| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: DDD2 Type B work units going out. |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 369
|
|
| Author |
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
GIBA,
This application uses quite a bit of page file (virtual memory). If your machine does not have enough when boinc is starting the application, then it exits. The allocation of this space is pretty quick on a work unit, within 15 minutes of it starting the virtual memory usage should be consistent after that. SHINN, Do you by chance remember what the percentage complete was on the work unit before you restarted? -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Speaking of percentage - my first WUs to be completed (hopefully) are now at 106% and 105% (WUs BETA_erlc_a080_se0000_0 and BETA_erlc_a060_se0000_1) - let's see how far they go...
|
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
please go into your slots directory and try to capture result.out file. just copy it to your desktop. Should be in your boinc data directory.
-Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Did you encounter this issue with the pe work units at all? -Uplinger The pe works look normal. I am uncertain the pe work units when they close to complete.  I don't know what happens to got the errors once other processes are running in this machines. The error result log contain a code and message that is something related with page files be a little small to do the operation: I set one of my i7 pagefiles vaule to 2.5G. The i7 got some error,too. Every type B work unit needs to be allocate 500mb virtual memory. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Got both files length is 31.780k, most lines are like
EXELEC> Atom displacement from last update exceeds warning distance. Maximum distance = 2.05 Atom = 1234 Average distance = 0.62 EXELEC> Atom displacement from last update exceeds warning distance. Maximum distance = 2.06 Atom = 1234 Average distance = 0.62 EXELEC> Atom displacement from last update exceeds warning distance. Maximum distance = 2.06 Atom = 1234 Average distance = 0.63 Do you need the whole files? Percentage is now 110% and 111%. Hope they will finish someday - need the time... ;-) stderr.txt holds only the usual line, no further error messages or warnings. Matthias |
||
|
|
roundup
Veteran Cruncher Switzerland Joined: Jul 25, 2006 Post Count: 831 Status: Offline Project Badges: 
|
My first one is completed:
----------------------------------------BETA_ erlc_ a092_ se0000_ 1-- No issues from my side. Progress during runtime was always normal, no percentage above 100% observed as reported by mweisensee. My wingies did not report yet. Edit: This unit ran with HCC units concurrently. [Edit 1 times, last edit by roundup at Jan 27, 2010 9:21:19 PM] |
||
|
|
GIBA
Ace Cruncher Joined: Apr 25, 2005 Post Count: 5374 Status: Offline |
GIBA, This application uses quite a bit of page file (virtual memory). If your machine does not have enough when boinc is starting the application, then it exits. The allocation of this space is pretty quick on a work unit, within 15 minutes of it starting the virtual memory usage should be consistent after that. SHINN, Do you by chance remember what the percentage complete was on the work unit before you restarted? -Uplinger Humm, your explanation appear be consitent with what happens. The machine have 4 MB of RAM running over W Vista 32 bit, so W Vista just see 3,25 GB of physical RAM, and allocates more 6,68 Gb of virtual (according BOINC client), probably some more dinamically if need (W Vista manage it when need). The computer log shown me that the machine suspended BOINC client for few minutes exactly at same time that the error appear, probably it happens due some scheduled task from other applications which run there. This applications are memory eaters for instants, so probably happens a lack of physical and virtual memory to be used by boinc client during error event. It is my bet for now. Thanks.
Cheers ! GIB@
 Join BRASIL - BRAZIL@GRID team and be very happy ! http://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=DF99KT5DN1 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Do you by chance remember what the percentage complete was on the work unit before you restarted? I got the two work units at 2010/01/28 00:24 (UTC+8),and i rebooted the machine at 2010/01/28 01:00 (UTC+8). The work unit took 2.81 hours CPU time(i7 920 oc 4G). I guess the percentage is about 15-20%. |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
Matthias,
Did you restart at all on that se work unit? There is a bug on all of the se work units and we are working on fixing it. It appears to mainly be an issue when the work units are restarted, if you continue to restart over and over again, then you're work unit will not finish until the deadline is hit and boinc cancels it. Chances are you'll hit another limit before the time limit, but the work units are not good :( I should be able to fix these work units and send them out again. Also, Matthias, if you could zip up the result.out file and send it to me at support@worldcommunitygrid.org that would be great. It should answer my questions on the bug. Thanks, -Uplinger |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
SHINN,
Thank you for the explanation. That is the behaviour that we are expecting with this bug. -Uplinger |
||
|
|
|