Bad slaves

ROOT
1

Hi,

It seems that I’m having problems connecting to the slaves. My proof.conf looks like

node prod-frontend.hepgrid.uerj.br image=nfs

slave prod-frontend.hepgrid.uerj.br perf=100 image=nfs
slave prod-frontend.hepgrid.uerj.br perf=100 image=nfs
slave compute-0-0.hepgrid.uerj.br perf=100 image=nfs
slave compute-0-0.hepgrid.uerj.br perf=100 image=nfs

user cms on prod-frontend.hepgrid.uerj.br

I always get 2 active slaves (from the master) and things work fine, but the others are given as bad slaves. I also get no error messages.

Thanks,

Antonio

2

With which version do you have these problems?

Cheers, Fons.

3

Hi Antonio,

could you please start the PROOF daemons with

proofd -d 3 -f …

and also the ROOT session

in .rootrc :

Root.Debug : 4

in the debug mode and send the output ?

Cheers,
Carsten.

4

Ok, I can see the slaves now.

I had the wrong naming convention here in the cluster. That’'s why I couldn’t see them - they didn’t exist! :slight_smile:

Thanks anyway and sorry about the mess.

Cheers,

Antonio