none
SNA 3270 PU's noty connecting to MainFrame after system move RRS feed

  • Question

  • Our supported customer uses 3270 LU's for connecting to z/OS VTAM PU's. BizTalk 2010 servers connect via Enterprise Extender to Mainframe

    We have 2 servers under 1 SNA Manager Domain (13 PU's on 1 server, 14 PU's on other server).

    On a 6 week rotation, the customer moves their primary network from one Lpar to another to allow for IPL of primary Network Node. Both Lpars are in the same NETID, so not possible to have the PU's active on both Lpar's at the same time. After 24 hours, the PU's are moved back to primary Network Node.

    During the mainframe network moves, the PU's and the EE Switched Majnodes are inactivated on one system, some other applications are shutdown, then after 3-4 minutes the EE Switched Majnodes and PU's are activated on the other system.

    We are finding that 1 or 2 of the PU's are not recovering in SNA Manager, and a Stop/Start has to be performed to recover. It is not the same PU's each time, seems to be random which PU's won't recover.

    I noticed in the Application Logs that there is PU error with SENSE=08060000 (which to me indicates PU already is use - almost like it did not fully get disconnect from previous system), but unfortunately the error mesage does not indicate the actual PU name. Errors stop once STOP/START performed on any impacted PU.

    Is there setting we should adjust to have the PU's attempt connection for longer period ?? Or possibly a suggested time period we should wait after PU's inactivated on one Mainframe system before we attempt to activate on other system in same NETID ?? 

      

    Thursday, November 8, 2012 4:43 PM

Answers

  • You can do the following to get traces, but these are usually analyzed by us as part of a support case. Doing this through a forum posting is very difficult as the traces can be large and hard to parse via cut and paste.

    - Run snatrace.exe
    - Click the Tracing Global Properties tab.
    - Add a zero to the Trace File Flip Length value.
    - Enable the "Allow HIS Administrators to perform tracing" option.
    - Click Apply.
    - Click the Trace Items tab.
    - Highlight SNAIP1 in the List of Trace
    - Click Properties
    - Click the Message Trace tab.
    - Enable the Data Link Control and Level 2 Messages option.
    - Click OK.
    - Minimize the SNA Trace Utility window.

    Reproduce the scenario. After the problem occurs, restore the SNA Trace Utility window and then click "Clear All Traces" to stop the traces.

    The trace files will be in the C:\Program Files\Microsoft Host Integration Server 2010\traces folder (by default).

    The traces are readable with the Trace Viewer utility under the Tools section of the HIS 2010 Start Menu program item. Youcan also double-click the trace files to open with the Trace Viewer.

    Thanks...


    Stephen Jackson - MSFT

    Monday, November 19, 2012 10:08 PM

All replies

  • The 08060000 sense code is defined by IBM as follows:

    0806 Resource Unknown: For example, the request contained a name or address not identifying a PU, LU, SSCP, link, or link station known to the receiver or the sender.

            0000 No specific code applies

    The problem shouldn't be that the PU is in use. It sounds like the PUs are correctly activated.

    When you say the recovery is to stop/start the PU, is this in VTAM or on the HIS Server?

    What do the VTAM logs show when this occurs?

    Thanks...


    Stephen Jackson - MSFT

    Friday, November 16, 2012 11:11 PM
  • Stephen,

    The stop/start of the PU is performed on the Server to recover. No action needs to be taken in VTAM

    There is nothing in the syslog to indicate an issue with the PU or LU's in VTAM. The VTAM Majnode for the PU shows Active after Activate command entered in VTAM, but the PU itself is in CONCT state in VTAM (if I remember correctly, the SLU and PLU for PU is DISABLED initially).

    The next rotation is not scheduled until Dec 15/16 weekend. I will verify the status in VTAM of any server PU's that do not recover after system move on Mainframe.

    Just Curious - are there any traces specific for PU connectivity I can activate in SNA Manager on the server for the 10 minutes that MF network is being moved ?? 

    Monday, November 19, 2012 9:58 PM
  • You can do the following to get traces, but these are usually analyzed by us as part of a support case. Doing this through a forum posting is very difficult as the traces can be large and hard to parse via cut and paste.

    - Run snatrace.exe
    - Click the Tracing Global Properties tab.
    - Add a zero to the Trace File Flip Length value.
    - Enable the "Allow HIS Administrators to perform tracing" option.
    - Click Apply.
    - Click the Trace Items tab.
    - Highlight SNAIP1 in the List of Trace
    - Click Properties
    - Click the Message Trace tab.
    - Enable the Data Link Control and Level 2 Messages option.
    - Click OK.
    - Minimize the SNA Trace Utility window.

    Reproduce the scenario. After the problem occurs, restore the SNA Trace Utility window and then click "Clear All Traces" to stop the traces.

    The trace files will be in the C:\Program Files\Microsoft Host Integration Server 2010\traces folder (by default).

    The traces are readable with the Trace Viewer utility under the Tools section of the HIS 2010 Start Menu program item. Youcan also double-click the trace files to open with the Trace Viewer.

    Thanks...


    Stephen Jackson - MSFT

    Monday, November 19, 2012 10:08 PM