| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: All my linux cep2 wus now failing |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 11
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Help, please. Starting tonight all existing and new cep2 work downloading to my linux host (andLinux under windows, actually) is failing, details in a minute. It seems to be running on Windows, and other WCG projects are running on the andLinux. I use andLinux to run only CEP2 under normal circs and have done so for many months with no problems.
----------------------------------------Sample stderr excerpt is: process exited with code 195 (0xc3, -61) Application exited with RC = 0x100 [ERROR] Failed to open either source or destination files while copying C.32.C28H18N2SSi.00651123.2.noopt.bp86.sto6g.n.sp/53.0 to C.32.C28H18N2SSi.00651123.2.noopt.bp86.sto6g.n.sp.53.0. Error: 2 [03:13:52] Finished Job #0 03:13:52 (2500): called boinc_finish All the errors look like this, trying to copy < something > .n.sp/53.0 to < something > .n.sp.53.0 I tried downloading and copying the zip file and inspecting it elsewhere. I can't see anything different in its contents than what I find in cep2 zip files on my windows host. Taking the example above, if that "from" file is a literal pathname then no, there is no such file, there are no files with a directory in the pathname, the zipfile is flat. No files with "53 "anywhere in the name, so errno 2 makes sense, but not knowing what normal processing of the zip files is like I have no real idea what's going on here. Most of the wus show only in-progress wingmen. I've found at least one wu where a wingman got the same error, but the other wingman is in pending validation... its result was returned earlier, so is there something screwy since the maintenance? If not, can someone help me understand how to troubleshoot at my end? Thanks ETA: more wus showing wingman errors but some with wingman success. Really would like to understand what this problem is about and to get this host back crunching ceo2. Any help appreciated. [Edit 3 times, last edit by Former Member at Mar 18, 2013 12:52:58 PM] |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7846 Status: Offline Project Badges: 
|
My first inclination for you would be to reboot, unless you have tried this already.
----------------------------------------Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sorry I should have said, yes I have restarted to no avail. Since all the work had been aborted all slots were also empty of files. Can't pursue this until I'm home again this evening... but other than watching it error out again I'm not sure what I can do :(
----------------------------------------Reinstalling andLinux and boinc -- the very latest boinc won't run on it -- is more trouble and time than I'm willing to spend without more than a vauge hope it would make a difference. Since some of my wingmen are getting this same error there must be some known cause? It might help if I understood where this *.n.sp/53.0 file is supposed to come from. Nothing like that is in any cep2 zip file I've seen on linux or windows host. What script or software is doing what, to which of the zip file's contents, that wants such a file? [Edit 2 times, last edit by Former Member at Mar 18, 2013 6:05:01 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
1) Reset project since the cores are idle anyhow.
2) Run BOINC installer again and choose repair, to reconfirm/reset various file ownerships |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The first I will do when I get home, thanks should have thought of it.
----------------------------------------The second, I don't recall seeing as an option under linux but will check. Thanks for your response. p.s. um, I did say linux in this thread, and it's also the thread title.... anyway I appreciate all the info. I think flushing the slot with another wcg project may have fixed things ans the one cep2 wu I planned to let abort before resetting, is now running. I will wonder what the /53.0 errors were about, but the older I get the less time that kind of wondering lasts... [Edit 1 times, last edit by Former Member at Mar 18, 2013 10:37:10 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well, you did not specify which platform [Will not attempt to remember from past threads you may have posted this in... members change their machine(s) without notice ;>], Running the installer under Linux does in effect same, in a roundabout way Remove/Install. In Synaptic you have the "Mark for Reinstallation". It's lossless, at least for me has not overwritten custom cc_config.xml
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
well I got 2 tasks through then it went south again and blew my quota since it happened over night. I wish I knew prcisely what the #@$%^&* messages quoted above meant was really happening. Clearly it's happening to some others as I see the same messages in their error status.
----------------------------------------(link to that post) Resetting & detaching/re-attaching haven't helped. Will have to see what I can find in linux client repair options next (platform is in the thread title as well as the 1st post Can't try anything for another day because of the blown quota. [Edit 1 times, last edit by Former Member at Mar 19, 2013 5:53:33 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sorry, haha the platform was in the title
 195, *with* a minus sign is science app generated, *without* a minus sign is system/OS related. Post the whole log, for the readers satisfaction... 4 eyes see more. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks SekeRob, according to that it's app/OS Here's a log from yesterday, followed by a similar one from a wingman. It's the from/to filenames that are bugging me, as nothing like the "from" name is in the zip file, and none of the running jobs had file names like the "to name. I bolded the relevant characters otherwise the log text is unchanged.
----------------------------------------I am running an old client as I couldn't get the newer one to install under andLinux and this had been running fine for so long. The wingman however is on a 7.x client my result log Result Name: E212308_ 720_ A.35.C22H8N8OS4.16.4.set1d06_ 1-- <core_client_version>6.10.56</core_client_version> <![CDATA[ <message> process exited with code 195 (0xc3, -61) </message> <stderr_txt> INFO: No state to restore. Start from the beginning. [03:11:42] Number of jobs = 16 [03:11:42] Starting job 0,CPU time has been restored to 0.000000. [03:11:42] Starting new Job [03:11:42] Qink name = fldman [03:11:43] Qink name = gesman Application exited with RC = 0x100 [ERROR] Failed to open either source or destination files while copying A.35.C22H8N8OS4.16.4.noopt.bp86.sto6g.n.sp/53.0 to A.35.C22H8N8OS4.16.4.noopt.bp86.sto6g.n.sp.53.0. Error: 2 [03:11:43] Finished Job #0 03:11:43 (2482): called boinc_finish </stderr_txt> ]]> wingman result log: Result Name: E212308_ 720_ A.35.C22H8N8OS4.16.4.set1d06_ 0-- <core_client_version>7.0.27</core_client_version> <![CDATA[ <message> process exited with code 195 (0xc3, -61) </message> <stderr_txt> INFO: No state to restore. Start from the beginning. [19:25:06] Number of jobs = 16 [19:25:06] Starting job 0,CPU time has been restored to 0.000000. [19:25:06] Starting new Job [19:25:06] Qink name = fldman Application exited with RC = 0x100 [ERROR] Failed to open either source or destination files while copying A.35.C22H8N8OS4.16.4.noopt.bp86.sto6g.n.sp/53.0 to A.35.C22H8N8OS4.16.4.noopt.bp86.sto6g.n.sp.53.0. Error: 2 [19:25:07] Finished Job #0 19:25:07 (9807): called boinc_finish </stderr_txt> ]]> [Edit 1 times, last edit by Former Member at Mar 19, 2013 6:31:38 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It's that "failed to open" is indeed the bugger. If the wingman does same, then a bad WU build would be a suspected cause, but my Linux 12.10 with 7.0.27 and since a few months 7.0.39 has been doing fine with CEP2. Have not tried my 13.04 with 7.0.56 client (x86), but that does not matter as it will handle 64 bit sciences just fine, albeit CEP2 is only 32bits. Maybe the upgrade will have that [don't know when, this year]. At start the science unpacks some 6700 files... not like you have a space problem, do you?
|
||
|
|
|