IBM Support

PH70620: NETWORK EXPRESS (OSH) ARP REQUESTS ARE FAILING EQENET/EQDIO ADDARPCACHE FAILURE 2019

Subscribe to this APAR

By subscribing, you receive periodic emails alerting you to the status of the APAR, along with a link to the fix after it becomes available. You can track this item individually or track all items by product.

Notify me when this APAR changes.

Notify me when an APAR for this component changes.

 

APAR status

  • Closed as program error.

Error description

  • OSH ARP requests are failing with a hardware error code 0x2019
    (seen in systcpip exception trace records). The ARP Requests are
    being rejected by OSA firmware because the source and target IP
    addresses are not on the same subnet. In this case a static VIPA
    was used as the source address in the ARP request which was in a
    different subnet then the destination IP address. This will have
    different effects on IP packet delivery. For example TCP traffic
    will be re-transmitted, and hopefully delivered to the
    destination using a different route over a different interface.
    PINGs will timeout, and UDP packets will not delivered
    

Local fix

  • none
    
    additional symptoms:
    
    systcpip exception record
    XXX     VTAM      1001022A  10:59:28.348110  !Unexpected EqeNet
    Set Assist Pa
    HASID..00CC      PASID..00CC      SASID..00CC      USER...TCPIP
    TCB....00000000  MODID..EZBIFIND  REG14..41372B42
    DUCB...2900240D
    CID....00000000  PORT...0         CPUA...12
    IPADDR. 0.0.0.0
    ADDR...00000000  00000000_7F358068  LEN....000010  Device Name
     +0000  D6E2C1C9  C6F1F040  40404040  40404040   | OSAXYZ|
    ADDR...00000000  00000050_002D6FB8  LEN....000044  Inbound
    EqeNet Ipa Control Rec
     +0000  B3000024  20190100  13040000  0000007B
            00000000  00000001  002C0004  00000101
     +0020  00030000  00000000  5A49424F  43F10000
            srcip     dstip     00000000  00000000
     +0040  FFFFFF00 (subnet Mask)
    
    ADDR...00000000  00000050_002D6FB8  LEN....000001  EQENET Ipa
    Command    +0000  B3
    
    ADDR...00000000  00000050_002D6FBC  LEN....000002  EqeNet
    SetAssist Parms Return C
             +0000  2019
    
    netstat arp shows zero hardware/mac addresses:
    
    INTERFACE: OSAXYZ             ETHERNET: xxxxxxxxxxF1
    QUERYING ARP CACHE FOR ADDRESS y.y.y.60
    INTERFACE: OSAXYZ             ETHERNET: 000000000000
    QUERYING ARP CACHE FOR ADDRESS y.y.y.90
    INTERFACE: OSAXYZ             ETHERNET: 000000000000
    QUERYING ARP CACHE FOR ADDRESS y.y.y.1
    INTERFACE: OSAXYZ             ETHERNET: xxxxxxxxxx87
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All users of the IBM Communications Server   *
    *                 for z/OS 2.5, 3.1 and 3.2                    *
    *                 Network Express (OSH)                        *
    ****************************************************************
    * PROBLEM DESCRIPTION: Outbound IPSEC encrypted packets being  *
    *                      transmitted over EQENET interfaces are  *
    *                      being discarded by TCP/IP due to ARP    *
    *                      resolution failures.                    *
    *                                                              *
    *                      E9175/K                                 *
    ****************************************************************
    * RECOMMENDATION: Apply PTF                                    *
    ****************************************************************
    The problem is summarized as follows:
    1) Failure occurs when outbound packet initiating ARP resolution
       to target IP address is using IPSEC encryption.
    2) The IPSEC code path led to the IP assist ARP request being
       invalidly built by TCP/IP.
    3) The source IP address is being set to a source VIPA address
       which is outside the destination IP address subnet.
    4) This leads to the ARP IP assist being rejected by OSA
       firmware.
    5) This causes all packets waiting for ARP resolution to this
       next hop IP address to be discarded causing packet loss,
       delays and potential retransmissions.
    
    Notes: For non-IPSEC packets which initiate ARP processing, the
           ARP cache is correctly resolved and no problems are
           encountered.
           Environments with heavy IPSEC usage increases the
           probability of repeatedly hitting this problem which
           may cause loss of connectivity.
    

Problem conclusion

  • ezbifarp.plx - Has been corrected to always set the source IP
    address for outbound ARP requests to the home IP address of
    the EQENET (OSH) interface.
    

Temporary fix

Comments

APAR Information

  • APAR number

    PH70620

  • Reported component name

    TCP/IP MVS

  • Reported component ID

    5655HAL00

  • Reported release

    310

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    YesHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2026-03-20

  • Closed date

    2026-04-02

  • Last modified date

    2026-04-03

  • APAR is sysrouted FROM one or more of the following:

  • APAR is sysrouted TO one or more of the following:

    UO07394 UO07395 UO07396

Modules/Macros

  • EZBIFARP
    

Fix information

  • Fixed component name

    TCP/IP MVS

  • Fixed component ID

    5655HAL00

Applicable component levels

[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSSN3L","label":"z\/OS Communications Server"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"310","Line of Business":{"code":"LOB35","label":"Mainframe SW"}}]

Document Information

Modified date:
03 April 2026