First, you don't use NFS, you use a SQL database server. Without some
additional work, though, this still creates a single point of failure. See a
thread currently in progress on this list called "advice requested on high
availability configuration" for some more elaboration.
Mike Nerone <mailto:[EMAIL PROTECTED]>
Network Operations Manager
Internet Direct, Inc.
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On
> Behalf Of Dave Kitabjian
> Sent: Friday, 07 April 2000 1050
> To: '[EMAIL PROTECTED]'
> Subject: (RADIATOR) Coordinating multiple radiator instances
>
>
> We will be bringing a couple more Radiators online to work in parallel
> to our main one. I could use some guidance here:
>
> 1) logfile - Since they all write atomically, could they all share the
> same logfile if it resides on a common NFS volume? Is there a non-NFS
> way to do this?
>
> 2) SessionDatabase - To have them all share the same SessionDatabase
> (which is the only useful scenario, right?), the DB would also have to
> sit on something in common like NFS? Other options?
>
> I believe my boss will not want the NFS option because it will create a
> single point-of-failure, defeating much of our purpose behind having
> multiple servers. Any comments?
===
Archive at http://www.starport.net/~radiator/
Announcements on [EMAIL PROTECTED]
To unsubscribe, email '[EMAIL PROTECTED]' with
'unsubscribe radiator' in the body of the message.
