Message boards : News : Screenshot display
Message board moderation

To post messages, you must log in.

AuthorMessage
nikogianna
Project administrator
Project developer
Project tester

Send message
Joined: 21 Feb 17
Posts: 21
Credit: 195,770
RAC: 0
Message 4857 - Posted: 25 Apr 2017, 12:55:34 UTC

A new feature has been added. The screenshot that is captured in the event of an error during the job execution is now displayed at the very bottom of the result page. So, if a task errors out, the state of the VM at the time of the error can be seen in the task's results.

We would greatly appreciate any feedback on this new feature.

Thank you for your invaluable support!
ID: 4857 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 781
Credit: 12,422,653
RAC: 1,238
Message 4858 - Posted: 25 Apr 2017, 19:56:12 UTC

ID: 4858 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
nikogianna
Project administrator
Project developer
Project tester

Send message
Joined: 21 Feb 17
Posts: 21
Credit: 195,770
RAC: 0
Message 4861 - Posted: 26 Apr 2017, 7:42:59 UTC
Last modified: 26 Apr 2017, 7:43:19 UTC

It seems that the actual state of the VM was the one displayed by the screenshot since as far as I can see it is working for other cases e.g.

https://lhcathomedev.cern.ch/lhcathome-dev_ops/db_action.php?table=result&id=322790
ID: 4861 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Crystal Pellet
Volunteer tester

Send message
Joined: 13 Feb 15
Posts: 1188
Credit: 862,257
RAC: 9
Message 4864 - Posted: 26 Apr 2017, 9:05:19 UTC
Last modified: 26 Apr 2017, 9:07:24 UTC

Disable the screen blanking in the VM's for all sub-projects is the solution and already a long term wish from the volunteers.

It would also be useful for test purposes when the tester wants screen capturing every few minutes and not able/want to be at the machine all the time.

btw: your link is only accessible to admins.
ID: 4864 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
nikogianna
Project administrator
Project developer
Project tester

Send message
Joined: 21 Feb 17
Posts: 21
Credit: 195,770
RAC: 0
Message 4866 - Posted: 26 Apr 2017, 12:41:05 UTC

Sorry about the link. Here is one that should be working:

https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=323055

About the screen blanking, it is actually disabled for Theory and it will be for the rest of the apps as soon as we feel satisfied with the way Theory works.

About the screen capturing, the most important and crucial part of the VM's life is the moment a task errors out, so that is what we are trying to capture. Periodical screen capturing could be done but it would have to be on an ad hoc basis (e.g. using a script).
ID: 4866 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Crystal Pellet
Volunteer tester

Send message
Joined: 13 Feb 15
Posts: 1188
Credit: 862,257
RAC: 9
Message 4868 - Posted: 26 Apr 2017, 19:57:13 UTC - in response to Message 4866.  
Last modified: 26 Apr 2017, 20:17:21 UTC

Yeah, I already found your example result with the captured screen.

I had a script in the past for capturing, but it was not useful when the screen turns to black after a few minutes.

My windows-script (NL-time format) (where of course the vm-name has to be changed):


@echo off
set nap=60

:START
set dategroup=%date:~9,4%%date:~6,2%%date:~3,2%
set timegroup=%time:~0,2%%time:~3,2%%time:~6,5%
set timegroup=%timegroup:.=%|| delete seperator
set timegroup=%timegroup:~0,6%|| trim hundreds
set timegroup=%timegroup: =0%|| insert leading zero for hours < 10

"%vbox_msi_install_path%\VBoxManage" controlvm boinc_a6f079d732a6ace5 screenshotpng D:\vLHC\Images\%dategroup%-%timegroup%.png
timeout /t %nap% /NOBREAK >NUL
goto START

exit


The disabling of the screen blanking was announced earlier, but did not work.
Maybe in this latest version. I'm testing it now.

Tested: The screen blanking is still enabled. The screen turns to black after about 10 minutes.
ID: 4868 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
nikogianna
Project administrator
Project developer
Project tester

Send message
Joined: 21 Feb 17
Posts: 21
Credit: 195,770
RAC: 0
Message 4869 - Posted: 27 Apr 2017, 15:11:50 UTC

The screen is blanking indeed. I will look into it. As a temporary workaround you could add a keyboard space (0x39) input to your script right before the screenshot to "wake" the VM. That is what the vboxwrapper is currently doing.
ID: 4869 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Crystal Pellet
Volunteer tester

Send message
Joined: 13 Feb 15
Posts: 1188
Credit: 862,257
RAC: 9
Message 4870 - Posted: 28 Apr 2017, 8:59:38 UTC - in response to Message 4869.  

The screen is blanking indeed. I will look into it. As a temporary workaround you could add a keyboard space (0x39) input to your script right before the screenshot to "wake" the VM. That is what the vboxwrapper is currently doing.

Thanks for pointing me to that possibility.
I use Left-Alt (0x38). That's what I normally do by hand when waking up a guest console. For the host I normally use Left-Shift.
For me 'space' is still an input character that theoretically could be stored in documents or whatever.
ID: 4870 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
computezrmle
Volunteer moderator
Project tester
Volunteer developer
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 28 Jul 16
Posts: 484
Credit: 394,839
RAC: 0
Message 4871 - Posted: 28 Apr 2017, 11:25:31 UTC

... adding the kernel parameter "consoleblank=0" to the bootmanager.

Have you ever tried this solution?

https://lhcathomedev.cern.ch/lhcathome-dev/forum_thread.php?id=324&postid=4454
ID: 4871 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
nikogianna
Project administrator
Project developer
Project tester

Send message
Joined: 21 Feb 17
Posts: 21
Credit: 195,770
RAC: 0
Message 4872 - Posted: 28 Apr 2017, 14:12:31 UTC

We were going with "setterm -blank 0", but it obviously wasn't working. Trying to understand why. Possibly not surviving reboots
ID: 4872 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
computezrmle
Volunteer moderator
Project tester
Volunteer developer
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 28 Jul 16
Posts: 484
Credit: 394,839
RAC: 0
Message 4873 - Posted: 28 Apr 2017, 14:55:59 UTC - in response to Message 4872.  

We were going with "setterm -blank 0", but it obviously wasn't working. Trying to understand why. Possibly not surviving reboots


You may read this:
https://superuser.com/questions/152347/change-linux-console-screen-blanking-behavior

To me it seems that setterm does not work on every terminal while consoleblank=0 as boot parameter works directly on the VM´s kernel.
ID: 4873 · 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: 6
Message 4874 - Posted: 29 Apr 2017, 10:18:10 UTC

this one worked but I assume it was lucky timing?

https://lhcathome.cern.ch/lhcathome/result.php?resultid=136767548
ID: 4874 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
computezrmle
Volunteer moderator
Project tester
Volunteer developer
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 28 Jul 16
Posts: 484
Credit: 394,839
RAC: 0
Message 4875 - Posted: 30 Apr 2017, 11:42:58 UTC

I tested consoleblank=0 on one of my own linux VMs.
It completely disables the console screensaver as expected and therefore does not need any workaround to wake a sleeping console.

Do I miss a requirement that makes it impossible to use this solution?
ID: 4875 · 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 4883 - Posted: 2 May 2017, 12:02:04 UTC - in response to Message 4875.  

As we are using CernVM, doing this is not straight forward which is why the the setterm method was tried. It is being followed up.
ID: 4883 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : News : Screenshot display


©2024 CERN