Clarification of Policy Agent (PAGENT) debug messages

Problem

During TCPIP address space termination, the following messages are seen in the Policy Agent (PAGENT) log file: SYSERR :008: remote_listen_thread: accept() failed for remote PAPI connection, errno = EDC5122I Input/output error. LOG :008: close_or_recycle_remote_listen: EZD1578I PAGENT IS UNABLE TO PROCESS REQUESTS FROM SERVICES REQUESTORS SYSERR :008: remote_listen_thread: no active listen socket, exiting EVENT :008: terminate_thread_pool: timer id 19380D20 is canceled, num nodes = 136 EVENT :009: pool_thread_driver: Thread cleanup completed EVENT :010: pool_thread_driver: Thread cleanup completed EVENT :012: pool_thread_driver: Thread cleanup completed EVENT :007: pool_thread_driver: Thread cleanup completed EVENT :011: pool_thread_driver: Thread cleanup completed EVENT :014: pool_thread_driver: Thread cleanup completed EVENT :013: pool_thread_driver: Thread cleanup completed EVENT :016: pool_thread_driver: Thread cleanup completed EVENT :015: pool_thread_driver: Thread cleanup completed EVENT :020: pool_thread_driver: Thread cleanup completed EVENT :008: remote_listen_thread: Thread cleanup completed This set of messages is normal and expected when TCPIP is terminated while PAGENT is still running. The EDC5122I Input/output error message indicates that a transport provider (TCP/IP) is no longer available, which caused the PAGENT accept() function to fail. The follow-on messages are related to PAGENT cleanup processing for threads that were created. Prior to all of the above messages, the following series of messages was occurring every 10 minutes. EVENT :000: contract_thread_pool: thread pool already at minimum (10) - can't contract EVENT :000: thread_pool_maintenance: timer id 19380D20 is set for 10 minutes, queued = 2, num nodes = 137 These again are normal messages that are cut by default by PAGENT. The meaning of these messages follows:[

] PAGENT uses a thread pool for the centralized policy function, and for Configuration Assistant connections. The pool is dynamic, growing and shrinking as the need arises. However, there are minimum and maximum sizes of the pool - the first message indicates that the pool is already at the minimum of 10; the second message is setting a 10-minute timer to re-check the statistics and expand/contract the pool if needed.