This content has been marked as final. Show 4 replies
Mostly likely you have a network setting somewhere that disconnects after some interval of quiet communication. You can add a heartbeat table (go to MOS and query GoldenGate Heartbeat) or just add some dummy table of your own to replication and have it updated every X interval that is less than your timeout. Typically in production situations there's enough data coming through where this isn't an issue so you're probably running this on a test system.
You can also set AUTORESTART in your manager param file.
But either way, 10054 is a standard TCP/IP error and is technically outside the purview of OGG.
Hmm... I just caught that this is one system. Then you don't need to use RMTHOST. You don't even need to have OGG installed twice. Just use one extract, write to a local trail, have the local replicat read from that local trail and apply SQL changes.
best thing is to look at tcperr.log .. which will be somewhere in your GG home ... find it :)
maybe your firewall is blocking access? You could use telnet to see if you can connect remotley, to the port that should be enabled.
telnet 10.1.1.1 4056 -ladmin
where 4056 is the port number
On the server where you have the issue, you can also run the following dos command to see if the server is listening on the port, you expect to be using:
look at the State column where State is LISTENING
Check out ggserr.log on remote when you start replication...
Possibly dynamic range is not wide to give a connection to you or network issue...
Please let me know what you see in remote. ...