| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project Forum
Thread: Errors exit code 29 (0x1d) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 40
|
|
| Author |
|
|
toss
Senior Cruncher New Zealand Joined: Jan 3, 2007 Post Count: 220 Status: Offline Project Badges: 
|
Yesterday I had a machine finish DDDT and changed it to CEP.
While unattended for 12 hrs it errored about 50 WU's in quick succession. No units actually completed. Upon discovery I solved the problem by switching it to another project. It continues to run other projects without error. I have other machines on CEP without trouble. I checked some of the results and all appear to be exit code 29 (0x1d). Previously when I had some WU's with this error, I read in these forums about it and crunched other work until the problems were resolved. Mainly I post this just for information if anyone is interested. However I am curious as to what might be unique about this machine that causes it to error where others don't. Or is it an anomaly of this project? --------------------------- Result Name: E000848_ 388C_ 009e09815_ 0-- <core_client_version>6.6.20</core_client_version> <![CDATA[ <message> The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d) </message> <stderr_txt> Calling initGraphics() INFO: No state to restore. Start from the beginning. Encountered error. Exiting. </stderr_txt> ]]> --------------------------- 19/07/2009 8:54:35 p.m. Starting BOINC client version 6.6.20 for windows_intelx86 19/07/2009 8:54:35 p.m. log flags: task, file_xfer, sched_ops 19/07/2009 8:54:35 p.m. Libraries: libcurl/7.19.4 OpenSSL/0.9.8j zlib/1.2.3 19/07/2009 8:54:35 p.m. Data directory: C:\Documents and Settings\All Users\Application Data\BOINC 19/07/2009 8:54:35 p.m. Running under account xxxx 19/07/2009 8:54:35 p.m. Processor: 2 AuthenticAMD AMD Athlon(tm) 64 X2 Dual Core Processor 4600+ [x86 Family 15 Model 43 Stepping 1] 19/07/2009 8:54:35 p.m. Processor features: fpu tsc pae nx sse sse2 3dnow mmx 19/07/2009 8:54:35 p.m. OS: Microsoft Windows XP: Professional x86 Editon, Service Pack 3, (05.01.2600.00) 19/07/2009 8:54:35 p.m. Memory: 1023.48 MB physical, 2.21 GB virtual 19/07/2009 8:54:35 p.m. Disk: 34.18 GB total, 14.91 GB free 19/07/2009 8:54:35 p.m. Local time is UTC +12 hours 19/07/2009 8:54:35 p.m. Configured to not use coprocessors 19/07/2009 8:54:35 p.m. Not using a proxy 19/07/2009 8:54:35 p.m. World Community Grid URL: http://www.worldcommunitygrid.org/; Computer ID: 527660; location: (none); project prefs: default 19/07/2009 8:54:35 p.m. World Community Grid General prefs: from World Community Grid (last modified 19-Jul-2009 04:56:08) 19/07/2009 8:54:35 p.m. World Community Grid Host location: none 19/07/2009 8:54:35 p.m. World Community Grid General prefs: using your defaults 19/07/2009 8:54:35 p.m. Preferences limit memory usage when active to 972.31MB 19/07/2009 8:54:35 p.m. Preferences limit memory usage when idle to 1023.48MB 19/07/2009 8:54:35 p.m. Preferences limit disk usage to 15.03GB |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
What can we say... a regression error which we saw with previous CEP science versions and thought to have been resolved with version 6.32. This is a C type job Batch E000848. So far the once I have had through 842 did fine and working on an 850 just fetched, C type on a 6.2.28 client on XP 32 bit platform, only Intel in my house.
----------------------------------------What's the quorum partner results looking like (result status page, click on WU name)? Just wanting to ensure it's not a general event. thanks
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 2 times, last edit by Sekerob at Jul 20, 2009 11:11:46 AM] |
||
|
|
toss
Senior Cruncher New Zealand Joined: Jan 3, 2007 Post Count: 220 Status: Offline Project Badges:
|
Not many results in yet. Most in progress.
Sampled, and any seen were valid or PV. Seems unique to this machine - but I have no idea what or why. Appears to me to be a dud combination of project and particular machine. Machine runs other work OK, and CEP runs on other machines OK. But these 2 just won't get on together! Cheers. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
If a machine starts out of the blue flunkies, boot. Also verify your AV is not acting up and suddenly seeing ghosts. Exclude AV from scanning your BOINC data_dir C:\Documents and Settings\All Users\Application Data\BOINC would solve that. The memory part still get's scanned.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
toss
Senior Cruncher New Zealand Joined: Jan 3, 2007 Post Count: 220 Status: Offline Project Badges:
|
Thanks Sekerob.
I've added the AV (avast) exclusion. I'll reboot soon and try and grab 1 more WU to test that. Cheers |
||
|
|
toss
Senior Cruncher New Zealand Joined: Jan 3, 2007 Post Count: 220 Status: Offline Project Badges:
|
Test WU failed.
exit code 29 (0x1d) again. |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
toss,
I have looked at your machine and the various CEP work units it was running. It appears as though your machine is having issues writting one of the output files properly. This could be due to anything from your virus scanner locking the file up (also other malware scanners due this as well, so if you have a pop up blocker please check them). Also, it could be due to something simple as the slots directory is having issues with permissions. Could you please provide some additional information. Like how the agent was installed on the machine (service, normal...etc). Also, what are your settings for drive space usage and does the machine have plenty of spare room? Is it writting to a normal hard drive or is it using an external storage device? -Uplinger |
||
|
|
toss
Senior Cruncher New Zealand Joined: Jan 3, 2007 Post Count: 220 Status: Offline Project Badges:
|
Greetings uplinger,
Thank you very much for your interest and response. Firstly I should explain I'm a self taught hobbyist with little technical expertise. But I'm learning every day. 1/ Response to your post. IMHO AV shouldn't be a problem since I excluded BOINC per Sek's advice above. I do have pop ups blocked via F'fox options. Have no other malware scanners always running - I do others manually. You said.... how the agent was installed on the machine (service, normal...etc)... Don't know what that means. Will search & learn later. I have installed the same way on each machine since joining WCG. Just download BOINC and default install. Now that I have a home LAN I keep the Boinc package in a shared folder and grab it from there for any new install. Hope that answers the question. You said....the slots directory is having issues with permissions. More learning to do later! Drive space usage (for restore I presume) is 3%. Normal HDD inside the box and partition is 45% free. 2/ I have encounted similar with another machine! One of 4 WCG dedicated AMD skt A's I switched to CEP - but is now running Rice. (Another similar machine is running CEP fine - both configured the same.) These machines have a clean install of XP and run with minimal services. Slightly customised UI but otherwise largely default install. Only really running OS plus minimum Avast services and Boinc. No pop blockers or anything like that. This machine is o'clocked. My first such effort - and that's been fun. Hasn't been tweaked for a month or so and is otherwise stable. Exit code 29 errors much later after running for hours. 20/07/2009 1:07:18 a.m. Starting BOINC client version 6.6.20 for windows_intelx86 20/07/2009 1:07:18 a.m. log flags: task, file_xfer, sched_ops 20/07/2009 1:07:18 a.m. Libraries: libcurl/7.19.4 OpenSSL/0.9.8j zlib/1.2.3 20/07/2009 1:07:18 a.m. Data directory: C:\Documents and Settings\All Users\Application Data\BOINC 20/07/2009 1:07:18 a.m. Running under account xxxx 20/07/2009 1:07:18 a.m. Processor: 1 AuthenticAMD AMD Sempron(tm) [x86 Family 6 Model 8 Stepping 1] 20/07/2009 1:07:18 a.m. Processor features: fpu tsc sse 3dnow mmx 20/07/2009 1:07:18 a.m. OS: Microsoft Windows XP: Professional x86 Editon, Service Pack 2, (05.01.2600.00) 20/07/2009 1:07:18 a.m. Memory: 447.48 MB physical, 1.03 GB virtual 20/07/2009 1:07:18 a.m. Disk: 37.04 GB total, 34.76 GB free 20/07/2009 1:07:18 a.m. Local time is UTC +12 hours 20/07/2009 1:07:18 a.m. Configured to not use coprocessors 20/07/2009 1:07:18 a.m. Not using a proxy 20/07/2009 1:07:18 a.m. World Community Grid URL: http://www.worldcommunitygrid.org/; Computer ID: 959350; location: home; project prefs: home 20/07/2009 1:07:18 a.m. World Community Grid General prefs: from World Community Grid (last modified 19-Jul-2009 04:56:08) 20/07/2009 1:07:18 a.m. World Community Grid Computer location: home 20/07/2009 1:07:18 a.m. General prefs: using separate prefs for home 20/07/2009 1:07:18 a.m. Preferences limit memory usage when active to 443.01MB 20/07/2009 1:07:18 a.m. Preferences limit memory usage when idle to 447.48MB 20/07/2009 1:07:19 a.m. Preferences limit disk usage to 34.64GB 20/07/2009 1:07:19 a.m. Suspending computation - initial delay 20/07/2009 1:07:49 a.m. World Community Grid Restarting task CMD2_0018-IREB1A.clustersOccur-ITB5A.clustersOccur_1085_1 using hcmd2 version 614 As I said previously... Machine runs other work OK, and CEP runs on other machines OK. But these 2 just won't get on together! |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Hi.
----------------------------------------2. slots: There one create for each job active, including paused waiting to run. The are located in C:\Documents and Settings\All Users\Application Data\BOINC\slots\ with a number 0 1 2 3 etc. The permissions could not be propagating properly. I've read that doing a project reset should refresh that propagation and force a clean permission setting on all CEP parts in your case. May work. 1. BOINC runs in 2 installation forms a. protected aka service. These load prior to any user logging in to open a OS user session. b. user/all user, no service. These only load BOINC for the person signing-in to the OS, if BOINC is in the startup folder with a shortcut link. Else BOINC has to be started manually. The log indicates single core, with 443/448mb allowed ram use. Should be sufficient combined with 1gb VM to run a single CEP job.
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Jul 23, 2009 8:49:25 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yet another person reporting this problem and using BOINC 6.6.
Coincidence? |
||
|
|
|