/var/adm/messages is loaded with lots of Assignment to constant runpaths / runrlimit_core / runrlimit_cpu etc. (4375856)

/var/adm/messages is loaded with lots of Assignment to constant runpaths / runrlimit_core / runrlimit_cpu etc.

After upgrading to 7.2.3.3, the following errors are flooding up in /var/adm/messages.

Jan 23 12:36:31 itgcubsprod auth|security:err|error pmlogxfer7.2.3.3[32244074]: 1101 Assignment to constant runpaths
Jan 23 12:36:31 itgcubsprod auth|security:err|error pmlogxfer7.2.3.3[32244074]: 1101 Assignment to constant runrlimit_as
Jan 23 12:36:31 itgcubsprod auth|security:err|error pmlogxfer7.2.3.3[32244074]: 1101 Assignment to constant runrlimit_core
Jan 23 12:36:31 itgcubsprod auth|security:err|error pmlogxfer7.2.3.3[32244074]: 1101 Assignment to constant runrlimit_cpu
Jan 23 12:36:31 itgcubsprod auth|security:err|error pmlogxfer7.2.3.3[32244074]: 1101 Assignment to constant runrlimit_data
Jan 23 12:36:31 itgcubsprod auth|security:err|error pmlogxfer7.2.3.3[32244074]: 1101 Assignment to constant runrlimit_fsize

The pmlogxfer transfers the offline event and io logs to a policy server after an offline session has occurred

These particular messages may be safely ignored.

Unfortunately, there is no way to disable the pmlogxfer syslog messages. The cause of the issue has been identified and his has been identified as a product defect. 

The Enhancement Request no is: 444510

This has been fixed and will be released with the next version 7.3.

To limit the error messages, the following can be done.The pmlogxfer will only run after an offline session to transfer the offline log files to a valid policy server. There are a couple of things that could be done to reduce the number of offline sessions that occur:

1) Increase the offlinetimeout setting in the pm.settings file. For example, "offlinetimeout 3000" would increase the timeout from 1.5 seconds to 3 seconds.

2) Increase the number of policy servers in their environment. This will increase the load-sharing capacity of the environment and reduce the chances that any one policy server gets overloaded (and incapable of responding within the offlinetimeout time).