All the NetMaster SmartTraces (TCP, UDP, ICMP...) are not returning packets for a specific TCPIP stack only. The same traces for other stacks work without problems
NetMaster for TCPIP 12.2
The IBM real-time application-controlled TCP/IP trace Network Management Interface (NMI) is disabled.
NetMaster SSI is using IBM NMI for Packet Analyser:
NS0201 P-FILE: PATRACENMI=YES
On the NMSSI joblog, the following recurring messages for TCPIPROC stack:
NFSU81 TRACE NMI SUBTASK INIT FOR STACK TCPIPROC FAILED, RC: R15=00000000 CODE: 8 SUBCODE: 0
NFSU83 TRACE NMI ERROR, EZBRCIFR FUNCTION: OPEN RETVAL: FFFFFFFF RETCODE: 00000486 RSNCODE: 78607365
Per IBM documentation page Common EZBRCIFR return and reason codes, these codes mean that NMI is disabled:
Another IBM documentation page (Disablement of the NMI) provides additional info about this NMI problem:
If certain error conditions occur, the NMI is disabled. Message EZD2014I is issued to the console indicating the cause of the disablement. The DISPLAY TCPIP,,TRACE command output also indicates that the NMI is disabled. If your application receives a return code of ETCPERR and a reason code of JrDisabled, this indicates that the NMI is disabled. If your application has an open trace instance, the application should invoke the RCCClose request to close the trace instance. The TCP/IP stack must be recycled to reset the disabled condition.
Being so, this is a problem on the TCPIPROC stack, and this stack should have issued the message EZD2014I providing info about the reason of the disablement:
EZD2014I Real-time application-controlled TCP/IP trace NMI has been disabled – reason code rsncode
rsncode
A code that explains the error. Possible values are:
Before recycling the TCPIP stack, it could be worth checking with IBM support if they can provide any suggestion or if they need some documentation to find the root cause.