Rac Issue for EBS r12.2.4

[Samia]

Hi Erman,

How are you ? I hope fine .

I need some advise for an instance of a cluster database (RAC).
Only one node is working fine while the other when we start it nothing good with EBS ...

There are many other databses running well ... so may be the RAC is not the problem !! I don't know.

What could be root cause ? where shull we look ?

the node which causing issue show many events gc_freelist :
running: select * from dba_users can make more 10min (our DBA said, I didn't check that because it's stopeed now).


Thank you in advance for your help.

Regards,
Latifa

[ErmanArslansOracleBlog]

Hi Latifa,

Thanks. I hope you are fine too..
This issue is generic.

Ensure you configured RAC properly.. (software + network -- interconnect, rac-related db parameters and etc..)

What is the wait event of that session (select * from dba_users) ?

[Samia]

Hi Erman,

Thank you for your reply.

Our DBA said that configuration is identical between the two nodes nothing different.

for session select * from dba_users, the system try to liberate locks and never succeed. There are two many contention in the cache gc_freelist as he said.

it's very starenge because the other node of the RAC works fine !!

what can we can we check (cached parameters, system, ...) !

eventhough we clone the instance the problem will remain and if it solved by it so it more strenge !!


Regards,
Latifa

[ErmanArslansOracleBlog]

This issue is generic.
You just tell it with words, and it seems you are a middle-man in this scenario.. (I mean you don't have access to the database in the required level)

Run ORAchk or AHF and review the output.

[Samia]

Hi Erman,

What do you think about this Doc ID : 2734227.1

Our DBA think that it's the source of the issue, he might find error messages before that lead to this note.

Actually we cloned the environment and it's working fine.

Regards,
Latifa

[ErmanArslansOracleBlog]

The cause given in that node is specific. If you have similar symptoms , then okay, go ahead.
But you didn't send me anything related with the issue given in that node.. So..

Ref.
AUTO STATS Job Fails with ORA-30926: unable to get a stable set of rows in the source tables (Doc ID 2734227.1)