|
Dear All,
I have configured 128-node OSCAR in my company.
Since all nodes are kept in the same room some of the nodes fail often. I have
10 nodes dedicated to failure recovery. I have two queries in this
situation.
1. Is there any opensource cluster management
software that keeps track of the hardware details the systems in the cluster so
that when one node fails the other node (dedicated for failure recovery)
gets waked up automatically shutting down the failed node.
2. Can heart-beat connection be established in
OSCAR itself or HA-OSCAR is required. I want to have 10 nodes dedicated for
failure recovery. These 10 nodes do not form part of the normal operation of the
cluster. When nodes fails these nodes takes control of the failed node shutting
down the later and joins the cluster through cluster management
software.
Regards,
Rajiv
|
- [Oscar-devel] OSCAR-Failure Recovery through cluster mana... Rajiv
