41) Message boards : Number crunching : Communication with VM Hypervisor failed (Message 1254)
Posted 17 Oct 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Hi

I started again CMS-dev on my Mac to see how it goes, and it goes wrong , I find the WU stuck with status "Communication with VM Hypervisor failed" repeating all the time in boinc message log, I canceled one WU yesterday and this morning I find another one in the same status,

the stederr.txt says

2015-10-17 07:35:27 (97085): vboxwrapper (7.7.26175): starting
2015-10-17 07:35:27 (97085): Feature: Checkpoint interval offset (290 seconds)
2015-10-17 07:35:27 (97085): Detected: VirtualBox VboxManage Interface (Version: Unknown)
2015-10-17 07:35:27 (97085): Detected: Sandbox Configuration Enabled
2015-10-17 07:35:28 (97085): WARNING: Communication with VM Hypervisor failed.
2015-10-17 07:35:28 (97085): Could not communicate with VM Hypervisor. Rescheduling execution for a later date.


and the vbox_trace.txt

2015-10-17 07:35:27 (97085):
Command: VBoxManage -q --version
Exit Code: 0
Output:
sh: VBoxManage: command not found

2015-10-17 07:35:27 (97085):
Command: VBoxManage -q list systemproperties
Exit Code: 0
Output:
sh: VBoxManage: command not found

2015-10-17 07:35:28 (97085):
Command: VBoxManage -q list hostinfo
Exit Code: 0
Output:
sh: VBoxManage: command not found


It seems an issue with CMS not being able to run VB, but when starting boinc it finds VB correctly :

12-Oct-2015 20:14:34 [---] VirtualBox version: 4.3.30r101610


I have boinc 7.6.12 (stable version for Mac) with OS X 10.11.

Thanks for your help.
42) Message boards : Number crunching : Kaboom on my Mac (Message 428)
Posted 2 Jun 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
True :D

I was doing 2 threads at the same time (here + Atlas for another different issue) and I got confused ;)
43) Message boards : Number crunching : Kaboom on my Mac (Message 426)
Posted 1 Jun 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Thanks for the quick answer.

I forgot to mention : I use a Mac.
44) Message boards : Number crunching : Kaboom on my Mac (Message 423)
Posted 1 Jun 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Hi

Mmmm I don't think it's good :



(WU was still "running" when I found it like this)

Logs + screnshot zip (from the slot dir, before I canceled the WU) : http://cl.ly/bSFo/download/Logs%20cmsdev.zip

(tell me if you can't reach it)

Thanks
45) Message boards : News : A Message To All Our Volunteers (Message 170)
Posted 24 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Cool, thanks for the feedback, I'll be around :)
46) Message boards : News : A Message To All Our Volunteers (Message 167)
Posted 24 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Ok so now I got 3 WUs successfully ending on my iMac (and another one currently crunching) and I can see all seems to go well so what you say is that I can stop CMS for the moment and let other projects go on instead ?
47) Message boards : News : A Message To All Our Volunteers (Message 150)
Posted 22 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
OK I understand better, thanks.
48) Message boards : News : A Message To All Our Volunteers (Message 140)
Posted 21 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Hi

Thanks a lot for the information.

I'm a bit skeptical though : we are helping you starting an infrastructure for a new project, however when it works well you will move it to an existing infrastructure (vLHC) as "another project", so what's the point ? Am I missing something ?
49) Message boards : News : New Release (v45) (Message 100)
Posted 18 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
I got a second WU at noon and it seems much better, there are big juicy logs inside the VM full of interesting things, the console looks much better too and the CPU is happily crunching :)
50) Message boards : News : New Release (v45) (Message 92)
Posted 18 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Ah too bad I already aborted it, but there was also another reason : there was also a vLHC task running, and during the night the CMS task ended in the most hatred "unmanageable" state, I have this problem whenever (but not always) two VM based project run at the same time (Ben I think you know about that).

So for the moment I set vLHC and Atlas to "no more work" (RNA doesn't have any at the moment I think) and I raised CMS resource share, so I hope to get another task soon and see what happens.
51) Message boards : News : New Release (v45) (Message 88)
Posted 17 Mar 2015 by [AF>Le_Pommier] Jerome_C2005
Post:
Hi everybody

I just got for my very first (big) WU for my iMac but it's not running well :

tail: /home/boinc/stderr: file truncated
tail: /home/boinc/stderr: file truncated
tail: /home/boinc/stderr: file truncated
tail: /home/boinc/stderr: file truncated
etc

repeating all the time in the console,

[17/03/15 22:01:24] ERROR:root:No message received! Nothing to do!

all the time in the CMSJobAgent-stderr.log

I guess I have to abort it...


Previous 20


©2024 CERN