IBM i Global

IBM i 

A space for professionals working with IBM’s integrated OS for Power systems to exchange ideas, ask questions, and share expertise on topics like RPG and COBOL development, application modernization, open source integration, system administration, and business continuity.


#Power


#IBMi
 View Only

  • 1.  HPRip 7.4 vs 7.3 remote location

    Posted Fri September 20, 2024 07:27 PM

    Hi, I have a partition on 740 with HPR and in another location there is another with 730 with HPR (before they connected via anynet), they connect via HPRIP controlers but I have intermittent connections, thinking it could be some configuration between both operating systems, what could I check? In HPR is there a concept of collision like in TCPIP?
    Regards



    ------------------------------
    Miguel Peralta
    ------------------------------


  • 2.  RE: HPRip 7.4 vs 7.3 remote location

    Posted Fri September 20, 2024 09:18 PM
    Edited by Satid S Fri September 20, 2024 09:27 PM

    Dear Miguel

    It would be good if you could provide details on those "intermittent connections" incidents.  When it happens, does it cause you to have to vary off and then vary on the relevant controller descriptions to resume operation?   When it happens, is there any relevant message appearing in QSYSOPR message queue?

    Do the two IBM i servers connect to the same local LAN or are they in different physical locations that connect through a remote link?   If the latter, I used to experience similar problem that was caused by the fact that the remote IP router supported limited size of what is call MTU size in TCP/IP.  Please read the attached article on how to identify the proper MTU size so that it does not exceed what the IP router supported.  Basically, you ping from one location to another location and adjust the max packet size of the ping until fragmentation disappears and use that max packet size to set in IBM i TCP/IP routing entry's MTU with CHGTCPRTE command.   You may want to create a new routing entry with the proper MTU size for the specific target IP address used in HPRIP in both LPAR/servers.

    If this is not the case with you, you should start IBM i communication trace of a size about 1-2GB (and limit the captured trace by specifying the 2 IBM i servers' IP addresses involved in this incident) and let it run even when it's full (cyclic mode) and stop it when the problem occurs.  Then you dump the trace and use Wireshark to look at what happens at the moment of the incident.



    ------------------------------
    Satid S
    ------------------------------

    Original Message


  • 3.  RE: HPRip 7.4 vs 7.3 remote location

    Posted Fri September 20, 2024 09:22 PM
      |   view attached

    Sorry, I forgot to attach the article.  Here it is.



    ------------------------------
    Satid S
    ------------------------------

    Attachment(s)

    pdf
    TCPIP-MTU Size and Fragmentation.pdf   324 KB 1 version
    Original Message


Related Content