Clients all died again

[PCZ]

ALL
I thought the CLI had been fixed but it hasn't
When I woke this morning I found that they all had the 1K work packet problem

I like the Science behind this project and wanted to contribute but the current state
of the client is terrible and running it would test the patience of a saint.

Maybe it will be fixed one day and I can make a return.
Sorry to those folks who answered the call to support this project but I had no idea the client was this bad.

[IronBits]

Well that's the pits! :(

[PCZ]

ALL
I am going to give it another go this weekend.
Monday I will return most of the boxes to DF.

We have to catch US Distributed !!

[rshepard]

Have you tried the "fix" for this problem discussed in this thread:
http://forums.teamphoenixrising.net/showthread.php?threadid=18808

I've been using it on 9 clients for a couple of days, and only had one crash (on an unrelated error)

[PCZ]

Yes I am going to be using a variation of the "Fix".

The first nodes are already running but it is taking time to write the batch files and set up the task scheduler.
I also use instsrv/srvany to run LF as a service and my batch file does a net stop and net start, hopefully it will bring the dead ones back to life.

[PCZ]

All the clients made it through the night !!

The 1k Job packet bug didn't rear it's ugly head so I don't no how succesful the fix might be.

[Darkness Productions]

By the way OCZ, welcome back. I was beginning to think you were off having fun somewhere else. Also, how much GHz do you have on Lifemapper right now?

[PCZ]

The problems with the CLI really annoyed me so I took a break from it.
about 50ghz at a guess

BTW
I am only running the client for the weekend as a trial.

I added em up 45ghz consisting of a mixture of Athlons and P4's
The Athlons perform much better clock for clock than P4's on this project.

[Darkness Productions]

Jebus, 50GHz? /me cries.

It'll only be a matter of time....