> -----Original Message-----
> From: GRIDPP2: Deployment and support of SRM and local storage management
> [mailto:[log in to unmask]] On Behalf Of Govind Songara
>
> Same situation, we dont have enough space to drain.
> I think for re-install we have to mark node in disabled mode and that
> would required DT ??
>
Nah; just do it. The node will be offline and failing transfers
until you get it back up and working, but you should be looking
at ~20mins for an OS and DPM re-install, and if you do them one
at a time the rest of the storage and most of the jobs will still
be working.
It's worth draining a site for a head node upgrade, but for the
pool nodes I wouldn't bother putting the site in downtime. More
specifically, I _didn't_ put the site in downtime when I did
this at Oxford.
Ewan
|