Tuxedo cleanupsrv Server Is Killing The Wrong Process (Doc ID 2202717.1)

Last updated on NOVEMBER 14, 2016

Applies to:

Oracle Tuxedo - Version 12.1.3 and later
Information in this document applies to any platform.

Symptoms

 An application server is booted:

044106.529.tuxmachine!myapp1.exe.12444.3410264448.0: 05-23-2016: Tuxedo Version 12.1.3.0.0, 64-bit
044106.529.tuxmachine!myapp1.exe.12444.3410264448.0: LIBTUX_CAT:262: INFO: Standard main starting

--> Another application is triggered which shuts down the above application by doing a kill on the processes:

123339.870.tuxmachine!TMSYSEVT.3609.2636923264.0: LIBTUX_CAT:1477: ERROR: .SysServerDied: myapp1.exe, group GROUP1, id 543 server died
123339.872.tuxmachine!TMSYSEVT.3609.2636923264.0: LIBTUX_CAT:1475: ERROR: .SysServerCleaning: myapp1.exe, group GROUP1, id 543 server cleaning

-> Some other application server is started and makes use of the same process id# (or another process is started and a thread of it, in a mult-threaded server, uses the same pid#)

123344.598.tuxmachine!myapp2.exe.12444.985510304.0: 05-23-2016: Tuxedo Version 12.1.3.0.0, 64-bit
123344.598.tuxmachine!myapp2exe.12444.985510304.0: LIBTUX_CAT:262: INFO: Standard main starting

-> The bulletin board does not know that the myapp1 pid# has been reused and calls cleanupsrv to kill myapp1 but myapp2 is now using the pid# and gets killed

123346.994.tuxmachine!cleanupsrv.12103.3030925696.-2: CMDTUX_CAT:538: INFO: Process 12444 still running - forcing termination (via SIGIOT)

 The OS maxpid kernel value is low and increases the cycle rate of re-use of the process ID (PID) values.

Changes

 

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms