Message boards : ATLAS Application : Testing CentOS 7 vbox image
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 3 · 4 · 5 · 6

AuthorMessage
computezrmle
Avatar

Send message
Joined: 28 Jul 16
Posts: 401
Credit: 374,791
RAC: 48
Message 6773 - Posted: 22 Oct 2019, 14:40:38 UTC - in response to Message 6772.  

You may also test if the top service automatically restarts if you hit 'q' or 'CTRL-c'.
But be aware:
The systemd defaults allow not more than 5 restarts within 10 seconds.
If this rate limit is exceeded the service will not (automatically) restart any more.
ID: 6773 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
David Cameron
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 20 Apr 16
Posts: 154
Credit: 1,334,358
RAC: 549
Message 6774 - Posted: 22 Oct 2019, 15:29:10 UTC

There is now an experimental new version of console 2 available from computezrmle, unfortunately it doesn't work for me but I have put some WU in here for debugging. These are short test tasks with 10 events.
ID: 6774 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Crystal Pellet
Volunteer tester

Send message
Joined: 13 Feb 15
Posts: 1146
Credit: 752,572
RAC: 1,245
Message 6775 - Posted: 22 Oct 2019, 19:27:21 UTC - in response to Message 6774.  

montty2: permission denied
ID: 6775 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
David Cameron
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 20 Apr 16
Posts: 154
Credit: 1,334,358
RAC: 549
Message 6776 - Posted: 23 Oct 2019, 9:21:24 UTC - in response to Message 6775.  

After some fixes from computezrmle it works for me now:

ID: 6776 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Crystal Pellet
Volunteer tester

Send message
Joined: 13 Feb 15
Posts: 1146
Credit: 752,572
RAC: 1,245
Message 6777 - Posted: 23 Oct 2019, 9:25:41 UTC - in response to Message 6776.  
Last modified: 23 Oct 2019, 9:33:53 UTC

After some fixes from computezrmle it works for me now:


I want to see that myself ;) Kudos to computezrmle !!

btw: When I have no new tasks in queue and ask a new one, everytime the huge 0.84 vdi is downloaded again. Any idea??
ID: 6777 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
David Cameron
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 20 Apr 16
Posts: 154
Credit: 1,334,358
RAC: 549
Message 6778 - Posted: 23 Oct 2019, 10:43:35 UTC - in response to Message 6777.  

btw: When I have no new tasks in queue and ask a new one, everytime the huge 0.84 vdi is downloaded again. Any idea??


I also see this whenever I reconnect to the dev project after not running for a while, but I'm doing a steady stream of work it doesn't download it again. Not sure what the problem is but maybe it's due to re-activating deprecated versions.
ID: 6778 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
computezrmle
Avatar

Send message
Joined: 28 Jul 16
Posts: 401
Credit: 374,791
RAC: 48
Message 6779 - Posted: 23 Oct 2019, 10:53:10 UTC - in response to Message 6778.  

I don't see this issue.
Before I started the monitoring test I did a project reset, got the recent vdi and have a dry buffer while I think about changes in the script.
When I get fresh work the already downloaded vdi is used.
ID: 6779 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
David Cameron
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 20 Apr 16
Posts: 154
Credit: 1,334,358
RAC: 549
Message 6780 - Posted: 23 Oct 2019, 12:42:57 UTC

I put some real tasks in now, and it looks good:



Only 7 hours left to go...
ID: 6780 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
computezrmle
Avatar

Send message
Joined: 28 Jul 16
Posts: 401
Credit: 374,791
RAC: 48
Message 6781 - Posted: 23 Oct 2019, 15:24:26 UTC

I got a 6-core with 200 events.
11 events done.
Runtimes 2474-3730 s per event (!!)
Time left: 1d 10h 20m
ID: 6781 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
computezrmle
Avatar

Send message
Joined: 28 Jul 16
Posts: 401
Credit: 374,791
RAC: 48
Message 6782 - Posted: 24 Oct 2019, 17:19:11 UTC - in response to Message 6781.  

I got a 6-core with 200 events.
11 events done.
Runtimes 2474-3730 s per event (!!)
Time left: 1d 10h 20m

That task is still running.
It just finished an event after 4302 s.

182 events done.
Still 3h to go.
ID: 6782 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Ray Murray
Avatar

Send message
Joined: 13 Apr 15
Posts: 132
Credit: 2,384,668
RAC: 295
Message 6802 - Posted: 2 Nov 2019, 19:11:24 UTC
Last modified: 2 Nov 2019, 19:47:07 UTC

196 (0x000000C4) EXIT_DISK_LIMIT_EXCEEDED
Quite annoying after 4+ days as it otherwise ran fine.

Logs show
2019-11-01 23:36:09 (1796): Guest Log: HITS file was successfully produced

2019-11-01 23:36:42 (1796): Guest Log: Successfully finished the ATLAS job!
and
2019-11-01 23:36:49 (1796): Guest Log: *** Success! Shutting down the machine. ***
so it would seem the error happened during post-processing.

Forgot to mention
I'm also seeing the full download if there is a break in work. If work is contiguous it only downloads the stuff to run the new job but if there is a break between finishing and uploading a job and requesting a new one, the vdi is downloaded again.
I've not checked if the vdi is being deleted on completion or if it is being overwritten. I'll look 2moro.
ID: 6802 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
maeax

Send message
Joined: 22 Apr 16
Posts: 601
Credit: 1,376,546
RAC: 2,309
Message 6803 - Posted: 3 Nov 2019, 9:43:06 UTC - in response to Message 6802.  

This is from my last successful Atlas-task in -dev:
2019-10-24 22:59:19 (12188): Guest Log: total 449184
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 209689841 Oct 23 20:17 ATLAS.root_0
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 244613206 Oct 24 20:56 HITS.pool.root.1
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 8748 Oct 23 20:17 init_data.xml
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 268662 Oct 23 20:14 input.tar.gz
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 5355520 Oct 24 2019 result.tar.gz
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 815 Oct 23 20:14 RTE.tar.gz
2019-10-24 22:59:19 (12188): Guest Log: -rwxrwxrwx. 1 root root 8659 Oct 23 20:14 start_atlas.sh
2019-10-24 22:59:19 (12188): Guest Log: *** Success! Shutting down the machine. ***

This is from your unsuccessful task- 18.79 GByte result-file against 5.36 from my task.

2019-11-01 23:36:48 (1796): Guest Log: total 459028
2019-11-01 23:36:48 (1796): Guest Log: -rwxrwxrwx. 1 root root 209392385 Oct 28 18:51 ATLAS.root_0
2019-11-01 23:36:48 (1796): Guest Log: -rwxrwxrwx. 1 root root 241559602 Nov 1 23:32 HITS.pool.root.1
2019-11-01 23:36:48 (1796): Guest Log: -rwxrwxrwx. 1 root root 5923 Oct 28 18:51 init_data.xml
2019-11-01 23:36:48 (1796): Guest Log: -rwxrwxrwx. 1 root root 268665 Oct 28 07:44 input.tar.gz
2019-11-01 23:36:48 (1796): Guest Log: -rwxrwxrwx. 1 root root 18790400 Nov 1 2019 result.tar.gz
2019-11-01 23:36:48 (1796): Guest Log: -rwxrwxrwx. 1 root root 815 Oct 28 07:44 RTE.tar.gz
2019-11-01 23:36:49 (1796): Guest Log: -rwxrwxrwx. 1 root root 8659 Oct 28 07:44 start_atlas.sh
2019-11-01 23:36:49 (1796): Guest Log: *** Success! Shutting down the machine. ***

Maybe, you reached a limit because of 8 GByte RAM, or other limitations of the ending files in Boinc or Windows.
Very painful, after such a long time running successful.
ID: 6803 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 . . . 3 · 4 · 5 · 6

Message boards : ATLAS Application : Testing CentOS 7 vbox image


©2022 CERN