Understanding Node Eviction - RAC 10

In this video you will understand what is Node Eviction and why it happens..

Our Upcoming Online Course Schedule is available in the url below

[vidmayur]

excellent explaination, very important concept in RAC . Even today I revisit it sometimes when I have doubts .

[WorkRakuten-mcjp]

I am great fan of you. Clear expansion.

[yashwanth]

Sir.. I am a fan of urs from now :) .... You have a great way of identifying and explaining tricky situations in Oracle database.. Every video I watch, I come away surprised at how many things I do not know and how easily you help me understand... Thanks a lot Sir...

[valiantvimal]

Thanks a lot sir for all your videos on several concepts of Oracle and Infrastructure☺️

[balasahebshetkar]

@2:29 to 3:17 - Sir I have doubt in example you provided .... In clusterware each node must have access on more than 1/2 of VD to survive ... if we consider this ... then in your example no node would have evicted because first node have access on 02 VD and 2nd on 03 VD ...

[rajag]

It is good explanantion. Can you please explain the RAC starup sequence

[TheManubarcafan]

I have a cluster of two nodes, after power failure one of the nodes can be up in the cluster but the second does not, both nodes are pinging each other, and also tracecroute successfully, the logs say no network hearbeat...but the disk heart beat is ok

[kunal]

STONITH is "Shoot the other node in the head", that means sick node doesn't reboot by itself instead it is rebooted by healthy node.

[karanchauhan]

I am a bit confused as to why we have odd number of voting disks, Say if we have 4 voting disc even then if one of the voting disc is not available still the cluster can have access to 3 voting disk meaning more than 1/2 so why can't we have 4 voting disc(which is an even number)

[vijaybhaskarful]

How to fix this node evictions ?
Can you give me example

[up]

I have same problem my cluster is rebooting when I check my crs log it show network communication error

[btgrk]

Why would a node get evicted if it could still be able to access 2 of 3 VDs even though peer node accessing all 3 VDs?

[sanedba]

thanks alot sir ..._/\_ would please xplain about relocate resource in cluster where it gets updated how the processes get to knw the failover of resource to existing node ???