On Wednesday, June 26, 2013 08:35:07 AM Per Granath wrote:
> Did you try it with this configuration?
>
> chassis {
> redundancy {
> failover {
> on-loss-of-keepalives;
> on-disk-failure;I've always wondered whether it is useful for a failover to occur due to disk failure. If a disk in the primary RE has failed, and chances are you have the same one in the backup RE, the likelihood of imminent failure is quite high due to shared fate. I know some folk could decide to have different hardware revisions so that they reduce their chances of shared-fate failures, but this is too much work and might cause other unrelated problems due to different hardware revisions of each RE. If the RE can keep chugging along even with disk failure, but provide reasonable warning to the operator who can then decide to institute manual failover in to replace the bad disk or whole RE, that would be ideal. But then again, I suppose NOT configuring 'on-disk-failure' achieves the same thing. Perhaps this was more meaningful when disks were less reliable. I haven't had a disk fail on me on an RE before, but my corner of the world is quite small. Mark.
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ juniper-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/juniper-nsp
