Message boards : Number crunching : Update to VB 5.1.2
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,361,368
RAC: 3,981
Message 3958 - Posted: 5 Aug 2016, 3:10:25 UTC

Disaster

Never had this problem before and they are ALL crashing.

http://lhcathomedev.cern.ch/vLHCathome-dev/result.php?resultid=232086

So far Atlas and vLHC are not having any problem with this update.

I even tried a *reset*
Mad Scientist For Life
ID: 3958 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile tullio

Send message
Joined: 17 Aug 15
Posts: 62
Credit: 296,695
RAC: 0
Message 3959 - Posted: 5 Aug 2016, 7:51:22 UTC

My LHCb tasks are all crashing on my Linux box with VirtualBox 5.1.2. CMS tasks work perfectly. Multicore apps crash, also because my Opteron 1210 has only 2 cores.
Tullio
ID: 3959 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Toby Broom

Send message
Joined: 19 Aug 15
Posts: 46
Credit: 3,626,553
RAC: 69
Message 3970 - Posted: 5 Aug 2016, 21:16:14 UTC
Last modified: 5 Aug 2016, 21:33:30 UTC

I just updated today, seems OK so far with multicore.

ALICE & CMS failed:

2016-08-05 23:29:47 (8104): Error 0x80004001 in vbox51::VBOX_VM::create_vm (c:\src\boinc\boinc\samples\vboxwrapper\vbox_mscom_impl.cpp:359)
2016-08-05 23:29:47 (8104): Error: Getting Error Info! hr = 0x1
ID: 3970 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Toby Broom

Send message
Joined: 19 Aug 15
Posts: 46
Credit: 3,626,553
RAC: 69
Message 3975 - Posted: 6 Aug 2016, 8:33:04 UTC

Spoke too soon, just errors everywhere after running overnight.
ID: 3975 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile tullio

Send message
Joined: 17 Aug 15
Posts: 62
Credit: 296,695
RAC: 0
Message 3976 - Posted: 6 Aug 2016, 9:33:28 UTC

My Benchmark also failed on the Linux box with VBox 5.1.2
Tullio
ID: 3976 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Laurence
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 12 Sep 14
Posts: 1069
Credit: 334,882
RAC: 0
Message 3986 - Posted: 6 Aug 2016, 20:56:35 UTC - in response to Message 3976.  

Have informed Rom.
ID: 3986 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Rasputin42
Volunteer tester

Send message
Joined: 16 Aug 15
Posts: 966
Credit: 1,211,816
RAC: 0
Message 3988 - Posted: 7 Aug 2016, 10:06:41 UTC

CMS appears to be working, if you use vboxwrapper 26196.
ID: 3988 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,361,368
RAC: 3,981
Message 3994 - Posted: 7 Aug 2016, 21:21:30 UTC - in response to Message 3988.  

I just tried a couple CMS version with VB 5.1.2 and in less than 3 seconds they both crashed.

http://lhcathomedev.cern.ch/vLHCathome-dev/result.php?resultid=234118
Mad Scientist For Life
ID: 3994 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Rasputin42
Volunteer tester

Send message
Joined: 16 Aug 15
Posts: 966
Credit: 1,211,816
RAC: 0
Message 3995 - Posted: 7 Aug 2016, 21:23:55 UTC - in response to Message 3994.  
Last modified: 7 Aug 2016, 21:24:14 UTC

CMS appears to be working, if you use vboxwrapper 26196.



I have taken the one from theory and renamed it 26193.

Works fine.
ID: 3995 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,361,368
RAC: 3,981
Message 3996 - Posted: 7 Aug 2016, 21:30:45 UTC - in response to Message 3995.  

CMS appears to be working, if you use vboxwrapper 26196.



I have taken the one from theory and renamed it 26193.

Works fine.



I guess I will try Theory after I get that 225MB .vdi d/l'd

(it looks like that will take a long time on this DSL)
ID: 3996 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,361,368
RAC: 3,981
Message 4041 - Posted: 10 Aug 2016, 7:29:35 UTC
Last modified: 10 Aug 2016, 8:00:12 UTC

Ok I took a couple days off try this VB version and just a few minutes ago decided to give it another try (and I did see Rasputin got a Valid tasks using this 7.7.26197 wrapper)

So far it looks like it is running but just now when I checked the RDC it says this-----



And when I X'd out of the VB manager it PAUSED the task and then made it a computer error.

THEN I try it on an Atlas task that has been running fine and when I X'd out of the RDC it KILLED that task too.

Glad I didn't even try this with the vLHC X2 I also have running since I already just wasted 9 hours of that Atlas task.

SO.......either this version of VB is bad and needs a clean reinstall....or I just am having no %$#@ luck with it and should just go back to VB 5.0.14

Oy Vey

Edit: Ok I just did the regular Oracle VB *repair* and did a reboot and am going to give this another try.

And also check the RDC and see if it says it is killed and also just plain computer error out again........then I will decide if I am doing a clean reinstall of this VB version.......since as I mentioned I always have got Valid Atlas and vLHC tasks with this version (but didn't use the RDC and kill the tasks when X'ing out)

Either that or just reinstall VB version 5.0.14........never had this happen since way back with the 2011 version

OH and after I did this and went to start a new vLHC-dev task once again it is making me D/L the 680MB .vdi and as always my D/L's from Cern take HOURS and this never happens when I do this at the Einstein server......ridiculous 3 hours or more.
Mad Scientist For Life
ID: 4041 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Rasputin42
Volunteer tester

Send message
Joined: 16 Aug 15
Posts: 966
Credit: 1,211,816
RAC: 0
Message 4044 - Posted: 10 Aug 2016, 10:46:02 UTC
Last modified: 10 Aug 2016, 10:46:37 UTC




This is completely normal.
ID: 4044 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,361,368
RAC: 3,981
Message 4052 - Posted: 11 Aug 2016, 9:56:14 UTC - in response to Message 4044.  
Last modified: 11 Aug 2016, 9:58:01 UTC

It never went beyond that.

That doesn't happen with any of my Valids

And the stderr has this

<message>
The filename or extension is too long.
(0xce) - exit code 206 (0xce)
</message>
Mad Scientist For Life
ID: 4052 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
maeax

Send message
Joined: 22 Apr 16
Posts: 677
Credit: 2,002,766
RAC: 2
Message 4054 - Posted: 11 Aug 2016, 12:36:26 UTC
Last modified: 11 Aug 2016, 12:40:29 UTC

Hi Magic,

are this errors from CMS with Multicore?

Will test this with Virtualbox 5.0.26.
ID: 4054 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,361,368
RAC: 3,981
Message 4057 - Posted: 11 Aug 2016, 20:43:52 UTC - in response to Message 4054.  

Hi Maeax,

Well they are the CMS Simulation v47.40 (vbox64_mt_mcore)
windows_x86_64 tasks but I have my preferences set at *1* so it just uses one core.

The only problem is running these with VB version 5.1.2 and I only have one computer running with this version (it works no problem with vLHC and Atlas)

I have 2 other computers running these same vLHC-dev tasks with the VB 5.0.14 and they always get Valids

This is the host having the problem here with the new VB version http://lhcathomedev.cern.ch/vLHCathome-dev/results.php?hostid=272

It did run multi-core tasks when it had VB 5.0.14 for the original tests before they started here. (this host ran 6-cores for those tests)

I may try one more clean install of this VB version when I get a chance but right now I don't want to just waste the other vLHC X2 tasks while they are running with no problems (you probably remember why on our milestone thread over there)

Tonight is meteor shower night so I am hoping the clouds stay away for another 24 hours
Mad Scientist For Life
ID: 4057 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
maeax

Send message
Joined: 22 Apr 16
Posts: 677
Credit: 2,002,766
RAC: 2
Message 4060 - Posted: 12 Aug 2016, 3:41:36 UTC
Last modified: 12 Aug 2016, 4:01:51 UTC

Hi Magic,

have CMS(mcore) with ONE task and ONE cpu started. Vbox 5.0.26.

In the moment the third Condor process is running.

Have you message in stderr.log.

(Boincmanager -> show graphics -> machine logs -> Index of logs -> stderr.log

Meteor shower is behind the Clouds :-(
ID: 4060 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Toby Broom

Send message
Joined: 19 Aug 15
Posts: 46
Credit: 3,626,553
RAC: 69
Message 4073 - Posted: 18 Aug 2016, 17:38:05 UTC

I updated to 5.1.4 and VLHC (all LHCb) is fine as well as Atlas.

On here the CMS, ALICE task seem to fail, but the Theory and Benchmark apps are good.
ID: 4073 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Number crunching : Update to VB 5.1.2


©2024 CERN