Troubleshooting
Problem
On AIX hosts or VIO servers, Qlogic or Cavium IBM fibre adapters in Power Systems register as targets in the SAN fabric instead of initiators.
This issue applies to adapters with the following feature codes / CCIN's:
FC EN0Y &EN12; CCIN ENOY
FC EN0F & EN0G; CCIN 578D
FC 5708 & 5270; CCIN 2B3B
FC EN1E, EN1F; CCIN 579A
FC EN1J, EN1K; CCIN 579C
FC EN1G, EN1H; CCIN 579B
Historically, this has only been a concern for customers who are not doing single initiator zoning. That is, if you put 2 or more initiators (host-side HBA's) in the same zone and 1 or more of those is an adapter with one of the listed Feature Codes, you may see the issues described in this document. This is not a new issue, having been present for years. Thus, if the customer has not being seeing any issues with these adapters, there should be no concern.
That said, it's still best practice to modify the sw_prli_rjt default behavior to avoid issues in the future where configuration changes or zoning changes might expose the problem.
Symptom
Any one of a number of symptoms might be present, including:
1. NPIV client LPARs do not discover devices during scans by system firmware (SMS or ioinfo)
2. AIX hosts can fail to boot
3. The AIX error log might be filled with extraneous errors when SAN monitoring software runs, or even when cfgmgr runs, as the adapter attempts to log in to itself. The errors decode as name server query failures. Detailed SENSE DATA indicates the failure was against the physical adapter's own N_Port ID.
Cause
The behavior described is by design from Qlogic and Cavium. It's part of the adapter firmware design.
This issue applies to adapters with the following feature codes / CCIN's:
FC EN0Y &EN12; CCIN ENOY
FC EN0F & EN0G; CCIN 578D
FC 5708 & 5270; CCIN 2B3B
FC EN1E, EN1F; CCIN 579A
FC EN1J, EN1K; CCIN 579C
FC EN1G, EN1H; CCIN 579B
Environment
The problem occurs for the fibre adapters with the listed feature codes and CCINs in IBM Power Systems running AIX or VIOS.
Diagnosing The Problem
If any of the symptoms described occur, check to see what type of physical fibre adapters are in use on the systems. If they are in the list of impacted adapters, they are subject to this issue.
Resolving The Problem
The problem is correcting by modifying a nondisplayable adapter attribute as follows:
Setting this attribute correctly will correct the host-side errors and will correct alerts on certain switches
as well. By default, it's set to no. Sample from the PdAt entry in the ODM for an adapter of type
fcs0 Available 04-00 U78CB.001.WZS0N9S-P1-C3-T1 PCIe2 8Gb 2-Port FC
Adapter (77103225141004f3)
PdAt:
uniquetype = "adapter/pciex/77103225141004f"
attribute = "sw_prli_rjt"
deflt = "no"
values = "yes,no"
width = ""
type = "R"
generic = "U"
rep = "sl"
nls_index = 0
The attribute can be corrected on AIX by doing the following for each adapter:
chdev -l fcs# -a sw_prli_rjt=yes -P
bosboot -ad /dev/ipldevice
reboot
OR - to make the change without reboot (assuming MPIO is in use), change one path/adapter at a time:
rmpath -p fscsi#
rmdev -Rl fcs#
chdev -l fcs# -a sw_prli_rjt=yes
cfgmgr -vl fcs#
To correct the issue on a VIO server perform the same steps from the oem_setup_env prompt, then reboot the VIO server.
NOTE: It's also possible to make the change on a VIO server without reboot by doing one adapter at a time. The caveat is that before changing each adapter, any vfchosts mapped to the adapter must first be unmapped. Then after the cfgmgr step has been run, the vfchosts can be remapped and the next adapter modified.
Document Location
Worldwide
[{"Type":"MASTER","Line of Business":{"code":"LOB08","label":"Cognitive Systems"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SWG10","label":"AIX"},"ARM Category":[{"code":"a8m0z0000001fMuAAI","label":"AIX General Support"}],"ARM Case Number":"TS009810365","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]
Was this topic helpful?
Document Information
Modified date:
14 March 2023
UID
ibm16609886