Message boards :
Number crunching :
10ja10zz. WU's ???
Message board moderation
Author | Message |
---|---|
Bernd Noessler Send message Joined: 15 Nov 09 Posts: 99 Credit: 52,635,434 RAC: 0 |
I have got some of them. http://setiathome.berkeley.edu/workunit.php?wuid=1056302852 They all end with a div by zero exception. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Not got any here, but I'll keep an eye open for them. |
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 |
Hmm, Windows stock applications (both CPU and CUDA) aren't showing an error but quit very soon. See WU 1056302893 for instance. Both Linux and OSX do show the divide by zero, as well as at least one Lunatics Windows CPU app, see task 2584521118. The Windows dump there seems to indicate the error happens while doing baseline smoothing. Joe |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Out of the 4 I have processed so far 2 were completed by the stock app. Workunit: 1056405533 10ja10zz.7361.4578.6.10.227.vlar Workunit: 1056385201 10ja10zz.7361.897.6.10.78 Workunit: 1056325001 - Completed w/o error by Stock app. 10ja10zz.7591.2942.3.10.103.vlar Workunit: 1056308870 - Completed w/o error by Stock app. 10ja10zz.7591.1306.3.10.29.vlar EDIT: Across all of my systems I only found 1 more of these. Workunit: 1056385156 - Already errored by wingmate on anon app. 10ja10zz.7361.897.6.10.65_1 SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Bernd Noessler Send message Joined: 15 Nov 09 Posts: 99 Credit: 52,635,434 RAC: 0 |
This one is from a Windows 7 machine. It looks like a pre stock 6.03 version and gives a lot of debugging output. http://setiathome.berkeley.edu/result.php?resultid=2584566724 How can the Windows stock 6.03 clients (CPU) finish the tasks in 3-5 secs without an error or overflow ? I think they catch the math interrupt and finish without an result. |
VQ-2 Ghost Send message Joined: 18 Jul 02 Posts: 55 Credit: 1,165,715 RAC: 0 |
It seems that this is happening with a lot of us, there's got to be something wrong with these 10ja10zz vlar units. Two of my lowly machines started getting these units that end up with div by zero exception error. |
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 |
This one is from a Windows 7 machine. The WUs are another case where the polyphase filter method of splitting is in use, it's good news that now ends up with correct file size at least. However, the template being used for analysis_cfg clearly indicates they did not intend to actually deliver these WUs. There are multiple items with non-useful parameters, these 3 lines are enough to account for the observed issues: <analysis_fft_lengths>0</analysis_fft_lengths> <bsmooth_boxcar_length>0</bsmooth_boxcar_length> <bsmooth_chunk_size>0</bsmooth_chunk_size> A normal WU has: <analysis_fft_lengths>262136</analysis_fft_lengths> <bsmooth_boxcar_length>8192</bsmooth_boxcar_length> <bsmooth_chunk_size>32768</bsmooth_chunk_size> If an application doesn't error on the baseline smoothing of a 10ja10zz, it won't do any analysis since there are no FFT lengths defined. {edit:} I did fire off an email to the staff. Joe |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
The 'tape' ID of zz does seem to indicate this is something odd. Unless they did happen to have 676 'tapes' that day. Perhaps this is trying to get some of the old data that was unsplitable before. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Lint trap Send message Joined: 30 May 03 Posts: 871 Credit: 28,092,319 RAC: 0 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. I already had NNT set so they never had a chance! I can do this until they expire...:) Lt |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. For what all this efford? If they error out, they do it within few seconds... |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. Eithier way they'll end up being regarded as an Error, you could get them resent to the Stock app and complete them that way, Claggy |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. Sorry but the 1's that I had on my video cards were running between 5 & 10 min with no progress before I aborted them. Cheers. |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. Yeah, that would be my way of solving that problem too (if I needed to do so), but "modified vlar resend process"? |
Lint trap Send message Joined: 30 May 03 Posts: 871 Credit: 28,092,319 RAC: 0 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. They were cpu wu's that all ended with Integer Divide by Zero faults. The result texts show no cpu seconds and no runtimes. My Lunatics cpu app is AK_v8b2_win_SSE41.exe. The files were resent, but they did not reappear in client_state.xml...so I deleted the files a second time. Another manual update since then was made with no more activity from the servers. They will timeout on 9/8. Because they don't seem to be 'valid' workunits, I would just rather they all timeout than get computation errors. Lt |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
The 6 I received all ended in computation error (because I'm on 'anonymous platform'??), so I used a modified vlar resend process to delete them. Simply aborting them would have been to difficult? Then they would have been sent to another host instead of waiting 6-8 weeks to timeout and then be sent to another host. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Lint trap Send message Joined: 30 May 03 Posts: 871 Credit: 28,092,319 RAC: 0 |
Simply aborting them would have been to difficult? Then they would have been sent to another host instead of waiting 6-8 weeks to timeout and then be sent to another host. Yes, I would have aborted them if I had seen them before they errored. Anyway, they were put on a very short deadline after being resent and timed out a few hours later. Lt |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
Simply aborting them would have been to difficult? Then they would have been sent to another host instead of waiting 6-8 weeks to timeout and then be sent to another host. And what's wrong with reporting them as errors? I mean that would have happen automatically, no action from the user required... |
Matt Lebofsky Send message Joined: 1 Mar 99 Posts: 1444 Credit: 957,058 RAC: 0 |
In a word: oops. This is a test tape I generated with completely artificial data for testing/calibration purposes. It seems the splitter didn't like it, and obviously we sent some garbage through the whole system. Sorry about that. We will tweak a couple things and send this file out again. Please processes them normally, as you'll get normal credit, etc. - Matt -- BOINC/SETI@home network/web/science/development person -- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude |
Bernd Noessler Send message Joined: 15 Nov 09 Posts: 99 Credit: 52,635,434 RAC: 0 |
Thanks for the info. |
shizaru Send message Joined: 14 Jun 04 Posts: 1130 Credit: 1,967,904 RAC: 0 |
This is a test tape I generated... So that's why they're called JAZZ?:) |
©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.