OPS/MVS - MSF connections between LPARS

Products

• OPS/MVS Event Management & Automation

Issue/Introduction

We just upgraded to 13.5 last weekend (from 13.0).  

We use VTAM for our MSF connections between LPARS.  We find that when we IPL, many times there are usually just 2 of our 5 LPARS that dont connect toe each others successfully with  MSF.   We end up bouncing MSF on one of those LPARS and then all recovers.

Have you seen this?  Any idea what we can look at.  Not easy to recreate - we dont IPL often.  This has been going on for years - finally getting around to ask about it.  Any other suggestions for recovery rather than bouncing MSF?  A DEACTIVATE/ACTIVATE of the LAPRS in MSF dont seem to recover it.

Environment

Release : 13.5

Component : OPS/MVS

Resolution

The recommendation is to delay the MSFINIT execution until VTAM completes initialization.  

    01:10:47 OPS3520O ACB OPEN FAILED FOR VTAM APPL OPSMVSC, RC=8, ERROR=92

    01:11:14 IST020I VTAM INITIALIZATION COMPLETE FOR CSV2R3

The above messages indicated that VTAM completed initialization after the first attempt was made to allocate the APPL. This ERROR=92 indicates VTAM was not active as seen in the time stamps.

 

Another recommendation is to create a message rule to intercept the message IST020I and then initialize MSF.  In other words, execute the MSF START and ACTIVATE MSFID(ALL) from a rule that intercepts the IST020I VTAM INITIALIZATION COMPLETE message.