Hi Colin,
Have you upgraded to dCache 1.7 yet. This used to bug me with 1.6.6 but
they seem to have fixed it in 1.7.
Yours,
Chris.
> -----Original Message-----
> From: Testbed Support for GridPP member institutes
> [mailto:[log in to unmask]] On Behalf Of Colin Morey
> Sent: 22 November 2006 13:40
> To: [log in to unmask]
> Subject: Re: fabric management: cfengine
>
> On Wed, 2006-11-22 at 13:26 +0000, Greig A Cowan wrote:
> > > next on the list is getting cfengine to restart
> dcache-pools and the
> > > like (dcache really doesn't like it if the controlling terminal
> > > disappears.)
> >
> > Can you explain some more? What are you trying to do?
> >
> run "/etc/init.d/dcache-pool restart".
>
> if we ssh into a pool node and run that, we can't then close
> the ssh connection as there's some filehandles that are kept
> open by dcache, if we force Stderr to close with "2<&-" the
> pool dies when the connection to the machine is terminated.
>
> using such tricks as "screen -d -m /etc/init.d/dcache-pool
> restart" are a bit hit and miss.
>
> of course, I may have something mis-configured that's biting
> me on this one.
>
>
> Cheers,
>
> Colin Morey,
> Northgrid Systems Manager,
> Manchester Tier2 Centre
>
|