VM CPID Problem

[Bryan]

I have a Linux VM on this laptop and when I use it the project stats get moved to a different CPID. This has happened on 2 different projects. To get them showing in stats again under the correct CPID I can use the same machine running Windows and the project appear again.

I'm really at a loss as to how to get the Linux VM to pick up the correct CPID from the project. Any ideas?

[Maxwell]

A) (the forehead-smack question) Are you using the correct email address when signing up for those projects?
B) What projects and what are you running? The split CPID can often be solved by crunching several projects that you're also crunching on other computers with the correct CPID. Usually takes a day or two to propagate throughout the stats sites...

[Bryan]

I'm attaching to projects that have the correct CPID and have been run by all my computers. When I attach this laptop (no other machines crunching the project) then the CPID gets changed by the VM and the project gets moved away from my primary CPID.

It has happened on MindModeling and Wanless. When I then crunch a few WU with my Windows BOINC on the same machine the projects move back to the correct stats page. It takes a day for them to appear again.

[Maxwell]

Do you have WUProp and other NCI projects running on both Windows BOINC and the Linux VM on this machine?

Another thing to try is adding projects to the VM that you're running on other computers with the correct CPID. May even try crunching on both Windows and the Linux VM for a bit to try and get this sorted out...

[Fire$torm]

Does the VM have a unique machine name?

[Bryan]

Does the VM have a unique machine name?

Yes sir.

[Mumps]

Best thing to do is find a project you crunched *before* MindModeling and Wanless and crunch a single WU for it under the VM. That should force the two rogue CPID's on the VM to get corrected.

Locally, each host (or guest in this case) keeps track of the Oldest CPID it's seen for your email. If it ever gets joined to a project that reports it has an older CPID for you, that is what forces the right CPID to get replicated to the messed up projects.

[Bryan]

Best thing to do is find a project you crunched *before* MindModeling and Wanless and crunch a single WU for it under the VM. That should force the two rogue CPID's on the VM to get corrected.

Locally, each host (or guest in this case) keeps track of the Oldest CPID it's seen for your email. If it ever gets joined to a project that reports it has an older CPID for you, that is what forces the right CPID to get replicated to the messed up projects.

That's interesting, I didn't know that is how it works!

I guess I could crunch a Seti WU .... I haven't done anything there for years

[Bryan]

I ran 2 Seti WU so I'll see if Seti disappears from my FreeDC stats tomorrow. If it does then I'll have to think long and hard whether to run another WU to get it back

I'm expecting a personal thank you from DA for the help I gave them tonight

[Fire$torm]

I'm expecting a personal thank you from DA for the help I gave them tonight