Quick!!! Join LHC 2.0 Beta!

[waffleironhead]

post this on the official forums to let them know about it

I reported it, they are claiming some network issues. I'm sure they are going to blame something on my end, but just what exactly I'm not sure. Quite a few people are having the same exact issue.

 

[Philippart]

I reported it, they are claiming some network issues. I'm sure they are going to blame something on my end, but just what exactly I'm not sure. Quite a few people are having the same exact issue.

the thing you could do, just to be sure, is:

uninstall virtualbox, reinstall it paying special attention to the network drivers it installs

 

[waffleironhead]

the thing you could do, just to be sure, is:

uninstall virtualbox, reinstall it paying special attention to the network drivers it installs

I've reinstalled. Not sure where to observe the network driver it uses. Everything is fairly obscured when running the installer.

 

[Philippart]

new app version

Version 6.03 released
Dear all,

We are very happy to announce a new version of the wrapper. This new version has two new improvements:


A Windows performance improvement, and
A Separate preference for setting how much CPU should use the VM.


The first item fixes a performance issue reported by several volunteers in the Windows platform, and the second one, adds a new preference to the project.

We have included a specific project preference to set up how much CPU time will be used by the VM. This will allow you to choose how much CPU the VM should use in your computer without affecting at all your other BOINC projects (this was the previous behavior).

In order to change this preference, go to your account, and change the Preferences for the project. The item is called: Maximum CPU % for Virtual Machine below Maximum CPU % for graphics.

After changing the value, you have to update the project, and restart BOINC.

As usual any comments, suggestions, bug reports, etc. will be welcome!!

Regards,

The Team 9 Aug 2011 9:06:17 UTC

 

[HAL9000]

Like the OPs avatar

Also this is very interesting, I've signed up, hope to see this go somewhere.

 

[waffleironhead]

Damn. My first wu was at 23 hours and I as going to do a reset and accidentally hit snapshot.
ERROR.
The WU aborted and now I have zero credit.

I'm going to report this to the official message board. I'm probably going to wait till they have the connectivity issue figured out before doing anymore beta testing for them

 

[Sunny129]

Damn. My first wu was at 23 hours and I as going to do a reset...

what for?

 

[Philippart]

@ Fardringle, since the new app version lets you configure the VM cpu usage and/or you can lower the process priority (manually or through some thirdparty software), Fardringle issue should be fixed now.

@ HAL9000, I read you Dave Is this your first Boinc project? I haven't seen you before on the distributed computing forums

@all: the project admin just confirmed that although the wrapper and project is still in beta, the work we do in the VM is already real science and the results are inserted to the LHC database (linked on the project website)

 

[waffleironhead]

what for?

It was what the Admins recommended if the network error appears. You reset the VM, and then hope you get a WU that is able to be uploaded.

 

[waffleironhead]

Well I keep getting these network issues. The admins have admitted that this is a months old problem, found back in alpha testing. They have admitted they think this is a small issue to them. I dont have a lot of time to babysit the VM and reboot and lose work done.

So, I may be back when its out of beta, but so far I think Ive done enough beta testing for them. Issue found and reported. The ball is in their court for fixing it. Losing 8+ hours of cpu time because their system wont tell mine where to send the result is disheartening. I can contact and receive work, I would imagine the same system is used for download of work as for upload. I'll check back in a month to see if they have the issue fixed.

 

[Sunny129]

i've had the same network problems all morning too. but unlike previous times where a simple reset did the trick (thanks for explaining the reset function to me by the way), and despite the fact that the BOINC_VM is telling us that there is an acutal network communication issue, i checked the server status page only to find that there have only been a handful of tasks ready to send at any given time...and its been that way since this morning.

see THIS thread on the T4T@H forums.

 

[Philippart]

I've checked my computers,
actually if you let the VM run it will eventually get work (my computers did), the work supply is just a bit short of jobs at the moment (the upload servers are busy too). I bet it's the sudden success the project has gotten thanks to the BBC article.

http://www.bbc.co.uk/news/science-environment-14488797

 

[Sunny129]

PROBLEM WITH JOB SUPPLY FROM CERN