Message boards : Number crunching : Loads and loads of computing errors today
Author | Message |
---|---|
[BAT] tutta55 Send message Joined: 16 Sep 05 Posts: 59 Credit: 99,832 RAC: 0 |
On 2 of my machines I'm producing dozens of computing errors today. On the others I don't :-( I doubt it has anything to do with the machines themselves, because for most of these work units others are producing errors too. An example https://boinc.bakerlab.org/rosetta/workunit.php?wuid=342005 BOINC.BE: For Belgians who love the smell of glowing red cpu's in the morning Tutta55's Lair |
[AF>france>pas-de-calais]symaski62 Send message Joined: 19 Sep 05 Posts: 47 Credit: 33,871 RAC: 0 |
On 2 of my machines I'm producing dozens of computing errors today. On the others I don't :-( <core_client_version>4.19</core_client_version> <message>Incorrect function. (0x1) - exit code 1 (0x1) </message> <active_task_state>1</active_task_state> <signal>0</signal> <stderr_txt> |
David E K Volunteer moderator Project administrator Project developer Project scientist Send message Joined: 1 Jul 05 Posts: 1018 Credit: 4,334,829 RAC: 0 |
|
wdsmia Send message Joined: 25 Sep 05 Posts: 3 Credit: 2,109,013 RAC: 0 |
I just started having computing errors on one box looks like the 4.78 app is not playing well with others or aleast with the 4.19 client. <core_client_version>4.19</core_client_version> <message>Incorrect function. (0x1) - exit code 1 (0x1) </message> <active_task_state>1</active_task_state> <signal>0</signal> <stderr_txt> application version 4.78 |
Fuzzy Hollynoodles Send message Joined: 7 Oct 05 Posts: 234 Credit: 15,020 RAC: 0 |
@wdsmia and others. My 4.72 is working really fine and is really stable. You can download it from this, where you can find all versions of BOINC. The advantage of 4.72 in preference to 4.45 is that the time estimate is better, so it adjust itself after some WU's. This means that it downloads relatively correct instead of suddenly going in panic mode about some WU's which are estimated wrongly to take more time. [b]"I'm trying to maintain a shred of dignity in this world." - Me[/b] |
wdsmia Send message Joined: 25 Sep 05 Posts: 3 Credit: 2,109,013 RAC: 0 |
@Hollynoodles looking at several results from different people its looking more like there may be a problem with the new 4.78 app Im finding the same errors on boxes running 4.19, 4.45 and 4.72. The Windows/x86 4.78 app was released 21 Oct 2005 1:36:59 UTC If you are running with more than one days cache you may not have run one using the new app. |
mscharmack Send message Joined: 29 Sep 05 Posts: 2 Credit: 11,323 RAC: 0 |
I just had 42 computational errors in a row, then two test WU's were run and I am currently into my next WU. It seems to be running okay?????? |
wdsmia Send message Joined: 25 Sep 05 Posts: 3 Credit: 2,109,013 RAC: 0 |
ok after investagating a bit more. Its looking like most of these errors are on wu's with the name of xxxxx_abrelaxmode_random_gauss_xxxxx. why some boxes can process them with no errors and others error out? I havent a clue. |
AnRM Send message Joined: 18 Sep 05 Posts: 123 Credit: 1,355,486 RAC: 0 |
I had three boxes sharing R@H and LHC@H and had to run BOINC ver.4.19 because LHC@H have not upgraded to 5.x yet.....R@H started giving computing errors when R@H ver.4.78 came along and, unfortunately, rattled off a whole herd of failed WU's on all three before I caught it. Soooo.... I had to upgrade to BOINC ver5.2.2 to keep R@H running and wave goodbye to LHC@H until they get their act together. Bottom line: ver. 4.78 won't run on BOINC ver.4.19. FYI, the other boxes are using BOINC ver.5.2.1 and are happy. |
KWSN_Dagger Send message Joined: 19 Oct 05 Posts: 5 Credit: 2,803 RAC: 0 |
I noticed that with these errors, it depends on what machine your running. ie if you run an AMD and it errored out, a person with a pentium would have success and vice versa. |
[BAT] tutta55 Send message Joined: 16 Sep 05 Posts: 59 Credit: 99,832 RAC: 0 |
Upgrading to 4.45 does not help. A message like this occurs for those same work units: <core_client_version>4.45</core_client_version> <message>CreateProcess() failed - The process cannot access the file because it is being used by another process. (0x20) </message> BOINC.BE: For Belgians who love the smell of glowing red cpu's in the morning Tutta55's Lair |
Ulrich Metzner Send message Joined: 17 Sep 05 Posts: 22 Credit: 405,640 RAC: 0 |
greetz, Uli |
Webmaster Yoda Send message Joined: 17 Sep 05 Posts: 161 Credit: 162,253 RAC: 0 |
Same here, both my AMD Athlon powered PCs have had a number of work units crash with oxC0000005 errors. I upgraded both to BOINC 5.2.2 to see if it made any difference, which has reduced the number of errors but not eliminated them. My Pentium 4 (2.4, 2.8 and 3.4 GHz) machines have been fine. EDIT: The Athlon 64 has been stable for the last 10 or so WU, but the Athlon XP is having more errors than successes and I do keep the WU in memory. *** Join BOINC@Australia today *** |
AnRM Send message Joined: 18 Sep 05 Posts: 123 Credit: 1,355,486 RAC: 0 |
....Could it be related to some optimizations in the new executable 4.78? This error occurs on an Athlon TB 1400 without SSE support.[/quote] >Most of my boxes are AMD ie. Durons, Semprons, Athlons and since upgrading to BOINC ver.5.x I have not had any problems with the new R@H ver.4.78. IMHO, the main heavy in the piece is the version of BOINC used.....Cheers, Rog. |
Robert Nelson Send message Joined: 17 Sep 05 Posts: 4 Credit: 1,233,589 RAC: 703 |
....Could it be related to some optimizations in the new executable 4.78? This error occurs on an Athlon TB 1400 without SSE support. >Most of my boxes are AMD ie. Durons, Semprons, Athlons and since upgrading to BOINC ver.5.x I have not had any problems with the new R@H ver.4.78. IMHO, the main heavy in the piece is the version of BOINC used.....Cheers, Rog.[/quote] Have not had any errors of the type mentioned, using BOINC 5.2.2 the units lately have been 4.78, windows boxs AMD and Intel processors. I do keep processs in memory. Note has been problem mentioned in many threads concerning not keeping rosetta in memory. |
Ulrich Metzner Send message Joined: 17 Sep 05 Posts: 22 Credit: 405,640 RAC: 0 |
Have not had any errors of the type mentioned, using BOINC 5.2.2 the units lately have been 4.78, windows boxs AMD and Intel processors. I do keep processs in memory. Note has been problem mentioned in many threads concerning not keeping rosetta in memory. I presume, you are talking about Athlon XP's? Because the Athon XP has SSE instructions. The older Athlon Thunderbirds don't have SSE instructions. That's my point... greetz, Uli |
AnRM Send message Joined: 18 Sep 05 Posts: 123 Credit: 1,355,486 RAC: 0 |
.....I presume, you are talking about Athlon XP's? Because the Athon XP has SSE instructions. The older Athlon Thunderbirds don't have SSE instructions. That's my point...[/quote] >Point well taken!.....my Athlon is a XP3000+ with 1Gb RAM and running WinXP. I have not had any problems since upgrading to BOINC 5.2.1. I also leave Rosetta in memory and run R@H and CPDN 90/10 with 120 min changeover. Your idea certainly seems intresting. I am not having any problems with my 1.8Gz Durons or 2.8GHz Semprons, howerver. I don't know if they have SSE instructions capability but I doubt it?? Hope this helps.....Cheers,Rog. |
kb7rzf Send message Joined: 7 Oct 05 Posts: 16 Credit: 35,427 RAC: 0 |
I've done 2 of these WU's that have been talked about, and so far no problem, running 5.2.2, both WU's finished between 45 minutes and an hours worth of time. And did not error out. My computer is an Intel Celeron 2.6ghz with 512mb ram on WinXP home. Jeremy |
Ulrich Metzner Send message Joined: 17 Sep 05 Posts: 22 Credit: 405,640 RAC: 0 |
greetz, Uli |
Beezlebub Send message Joined: 18 Oct 05 Posts: 40 Credit: 260,375 RAC: 0 |
I have 6 errors out of 55 WU returned 4 from a P4 3.4ghz w/HT and 2 from a AMD XP 2800+ 2.1 ghz. running Boinc 5.2.2 and none from my P4 2.0 ghz Two were the old style WU(4.77) and the other 4 are the new WU(4.78). This is from 18 oct. thru 22 oct. e6600 quad @ 2.5ghz 2418 floating point 5227 integer e6750 dual @ 3.71ghz 3598 floating point 7918 integer |
Message boards :
Number crunching :
Loads and loads of computing errors today
©2024 University of Washington
https://www.bakerlab.org