A fix is available
APAR status
Closed as program error.
Error description
Sending network packets to a virtual NIC during LOGOFF of the destination user resulted in ABEND HTT001 in HCPVNSDG. Both users were coupled to the same guest LAN using IP Layer transport. The failure occurred after upgrading a working Disaster Recovery system to z/VM CP Release 7.1.0 with the latest support.
Local fix
N/A
Problem summary
**************************************************************** * USERS AFFECTED: VSWITCH or Guest LAN using IP Layer * * transport (IP vs ETHERNET option) * * may be affected. * **************************************************************** * PROBLEM DESCRIPTION: * **************************************************************** * RECOMMENDATION: APPLY PTF * **************************************************************** HCPVNSDG uses the results of a prior IP Table search when the sender is using the same IP destination that it used before. Before making that decision to use the cached NIDBK pointer, HCPVNSDG tests for NIDDXVER==LANDXVER (was the network changed after this search was made?). In the failing case, the destination guest was logging off and LANDXVER was incremented to indicate that the network was changing (the Virtual NIC is being removed and will soon be destroyed). However, the sender was sending another packet through HCPVNSDG after LANDXVER was updated and before the IP address (IPNBK) was removed from the LAN IP Table. In this case, the page containing that NIDBK was invalidated and the next reference to it led to the HTT001 ABEND. In other cases, the packets might just be discarded until the destination guest logs on (or some other guest connection is added or removed, forcing another LANDXVER update). It possible that a sender could see a persistent IP Address failure (instead of an ABEND HTT001) if HCPVNSDG gets the new LANDXVER after the IP address has been removed from the table (during LOGON processing for the target). In that case a NULL pointer is saved in cache for the target address. The result would be continued failures (even after the target interface is re-established) until either (a) the sender sends a packet to a different destination, or (b) some other action on the network increments LANDXVER (forcing a new lookup for the target IP address).
Problem conclusion
Functions in HCPIPN and HCPIPM that update the LAN IP Tables are enhanced to also increment LANDXVER. This forces HCPVNSDG to perform a new search the next time it is invoked for a given sender.
Temporary fix
********* * HIPER * ********* FOR RELEASE VM/ESACP/ESAR710 : PREREQ: VM66264 CO-REQ: NONE IF-REQ: NONE FOR RELEASE VM/ESA CP/ESA R720 : PREREQ: NONE CO-REQ: NONE IF-REQ: NONE
Comments
APAR Information
APAR number
VM66502
Reported component name
VM CP
Reported component ID
568411202
Reported release
710
Status
CLOSED PER
PE
NoPE
HIPER
YesHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2021-02-17
Closed date
2021-03-29
Last modified date
2021-09-30
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
UM35847 UM35848
Modules/Macros
HCPIPM HCPIPN
Fix information
Fixed component name
VM CP
Fixed component ID
568411202
Applicable component levels
Fix is available
Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.
[{"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG27M"},"Platform":[{"code":"PF054","label":"z\/OS"}],"Version":"710","Line of Business":{"code":"LOB16","label":"Mainframe HW"}}]
Document Information
Modified date:
02 October 2021