Oddity. BOINC trying to upload an unfinished wu

留言板 : Number crunching : Oddity. BOINC trying to upload an unfinished wu
留言板合理

To post messages, you must log in.

作者消息
peristalsis

发送消息
已加入:23 Jul 99
贴子:154
积分:28,610,163
近期平均积分:51
United States
消息 522120 - 发表于:23 Feb 2007, 13:36:29 UTC

and thank you!! (hit the post button before brain was fully engaged)..p
ID: 522120 · 举报违规帖子
peristalsis

发送消息
已加入:23 Jul 99
贴子:154
积分:28,610,163
近期平均积分:51
United States
消息 522118 - 发表于:23 Feb 2007, 13:35:44 UTC

you're welcome
ID: 522118 · 举报违规帖子
Richard Haselgrove Project Donor
志愿者测试人员

发送消息
已加入:4 Jul 99
贴子:14152
积分:200,643,578
近期平均积分:874
United Kingdom
消息 522077 - 发表于:23 Feb 2007, 10:08:31 UTC

I was discussing this on the BOINC development board last night. They tell me that that problem should be fixed in v5.8.11 - I won't run that version (it causes as many problems as it solves), but I'm trying out v5.8.15: unfortunately (fortunately?), the problems are very rare, so it'll be hard to know for certain.

Your messages match the pattern exactly, including the other result finishing one second earlier - thanks for taking the trouble to find them.
ID: 522077 · 举报违规帖子
peristalsis

发送消息
已加入:23 Jul 99
贴子:154
积分:28,610,163
近期平均积分:51
United States
消息 521711 - 发表于:22 Feb 2007, 20:46:17 UTC

For your viewing pleasure:

2007-02-21 11:09:47 [SETI@home] Starting 15au03aa.20522.4016.947138.3.134_0 ***
2007-02-21 11:09:47 [SETI@home] Starting task 15au03aa.20522.4016.947138.3.134_0 using setiathome_enhanced version 517 ***
2007-02-21 11:42:57 [SETI@home] Computation for task 15au03aa.20522.4016.947138.3.156_2 finished
2007-02-21 11:42:57 [SETI@home] Starting 15au03aa.20522.4016.947138.3.138_2
2007-02-21 11:42:57 [SETI@home] Starting task 15au03aa.20522.4016.947138.3.138_2 using setiathome_enhanced version 517
2007-02-21 11:42:58 [SETI@home] Computation for task 15au03aa.20522.4016.947138.3.134_0 finished ***
2007-02-21 11:43:58 [SETI@home] Starting 15au03aa.20522.4016.947138.3.134_0
2007-02-21 11:43:58 [SETI@home] Starting task 15au03aa.20522.4016.947138.3.134_0 using setiathome_enhanced version 517 ***
2007-02-21 12:16:21 [SETI@home] Computation for task 15au03aa.20522.4016.947138.3.138_2 finished


I notice at 11:42:58 reported as finished
Then at 11:43:58 its started again. Not "restarted".
I couldn't find another 'finished' report on the WU. But then my eyes are starting to cross and I might have missed it.
Hope the data helped...p
ID: 521711 · 举报违规帖子
Richard Haselgrove Project Donor
志愿者测试人员

发送消息
已加入:4 Jul 99
贴子:14152
积分:200,643,578
近期平均积分:874
United Kingdom
消息 521603 - 发表于:22 Feb 2007, 14:53:45 UTC
最近的修改日期:22 Feb 2007, 14:58:16 UTC

If you're interested, you can find an archive of all the old messages in the file 'stdoutdae.txt' in your BOINC directory.

Here are the results for the one I had last night: BOINC core client message boards.

Edit: the reference to 'State file error' means that the information which should be stored in 'client_state.xml' for that WU has gotten scrambled. By the time you see that message, BOINC will have corrected it, so it's nothing to worry about - you already knew you had a problem.

I haven't found any way of recovering a WU once it gets into this loop - probably best just to abort the transfer and move on.
ID: 521603 · 举报违规帖子
peristalsis

发送消息
已加入:23 Jul 99
贴子:154
积分:28,610,163
近期平均积分:51
United States
消息 521595 - 发表于:22 Feb 2007, 14:11:55 UTC

Using boinc 5.8. plus Simon's optimizer Revision: R-2.2|xW|FFT:IPP_SSE2|Ben-Joe.
My log doesn't go far enough back to when the WU 'went bad'. Initial post had it stuck but trying to upload without completion.
It is now shown as 100% completion and trying to (still) upload. Time is 32:19...
Current messages on the WU:
2/22/2007 8:36:02 AM|SETI@home|[error] State file error: result 15au03aa.20522.4016.947138.3.134_0 is in wrong state
AND
2/22/2007 8:36:02 AM|SETI@home|[error] State file error: result (­ not found
Right after the open parenthesis there is a 'square'.
STill won't upload, giving the above error messages.
AMD X2 4400, WinXPPRO SP2
I still dunno. I presume the 'state' its talking about is the 'state of confusion'. Only one I've seen so just slightly curious...p
ID: 521595 · 举报违规帖子
Profile zoom3+1=4
志愿者测试人员
Avatar

发送消息
已加入:30 Nov 03
贴子:63296
积分:55,293,173
近期平均积分:49
United States
消息 521350 - 发表于:22 Feb 2007, 2:26:48 UTC

Question is what version of Boinc is the OP talking about?
5.8.11 would do a string of small units like 10 or more sometimes in a minutes, I'm now running 5.4.11 again so I don't know If I can supply anything to back that up.
My Amazon Wishlist
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 521350 · 举报违规帖子
Richard Haselgrove Project Donor
志愿者测试人员

发送消息
已加入:4 Jul 99
贴子:14152
积分:200,643,578
近期平均积分:874
United Kingdom
消息 521243 - 发表于:21 Feb 2007, 22:44:53 UTC

I have a theory about these WUs, but we need further evidence.

Could you still go through the logs, the way I did in this thread, and see if it matches?

What I'm looking for is:
Did it download, start and finish crunching normally?
Did it finish crunching exactly 1 second after another WU from the same project?
Did it re-start crunching again?
Did it attempt to upload, without finishing the second crunch?

I've got another one stuck at the moment, but it's too late (UK time) to do the research now - I'll track it through in the morning.
ID: 521243 · 举报违规帖子
peristalsis

发送消息
已加入:23 Jul 99
贴子:154
积分:28,610,163
近期平均积分:51
United States
消息 521188 - 发表于:21 Feb 2007, 21:29:56 UTC

I have a wu showing 97.487% completion that boinc is trying to upload. Seems stuck..no big deal but shouldn't the software not allow this?
"2/21/2007 4:25:13 PM|SETI@home|[error] Error on file upload: socket read incomplete: asked for 14500, got 9630: No such file or directory"
Been that way for the last several hours..p

ID: 521188 · 举报违规帖子

留言板 : Number crunching : Oddity. BOINC trying to upload an unfinished wu


 
©2020 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.