We are working on Oracle 11.2g two node RAC . At the time of clusterware installation we choose normal redundancy disk group for OCR.
Afetr installation we check there are 3 vote disks with three failure groups.Is all voting disks contains some information and all nodes send heart to one voting disk.
Please explain "why odd number of voting disk" .
As you know, normal redundancy diskgroup for OCR and VOTE disks require 3 way mirror. HIGH redundancy require 5 way redundancy.
Best practice needed to have an odd number of voting disks to ensure that a tie-breaking disk exists when the interconnect between nodes fails
Odd numbers of Voting Disks are recommended because losing 1/2 or more of all of your voting disks will cause nodes to get evicted from the cluster, or nodes to evict themselves out of the cluster.
You must have an odd number of voting files because you must guarantee that
an intersect remains in the voting files that each node is accessing when access to
one file is lost. If you have only two files and a node loses access to it, you cannot
guarantee which file the other node can access, if any. If you have three voting files
and lose access to one, you know that even if the other node loses access to one,
both nodes will be accessing at least one common file. So with three voting files,
you can tolerate loss of access to one file, but a loss of access to two files will mean
a cluster reconfiguration has to happen
Please check : Always ODD number of Voting disks.
Mahir M. Quluzade
Thanks a lot for relpy.
As i mention we are using 2 node rac and for vote disks normal redundency diskgroup.that means only one voting disk but with 3 copies is it right.
Another thing is the with 1 voting disks and 2 node rac if interconnect between node1 and node2 are fails and both the nodes are able to connect vote disk then what happen.
Thanks in advance.