A fix is available
Closed as program error.
An ASSERT is seen in the sna.err log: 2012-03-11 08:04:01 EDT 4096-19(0-10) E (example.com) ASSERT: File name = ../../p/sappn/nrtpapi.c Line number = 144 Expression = !NB_MEMCMP(thdr->tcid, LOCAL.my_tcid, S_TCID) Stack: nrt_rtp_receive+278(3292F020,33993048,33993048,1) nrt_receive_proc+1A0(3292F020,1,33993048,0) nba_dispatch_input+294(32941720,33993048,F003F580,49B2F20) nba_dispatch_process+CC(32941720,33993048,F003F620,1) nba_schedule_process+154(32941720,2F580C50,200034E,21A) nba_send_ips+308(30001B3,1,3292F020,0) nhp_rtp_nlp+DF0(3292F020,2F580C50,32924820,2F49B048) nhp_queue_handler+5DC(3292F020,2000,2F580C50,0) nba_dispatch_input+294
The problem occurs when the following all occur: If a node is restarted quickly and low numbered RTP pipes are in use at the point that the node is stopped (eg @R000004) then a VRN partner will not detect the failure and when it polls after the VRN connection is re-established CS/AIX can mix up the old data with data from a new partner. This leads to the ASSERT (which detects the randomised TCID that CS/AIX generates).
Code has been added to ignore the old data.
Reported component name
CS FOR LINUX
Reported component ID
Last modified date
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
Fixed component name
CS FOR LINUX
Fixed component ID
Applicable component levels