| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: exited with zero status but no 'finished' file |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 14
|
|
| Author |
|
|
I need a bath
Senior Cruncher USA Joined: Apr 12, 2007 Post Count: 347 Status: Offline Project Badges: 
|
I keep getting this with CEP2 on only one machine. I reset the project to no avail.
----------------------------------------ideas?  |
||
|
|
X-Files 27
Senior Cruncher Canada Joined: May 21, 2007 Post Count: 391 Status: Offline Project Badges:
|
Got tons of these before.
----------------------------------------I don't know your setup but here's mine: with tons of "exited with zero..." -Dotsch/UX 1.2 x64 on USB stick -running boinc on hard drive with ntfs partition to no more errors -Dotsch/UX 1.2 x64 on USB stick -running boinc on hard drive with ext3 partition -set owner/permissions on the ext3 partition the error seems to have caused by ntfs partition as I tried to move around the boinc data it asked me if i want to overwrite which leads me to believe that the science app is generating same filename. windows = not case sensitive; linux = case sensitive.   [Edit 2 times, last edit by X-Files 27 at Jul 2, 2010 11:49:13 PM] |
||
|
|
I need a bath
Senior Cruncher USA Joined: Apr 12, 2007 Post Count: 347 Status: Offline Project Badges:
|
Yes, my OS is(was- I switched back to windows xp for the weekend) running on a USB too, not Dotsch but just ubunto 10.whatever. I put boinc on one of the hard drives. I don't know what ext3 partition means but I will google it. Thanks for the ideas.
----------------------------------------I also have this thing where Boinc manager takes forever to contact the client. Seems like intra-computer connectivity problems. |
||
|
|
X-Files 27
Senior Cruncher Canada Joined: May 21, 2007 Post Count: 391 Status: Offline Project Badges:
|
Yes, my OS is(was- I switched back to windows xp for the weekend) running on a USB too, not Dotsch but just ubunto 10.whatever. I put boinc on one of the hard drives. I don't know what ext3 partition means but I will google it. Thanks for the ideas. I also have this thing where Boinc manager takes forever to contact the client. Seems like intra-computer connectivity problems. ext3 is just the native linux partition, in windows its about fat32/ntfs. mine too takes forever to contact the client. my workaround is to suspend the project before quitting and resume once im ready to leave it alone. cep2 seems to be disk extensive. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Per the system requirement specs up to 2GB HD is required. During beta it was found that 1.5GB was not enough so it was upped.
----------------------------------------Zero Status has various known and illusive causes. The process is basically interrupted, then resumes from last checkpoint or from the beginning, sometimes over and over till BOINC determines there were too many and then aborts the task with "too many exits". See FAQ (in case you've seen it already): http://boincfaq.mundayweb.com/index.php?view=116
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Jul 3, 2010 6:27:41 AM] |
||
|
|
I need a bath
Senior Cruncher USA Joined: Apr 12, 2007 Post Count: 347 Status: Offline Project Badges:
|
I'm narrowing down the problem. I am using a quad and the zero status thing happens if I try to run more than 2 cores. Now I am trying to run 4 cores where 2 of the tasks are HCC and two are CEP2. This seems okay - so far.
---------------------------------------- |
||
|
|
I need a bath
Senior Cruncher USA Joined: Apr 12, 2007 Post Count: 347 Status: Offline Project Badges:
|
okay. It is clear that my quad running ubunto on CD cannot run more than 2 CEP2 at a time. Now running 2 cep2 and 4 HCC with no issue. CPU at 100% cores at 75%. Tons of memory and disk space available. Hmmmm
---------------------------------------- |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Wish we had an option to control the number of the same science app that can run concurrently... all sorts of bottleneck reductions could be automated. The future may bring such smartness in the client. knreed somewhere linked to a document that is with the developers that would let the optimizers drool... who knows, version 9.58 :D
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges:
|
Am getting these errors now, running mostly 4 x DDDT2 on a Q9650 (quad) that ran FAAH/HCMD for months without a hiccup.
----------------------------------------It also happened while running 4 x c4cw Beta WUs a few days ago, also 3 x DDDT2 plus 1 x HFCC. All 4 tasks exit and then restart about every 1 hour. XP-64, BOINC 6.2.19 32-bit. Have read Sek's FAQ: "Zero Status" & "If this h...atedly...." Messages The machine is overclocked & I thought it might be a hardware issue, so I posted a thread at Xtreme Systems: WCG tasks die "Task xxx exited with zero...;finished' file" Please go there for further details. [Edit 1 times, last edit by Rickjb at Aug 17, 2010 4:33:10 PM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Anything that interrupts communication between the core client and the science for 30 seconds can cause this. I can replicate this and reproduce this fail within 31 seconds it I start a certain Linux file indexer or load a mediaplayer that goes bonkers on not being able to connect to an unmounted resource even if networking is heavily taxed it can happen. The developers at some point in time were talking about redesigning this "Hello I'm Still Alive" stuff, so that the science would not restart from last checkpoint but longer not heard about this or at least make it more robust. Anything that virtually monopolizes the CPU time for > 30 seconds can cause this. So if doing heavy stuff which is likely to take longer than a jiffy, pause BOINC or you run risk it unloading the science app regardless of LAIM being on or not.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
|