This content has been marked as final. Show 22 replies
I have to be honest here: I am not going to help you further beyond recommending that you open an SR with Oracle Support.
1. You obviously have not read the docs which makes you very dangerous to both your database and your employer.
2. You have provided zero (not a small amount but ZERO) information on your environment. We do not know hardware, operating system, version, edition, error message, log file contents, nothing.
3. You seem to assume you can bluff your way through this even now though you have no experience, no RAC education, and have invested no time reading the docs, based on some total stranger giving you advice in an internet forum.
So why am I not willing to help you further? Because when you destroy it ... and it seems likely you will ... my fingerprints will be on your mess and that is something I am unwilling to help you accomplish.
So, again, you need to open an SR with Oracle Support.
And for other forum participants ... please do not interpret this as me getting back at Support for all of the times they've frustrated me over the years. It may be true but I just don't want you to interpret it that way. ;-)
this is my issue i want to ask you if we have 2 node and one node is running and other node is down actually the serverHis concern is slightly different.
not accessible if there is any affect on node 1 if i restart the server belong to node 2 ???that's it .this is all the story
Let me try: If I stop a node the other node will assume the connections and keep my environment running?
Q1: Did you have sure which this reboot will "fix" the problem or it's a guess?
Q2: How is the health of this host? (e.g Is swapping)
Q3: Why your SSH is not working? (maybe you need only restart the service, instead whole host)
Q4: How was configured your Database Service ? Did you can relocate service from server to another?
Q5: Your Client is configured to support this configuration? (i.e using VIP)
Q6: There is any problem on your clusterware which can cause split brain on reboot on other node? (i.e see alert_log of cluster)
Please don't try answer these question... I'm just pointing why that's not simple.
In an environment configured properly to support failover, the common is Clusterware maintain high availability of the application even if a node comes down.
But to make this happen dozens of factors should be cooperating for the proper functioning of this technology.
It is not simply push a button and everything will work.
Just follow recommendation of Damorgan:
My recommendation is that you go to your management and have them bring in a consultant for 30-60 days to work with you and teach you what you need to know.
Your company paid Oracle for a support contract ... you refuse to use it.
Your management wants stable operational systems ... you refuse to bring important information to their attention.
It would be interesting to hear your explanation for why you refuse to do what is clearly in your organization's best interest.