On Wed, 2007-07-11 at 13:42 +0100, HAWKER, Dan (external) wrote:
Not sure if this is strictly OT, however I'm having some trouble
configuring my Linux clients to failover to secondary FDS boxes quickly
enough.
Dan,
We had the same issue. It seems like nss_ldap fails over after a
reasonable timeout, but applications like autofs will take much longer.
Also, if you have two servers and all of your clients are configured to
talk to them in the same order (if server1 is down use server2), then
server2 would be idle until server1 is down.
We use Piranha to get around this. This way we can count on both
servers being equally utilized and clients fail over seamlessly (so
seamless they don't even realize they've failed over).
I don't know if this is the "prescribed" way of handling it, but it
works well for us.
-Steve