What Replicate task happened, if one of nodes downon RAC & ASM environment ?

[khchoy]

Question is only focus on two case. one is manual down such as maintenance , other is database crash.

As you know customer application is running without any problem. (I konw other CDC solution is no pproblem also)

Specially , maintenance is important because every customer​ stop one node for maintenance. I want to know that task is no problem or waiting until all nodes are running.

I created case but I've got answer that they need log files. But this is question so I have no logs.

Questions are ;

 Connection string is already configured fail-over (used tnsnames.ora)

1) If DBA stopped one of RAC nodes, Does replication task continued replication or waiting until startup stopped node ?

2) If one of RAC Node stopped caused by database or OS failure , Does Replicate task stop with error message or waiting with error message until node recovering ?

Regards,

Kwangho

[Madhavi_Konda · Accepted solution]

Hi,

For failover and load balancing you can use Oracle RAC SCAN name. The benefit of using the SCAN is that the connection information of the client does not need to change if you add or remove nodes in the cluster. Replicate has retry mechanism and will reconnect to the node after recoverable error of lost connection.

Thanks,

Madhavi

[khchoy]

Thank for answer. My question is focused on source endpoint. Actually, Replicate Task connected DB instance and ASM instance. Usally, one of nodes of source side  is stopped for maintenance. So Replicate task should to run without stopping or waiting. 

Is it same when source endpoint ?

Best Regards,

KwsngHo

[khchoy]

Hi,

I've got just short reply from support. Support team answerd that Replicate Task maybe not working .

I think this is related architecture. If Replicate designed that Replicate task check node down , Task can work.

I asked this  support team and community, but still not get clear answer.

Who knows answer this kind of question ?

Regards,

KwangHo