This content has been marked as final. Show 1 reply
A couple of things. First, you are probably aware that a kill -9 causes the process to exit without executing any exit handlers, thus Tuxedo doesn't know the process terminated. It won't be until the next BB scan that the dead process is detected and restarted.
Now on to your TSAM question/issue, I don't think TSAM 11gR1 has any means of alerting for server failures. The closest would be to enable service monitoring and catch any failed requests and report those as alerts. In TSAM12c we have integrated TSAM with Oracle Enterprise Manager, and OEM would allow you to alert on a server failure as a server is modeled as a target in OEM.
Alternatively, depending upon what you want to happen when a server fails, you could use the Tuxedo event broker and subscribe to the system events associated with a server dying and being restarted, but that would be outside TSAM, yet still allow you to take some action based upon those events occurring.
Oracle Tuxedo Chief Architect