Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 14
Posts: 14   Pages: 2   [ 1 2 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 2091 times and has 13 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
applause [RESOLVED] Status = Error on 6 out of 6 "Long" WU's

++ NEW SINCE ORIGINAL POSTING
++ I have changed the topic to show this problem as resolved
++

Of the 6 new longer wu's I have crunched all 6 have failed after running for anywhere from 0.04 to 5.54 hrs.
All have failed with variations on the following error:

Unhandled Exception Detected..
- Unhandled Exception Record -
Reason: Access Violation (0xc0000005) at address 0x7C910CAE
read attempt to address 0xFFC00000

Sometimes it is a write attempt.
The "at address" changes.
The read/write attempt address changes.

Any help or thoughts would be greatly appreciated! - John

System info from Boinc Manager:

||Starting BOINC client version 5.10.30 for windows_intelx86
||log flags: task, file_xfer, sched_ops
||Libraries: libcurl/7.17.1 OpenSSL/0.9.8e zlib/1.2.3
||Data directory: C:\Program Files\BOINC
||Processor: 1 AuthenticAMD AMD Athlon(tm) [x86 Family 6 Model 10 Stepping 0]
||Processor features: fpu tsc sse 3dnow mmx
||OS: Microsoft Windows XP: Professional Edition, Service Pack 3, v.3264, (05.01.2600.00)
||Memory: 1023.48 MB physical, 2.41 GB virtual
||Disk: 189.92 GB total, 89.27 GB free
||Preferences limit memory usage when active to 921.14MB
||Preferences limit memory usage when idle to 972.31MB
||Preferences limit disk usage to 0.00GB

Task info from Boinc Manager:

10:30:20 AM|World Community Grid|Starting dddt0401l0708_100023_1
10:30:20 AM|World Community Grid|Starting task l0708_100023_1 using dddt version 515
10:32:45 AM|World Community Grid|Computation for task dddt0401l0708_100023_1 finished wink Of the 6 new longer wu's I have crunched all 6 have failed after running for anywhere from 0.04 to 5.54 hrs.
All have failed with variations on the following error:

Unhandled Exception Detected..
- Unhandled Exception Record -
Reason: Access Violation (0xc0000005) at address 0x7C910CAE
read attempt to address 0xFFC00000

Sometimes it is a write attempt.
The "at address" changes.
The read/write attempt address changes.

Any help or thoughts would be greatly appreciated! - John

System info from Boinc Manager:

||Starting BOINC client version 5.10.30 for windows_intelx86
||log flags: task, file_xfer, sched_ops
||Libraries: libcurl/7.17.1 OpenSSL/0.9.8e zlib/1.2.3
||Data directory: C:\Program Files\BOINC
||Processor: 1 AuthenticAMD AMD Athlon(tm) [x86 Family 6 Model 10 Stepping 0]
||Processor features: fpu tsc sse 3dnow mmx
||OS: Microsoft Windows XP: Professional Edition, Service Pack 3, v.3264, (05.01.2600.00)
||Memory: 1023.48 MB physical, 2.41 GB virtual
||Disk: 189.92 GB total, 89.27 GB free
||Preferences limit memory usage when active to 921.14MB
||Preferences limit memory usage when idle to 972.31MB
||Preferences limit disk usage to 0.00GB

Task info from Boinc Manager:

10:30:20 AM|World Community Grid|Starting dddt0401l0708_100023_1
10:30:20 AM|World Community Grid|Starting task l0708_100023_1 using dddt version 515
10:32:45 AM|World Community Grid|Computation for task dddt0401l0708_100023_1 finished
----------------------------------------
[Edit 2 times, last edit by Former Member at Mar 10, 2008 9:07:02 PM]
[Mar 7, 2008 6:08:13 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Sekerob
Ace Cruncher
Joined: Jul 24, 2005
Post Count: 20043
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Some pref choices really make me scratch the back of my head:

Preferences limit disk usage to 0.00GB

Please increase this to 2.00GB and BOINC will be allowed to write excess data to disk. When done please post content of the global_prefs.xml and global_prefs_override file in the BOINC data directory. An Start Here forum FAQ shows which field to amend in the "Custom Profile". http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=17033

Let us know if things clear up and when so please edit your opening post title and insert [RESOLVED] before Status = Error on 6 out of 6 "Long" WU's

cheers
----------------------------------------
WCG Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All!
----------------------------------------
[Edit 1 times, last edit by Sekerob at Mar 7, 2008 6:33:52 PM]
[Mar 7, 2008 6:31:15 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

This is a memory error - probably the most common one, and unfortunately there are many potential causes.

The first thing that jumps out at me looking at your log is that you have somehow limited BOINC's disk usage to zero. This is going to cause problems, although not (in theory) the problems you are seeing. But anyway - please fix that first, and see whether anything changes. You have plenty of free disk space, you just need to allow BOINC to use it. Change your settings here: http://www.worldcommunitygrid.org/ms/device/v...iguration.do?name=Default
[Mar 7, 2008 6:33:26 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Thank you for the replies!
I have reset profile to the defaults - cleared local preferences - shut down and restarted Boinc Manager.
Will post a follow-up as soon as I have another WU finish.
- John

3/7/2008 1:10:51 PM||Starting BOINC client version 5.10.30 for windows_intelx86
3/7/2008 1:10:51 PM||log flags: task, file_xfer, sched_ops
3/7/2008 1:10:51 PM||Libraries: libcurl/7.17.1 OpenSSL/0.9.8e zlib/1.2.3
3/7/2008 1:10:51 PM||Data directory: C:\Program Files\BOINC
3/7/2008 1:10:51 PM||Processor: 1 AuthenticAMD AMD Athlon(tm) [x86 Family 6 Model 10 Stepping 0]
3/7/2008 1:10:51 PM||Processor features: fpu tsc sse 3dnow mmx
3/7/2008 1:10:51 PM||OS: Microsoft Windows XP: Professional Edition, Service Pack 3, v.3264, (05.01.2600.00)
3/7/2008 1:10:51 PM||Memory: 1023.48 MB physical, 2.41 GB virtual
3/7/2008 1:10:51 PM||Disk: 189.92 GB total, 89.30 GB free
3/7/2008 1:10:51 PM||Local time is UTC -7 hours
3/7/2008 1:10:51 PM|World Community Grid|URL: http://www.worldcommunitygrid.org/; Computer ID: 426630; location: home; project prefs: default
3/7/2008 1:10:51 PM||General prefs: from World Community Grid (last modified 07-Mar-2008 13:03:14)
3/7/2008 1:10:51 PM||Host location: home
3/7/2008 1:10:51 PM||General prefs: no separate prefs for home; using your defaults
3/7/2008 1:10:51 PM||Preferences limit memory usage when active to 767.61MB
3/7/2008 1:10:51 PM||Preferences limit memory usage when idle to 921.14MB
3/7/2008 1:10:51 PM||Preferences limit disk usage to 3.73GB
3/7/2008 1:10:51 PM|World Community Grid|Restarting task dddt0401m0723_100052_0 using dddt version 515
[Mar 7, 2008 8:19:07 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
confused Re: Status = Error on 6 out of 6 "Long" WU's

Well .... Here is the result of the most recent WU. Note that it was "restarted" after making the changes to the profile and restarting the Manager; though I can't see why it should matter.
I have seen this before - had a bunch of "short" WU's fail out like this just prior to getting the "long" WU's.
I've run DDDT WU's for quite awhile and have only been having these issues since in the past few days.
Out of the last 33 DDDT WU's listed in my Results Status:
11 Valid (All short wu's)
1 Invalid
1 Pending
20 Error

Again, thank you all for the help - John
3/7/2008 1:10:51 PM|World Community Grid|Restarting task dddt0401m0723_100052_0 using dddt version 515
3/7/2008 1:50:54 PM|World Community Grid|Aborting task dddt0401m0723_100052_0: exceeded disk limit: 1675.40MB > 190.73MB
3/7/2008 1:50:59 PM|World Community Grid|Computation for task dddt0401m0723_100052_0 finished
3/7/2008 1:51:03 PM|World Community Grid|Output file dddt0401m0723_100052_0_1 for task dddt0401m0723_100052_0 exceeds size limit.
3/7/2008 1:51:03 PM|World Community Grid|File size: 1750270410.000000 bytes. Limit: 8000000.000000 bytes


[Mar 7, 2008 9:31:00 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Another disk space error? This is odd.

Please will you track down the result status for this work unit, and see whether the other computers running it had a problem?

You seem to have a number of problems: messed up preferences, memory errors, and now an out-of-control result file. I'd say your computer has a stability issue. Do you overclock?
[Mar 7, 2008 9:55:05 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Minor over-clocking.
But the system has been up and running for 18 months or so in it's present configuration without problems.
Ran the UD Client on Grid.org, then the UD client on WCG, and now the Boinc client on WCG (21K credit).
Been running FAAH, HPF, HCC, and DDDT wu's all without problems ... 'till now .... and only seeing problems with DDDT.

But ... I can and will stop over-clocking to see if that is the root cause.

To that end, I've stopped accepting new WU's and suspended the one I had waiting to run.
Since it is the first WU to start from scratch since changing the profile, I will allow the currently running WU to complete as it is 2+ hours into the run with 3+ hours remaining.
If it runs to completion OK, I'll release the next WU.
If the current WU fails, I will reset the clock and then try the next WU.

I will post as soon as I have new information.

As to other WU's w/errors. I can find only one.

dddt0401l0671_100456 ("long" WU) shows
1 Error in addition to mine
1 Pending
1 In Progress

Thanx - John
[Mar 7, 2008 11:24:08 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Hello TheDawgz,
Thanks for giving us so much information in your posts. Now that you have reset the disk limits to allow a work file to be stored, I think that your new work units that have started since then will complete successfully. The DDDT work file has been radically reworked to include several results in each work unit. I do not know the details, but I would not be surprised to find that a disk file is necessary to return all the results in a work unit.

Of course, I am often wrong! smile

Lawrence
[Mar 8, 2008 12:13:58 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Sorry, Lawrence - this is one of those times!

The new disk error is just for a specific file going over the limit. And call me crazy, but a file with an expected size of a few hundred K and a limit of 8 MB should never, ever, exceed a gigabyte - unless the program blew a fuse, as appears to be the case.
[Mar 8, 2008 1:10:22 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Status = Error on 6 out of 6 "Long" WU's

Status Update ---- Sorry for the delay!
Life always has a habit of getting in the way of my best laid plans. wink

To Lawrence's defense - this might not be one of those times!

The WU I allowed to continue running - did; to completion (5+ hrs) and is now Pending Validation!!!

The suspended WU is now running and I will keep you posted.

laughing
Although I may be in the top percentile in the O&O ratings*; without providing the volume of info - I can't imagine how anyone can give me any help - not that I would need it (either the help or the info).
laughing * Omnipotence & Omniscience

So I graciously and without any humility accept Lawrence's compliment!

And I'll take my compliments where I can get them ... 'cuz this is starting to look like a bad case of Operator Error. blushing

-John
[Mar 8, 2008 7:41:11 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 14   Pages: 2   [ 1 2 | Next Page ]
[ Jump to Last Post ]
Post new Thread