I had recently attached several computers to this project, but it seems that 3 of them simply aren\'t up to the task of finishing a model (or, at least, anytime soon - after about a half month they all showed < 0.5%...), and, to top it all off, one of the more capable computers that I put it on had recently stopped working (read as: Hard Drive = Dead). If I posted which computers would either take till 2015 to finish one model, along with the computer that had a model working before it BSoD\'d and thereafter told me that no HD was found (I tried reseating the HD and all the troubleshooting steps, including attaching it to another computer), could an admin forcibly disassign the WU from those computers on the server side and reassign them to other computers?

Thanks in advance,
-Harrison N.

It isn\'t necessary for anything to be done server-side. If you move the BOINC installation from one computer to another, the server will figure out that a new computer is being used. You won\'t, however, be able to run two different BOINC installations on the same computer. If you already have some WUs being processed on your target computer, then you should finish up one lot of WUs, delete that installation, and start up the next.

There is a link to a thread on moving models towards the bottom of this README post.

By the way, bad luck with the PCs ...

What a lot of computers, Harrison! I expect it\'s the 3 at the bottom of your list that would almost run the models in real time. One of the 3 hasn\'t got enough memory for climate models anyway.

You could let these less powerful computers run APS@Home

http://www.apsathome.org/

because the WUs are pretty short and there don\'t seem to be any minimum computer specs at all.

When you move your models to newer machines, I expect they\'ll crunch much faster. Speeding models up a lot can potentially cause a problem that will crash the models before they complete. I think this is unlikely to happen to yours because they\'ve made so little progress on the slower computers. But just in case, you should make regular backups of the complete contents of the boinc folder. With a backup, this problem (and almost any other problem) can be solved and the model completed.

There\'s a selection of backup methods in the READMEs.

---

Cpdn news

Hi Harrison

This is the first time that I\'ve seen someone with a Transmeta Crusoe powered computer.

If at any time you want to \'reassign\' a model, the only person who can do this in a hurry is you.
You do this by Aborting the model(s) concerned. When the server gets this message, it will flag the model for possible reassignment, although it doesn\'t necessarily get reissued.
If you just let a model \'die\' without informing the server, it will be flagged as \'lost\' in about 6 weeks or so.

BEFORE you Abort models, it is VERY important that you go to the Project tab and click on the project name. (cpdn)
Then click on the \"No new tasks\" button.
This will prevent the download of a new model immediately after you abort the current model.

Also do this if you want to Suspend a model in the Tasks tab! Otherwise, if it\'s the only thing there, you\'ll get a new model. The only way to suspend a model, is to also suspend BOINC, and you do this by using Suspend in the menu.

---

Backups: Here