, Einstein was up and delivering work, in fact there was a small Einstein fetch request at nearly the same time as this--correctly fulfilled. Your tentative diagnosis does not fit the facts. You seem to finish the short WU's in 15 minutes and moderately large ones in 40. I don't know if the 1 second request is a glitch. Since your boinc is trying to fill your queue. The problem being discussed pertains to excessive WU's received that are far beyond requested. So you don't think getting 20 distinct WUs requiring far over a week of the allocated resource to finish in response to a 1 second request is "excessive WU's received that are far beyond requested"? I think others on this thread will recognize that my event fits the pattern of their concerns. |
©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.