Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
World Community Grid Forums
Category: Beta Testing Forum: Beta Test Support Forum Thread: BETA for type C in DDD2 |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 95
|
Author |
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
It's a question of whether the compiler libs are changed to also allow Kremmen's computers to work on DDDT-2. Think his rigs were amongst the few with a very high fail rate, at least from reading what was reported on the forums.
----------------------------------------
WCG Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All! |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It's a question of whether the compiler libs are changed to also allow Kremmen's computers to work on DDDT-2. Think his rigs were amongst the few with a very high fail rate, at least from reading what was reported on the forums. Interesting. If it's a library issue, why would it just be the sX subclasses of type C WUs that fail, when other C's and A's are fine? And why are there no error messages logged? That would seem to be a fault, no matter what. The intriguing part to me is that the machine which got the errors (an underclocked, non-HT P4) happily completing a type A. I guess some particular subset of one particular sub-class of type C WUs perform some action that doesn't happen in any of the other DDD2 family of WUs? Apart from that, I would have thought the inability to restart WU's without them going back to the start again would have been more important. (Maybe not so much damage in the case of type C, as they're so quick, but for type A, surely that has to be fixed?) [Edit 1 times, last edit by Former Member at Feb 13, 2010 5:10:09 AM] |
||
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: |
I think that when describing any kind of strange WU behaviour here (or in any other WCG forum), we should include a brief description of our problem machine's hardware & OS. It might help in tracking down the source of the problem. If you are overclocking, state this too, especially for Beta tests. The new project software might just trigger hardware errors that don't happen on other projects.
----------------------------------------@Kremmen ... or did I miss your device's description? Re: Virtual memory Size for DDDT2 Betas (eg member tluehr) (I'm only running Windows, 2k, XP-32 and XP-64. Things could be different under different OSes). The VM You may bump into the size limits of your swapfile(s)/pagefile(s). If these file(s) is/are allowed to grow automatically, the limit may be imposed by the amount of space available your hard disc. (On XP, do a Help search for "virtual memory" to find how to set your VM size. It's also set in Control Panel >> System >> Advanced tab >> Performance >> Advanced) - HTH [Edit]: The VM Sizes above are those that show in Task Manager. I assume that they have to be backed up by available space in page/swap files, but documentation on Windows VM/swap behaviour is hidden in places unknown to me. [Edit 1 times, last edit by Rickjb at Feb 13, 2010 2:37:57 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
@Kremmen ... or did I miss your device's description? I edited in the hardware type (P4) before you posted, as it happens, but maybe you missed it? OS is Debian Etch linux with 2.6.18-6-686 kernel. Re: Virtual memory requirements for DDDT2 Betas (eg member tluehr) ... You may bump into the size limits of your swapfile(s)/pagefile(s). Does Windows require swapfile space for unused memory? If so (aside from showing Microsoft still can't even write an OS properly) it would be a huge advantage for Windows users if the WCG programs didn't declare all that unused space. As far as Linux is concerned, the whole concept of a VM "requirement" for these programs is a misnomer. Linux will only use memory as it is required. The fact that these programs declare about 348MB of uninitialised space (bss size on the recent beta: 365,866,724 bytes) is completely irrelevant to anything. Actual (resident) memory usage is the only relevant data, and it is consistently about 340MB less than VM. (This suggests that maybe 8MB of the 348MB is used.) A type A WU could run on a linux box with 256MB RAM and no swap. |
||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Windows (Vista and W7 at least) requires NONE, as I ran for weeks in January with the VM set to hard zero... except when running 4 DDDT-2 Beta concurrent it became very tight so allowed it again to overflow to disk.
----------------------------------------
WCG Global & Research > Make Proposal Help: Start Here!
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: |
1. Edited my post above.
2. @Sek: Please explain what the 1.3GB VM Size per Type A beta that I saw in Jan meant, and how 4 of these tasks ran on your machine(s) without page/swap file space. |
||
|
GIBA
Ace Cruncher Joined: Apr 25, 2005 Post Count: 5374 Status: Offline |
Uplinger or someone from tech team which could looking at Beta results.
----------------------------------------I just saw that this mine 3 Beta results below, remain as Too Late status in the results page. Is it really right ? I guess that maybe could be changed for Valid or Invalid status, just to count as result in the total (once probably due the errors found at that batch at that time, the remain batch was aborted and for few crunchers that returned results before it, the credits was launched manually for tech team). BETA_ erlc_ a079_ se0000_ 3-- GIBAQX9770F Too Late 27/01/10 16:52:19 27/01/10 20:05:04 2.51 52.9 / 52.9 BETA_ erlc_ a080_ se0000_ 3-- GIBAQX9770F Too Late 27/01/10 16:52:19 27/01/10 20:05:04 2.47 52.3 / 52.3 BETA_ erlc_ a081_ se0000_ 3-- GIBAQX9770F Too Late 27/01/10 16:52:19 27/01/10 20:05:04 2.49 52.6 / 52.6
Cheers ! GIB@
Join BRASIL - BRAZIL@GRID team and be very happy ! http://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=DF99KT5DN1 |
||
|
GIBA
Ace Cruncher Joined: Apr 25, 2005 Post Count: 5374 Status: Offline |
Uplinger or someone from tech team which could looking at Beta results. I just saw that this mine 3 Beta results below, remain as Too Late status in the results page. Is it really right ? I guess that maybe could be changed for Valid or Invalid status, just to count as result in the total (once probably due the errors found at that batch at that time, the remain batch was aborted and for few crunchers that returned results before it, the credits was launched manually for tech team). BETA_ erlc_ a079_ se0000_ 3-- GIBAQX9770F Too Late 27/01/10 16:52:19 27/01/10 20:05:04 2.51 52.9 / 52.9 BETA_ erlc_ a080_ se0000_ 3-- GIBAQX9770F Too Late 27/01/10 16:52:19 27/01/10 20:05:04 2.47 52.3 / 52.3 BETA_ erlc_ a081_ se0000_ 3-- GIBAQX9770F Too Late 27/01/10 16:52:19 27/01/10 20:05:04 2.49 52.6 / 52.6 Hi, past more than two months since results was returned, this issue continues and WU's status don't changed. Is the first time that I saw WU's appearing in my results for so long time. Could someone please take a look at and return me with a position about it ? Thank you in advance.
Cheers ! GIB@
Join BRASIL - BRAZIL@GRID team and be very happy ! http://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=DF99KT5DN1 |
||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
GIBA,
----------------------------------------for as far as I can see these 3 results, in result count, time contributed and in credit have already been added to your My Grid page totals. They merely have not been moved off the Result Status page... the less essential part. :<)
WCG Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All! |
||
|
GIBA
Ace Cruncher Joined: Apr 25, 2005 Post Count: 5374 Status: Offline |
Sekerob,
----------------------------------------I think that they haven't been moved off yet due don't have a definition in the status, once normaly the status associated with results counted, time contributed and credit granted is just valid or invalid, as I suggested and appointed before. So, it isn't normal in my opinion. Once at that time, some crunchers like me returned WU's before the project abort action, I guess that maybe there are more people in the same situation, and guess too that maybe this subject need attention to be finished / solved. Anyway, thank you for your insight and have a happy Easter time.
Cheers ! GIB@
Join BRASIL - BRAZIL@GRID team and be very happy ! http://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=DF99KT5DN1 |
||
|
|