| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 38
|
|
| Author |
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Yep, going thru the long log you posted with the AC@H failing on the same machine saw this bit.
----------------------------------------2007-09-01 23:30:52 [World Community Grid] Aborting task dddt0101a0030_ZINC00637125-0001_01_4: exceeded disk limit: 72.52MB > 71.53MB 2007-09-01 23:30:52 [World Community Grid] Deferring communication for 1 min 0 sec 2007-09-01 23:30:52 [World Community Grid] Reason: Unrecoverable error for result dddt0101a0030_ZINC00637125-0001_01_4 (Maximum disk usage exceeded) Coincidence, as various machines here are churning them out flawlessly in between FA@H and HPF2. What CPU?
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Coincidence, as various machines here are churning them out flawlessly in between FA@H and HPF2. What CPU? Sek, I don't see any rhyme or reason to the DDDT max disk usage PCs... some are Core2Duo, some are AMD64X2. Most are on WinXP but a few are on Vista. I came back from a weekend trip to find 6 pages of DDDT error WUs Spot-checking the quorums...it looks like they had 7 error statuses in the quorum (mine included). I don't see any Valid for the ones I've spot-checked. [Edit 2 times, last edit by Former Member at Sep 3, 2007 11:42:48 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Here the same:
Maximum disk usage exceeded 7 of 48 WUs between 0.00 till 0.20 hours of work. PC Specs: 13GB free Space on the BOINC-Harddisk 2 GB RAM Q6600 not overclocked failed WUs since 30.08.2007: (5 WUs with 7-Error-Status, the other 2 get this 7-E-S soon )dddt0101a0023_ZINC04092757-0000_04 dddt0101a0023_ZINC00920493-0000_06 dddt0101a0023_ZINC01222982-0000_04 dddt0101a0026_ZINC00528394-0001_00 dddt0101a0034_ZINC04136435-0002_03 dddt0101a0035_ZINC02476765-0000_05 dddt0101a0035_ZINC04137599-0001_04 There are absolut no failures in other BOINC-Projects(QMC,Rosetta,...) especially other WCG-Projects (Fight AIDS, HPFP2) Greetings tazrt |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I think I see a trend here.. max diskspace or out of memory issues can be attributed to a few things.. maybe this will help most of you out...
I been working with BOINC for over a day now since migrating from the UD agents on all my rigs.. Disk space I left as default.. 40 GB which seems to be plenty.. when BOINC is processing it will basicly leave 1 completed result in queue, running one, and will leave one undone in queue. when on multi processor rigs just add one of those pretty much for each core. for disk space for those with small drives I would not recommend going lower then 4 GB of drive space for BOINC drive space use... again add 1X space per core.. Memory issues Virtual Memory.. BOINC loves it.. if youre computer runs low on virtual memory BONIC will try to use any remaining Physical memory to process. typically normal use looking at about 500 MB or VM per core and can go up and down based on WU and WU progress.. typically you want to have 1 GB of Virtual Memory per core.. on your computer.. this will ensure that each task running will have some elebow room to work with.. Hope this helps most of you that have been having troubles... |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
Greetings,
We have figured out what was causing the maximum disk usage exceeded. Some of the DDD-T work units had more atom lines than a usual FAAH work. Thus we have upped the estimated disk space needed per work unit to get around this issue. Each work unit for DDD-T will have a quota of 100MB. Before it was set to 75MB. I am restoring those error work units today and they will be re-run and sent to the researchers with the other work units from the batches. Thank you for your patience during this and you participation in World Community Grid. -Uplinger |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
@choj01,
----------------------------------------Interesting but fear not as the processes are designed to go into a 'waiting for memory' state if there really is not enough available. As eluded the 'max disk usage exceeded' is a secondary error resulted from an other fault. When the error occurs, the program is to dump the memory state to disk for analysis, except the dump is so large it runs into the restriction set for this particular project which seemingly is around 72mb. Consider the upload result is only 200kb. At times thought is was related to 5.10.x, but not that either. Windows had some interesting patches. ATM I cant remember this having been reported for either Linux or Mac. Esoteric17 reported different systems and OSses, but all windows. knreed would be looking at it as he explained that if a unit failed, all 7 send out fail. It would be too much of coincidence that 7 systems had the same shortcoming. Added: See that while writing, Uplinger posted the resolution, so another problem send to the hereafter.
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Sep 5, 2007 7:03:25 PM] |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
As a side note, the quota's set on each work unit are there to prevent the work unit from writing too much information to the hard disk. This keeps the work unit using up more disk space then is needed.
-Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks uplinger. I've re-checked DDDT - hopefully I won't get any more errors
![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes thank you for solving this issue as I noticed this all error thing also.
|
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
There may be a few that I missed but a majority of them if not all should be fixed.
Thank you again! -Uplinger |
||
|
|
|