Here is the scenairo:
I got two PCs, one is Linux(RH4), which is installed a Java Oracle related application, and the other is Windows(2000), where the Oracle database stayed.
I want response quickly when the network between those PCs is suddenly down. So i notice that linux have KeepAlive attribute which can solve this.
I set the following KeepAlive attributes:
and also i add the "ENABLE=BROKEN" property to JDBC driverUrl to enable application to support TCP KeepAlive of OS.
but the result is not what i expected. The application just hangs and didn't report exception until around 15 mins after i cut the network.
What's yours idea? is KeepAlive set correctly? or how do i check if the KeepAlive package is sent?
I got two PCs, one is Linux(RH4), which is installed a Java Oracle related application, and the other is Windows(2000), where the Oracle database stayed.
I want response quickly when the network between those PCs is suddenly down. So i notice that linux have KeepAlive attribute which can solve this.
I set the following KeepAlive attributes:
and also i add the "ENABLE=BROKEN" property to JDBC driverUrl to enable application to support TCP KeepAlive of OS.
but the result is not what i expected. The application just hangs and didn't report exception until around 15 mins after i cut the network.
What's yours idea? is KeepAlive set correctly? or how do i check if the KeepAlive package is sent?