devzero
Expert
Expert

QLA2462 HBA - driver issues with VMDirectPath on 6.7

I'm using virtual machine with QLA2462 & SCST ( SCST: A Generic SCSI Target Subsystem for Linux  )  via Passthrough.


It's working fine so far, there is ony one problem - on initial connect between target and initiator, after 60s (which is default for scsi_transport_fc) there is a timeout happening which leads to device-loss and interruption.

after reconnect everything is working fine and stable. no more timeouts.

this does not seem to happen if the scst fc target is run on real hardware, at least i was not yet able to reproduce.

does anybody have a clue what could cause this issue and where to look at?

- latency/timing - looking at the scst/linux driver code for time/latency critical sections?

- lost interrupts?

- "other" issues related to VMDirectPath ?

- VMDirectPath related tunables  ?

Anybody running fibrechannel via Passthrough between 2 VM nodes?

I need to put my nice poor-man's-2node clustering project to thrash if i don't get this resolved Smiley Sad

roland

node1

[ 48.284747] qla2xxx [0000:13:00.0]-505f:35: Link is operational (1 Gbps).

[ 48.288458] [94]: qla2x00t(35): RSCN registration failed: 0x2 (OK for non-fabric setups)

[ 49.475659] qla2xxx [0000:13:00.1]-505f:36: Link is operational (1 Gbps).

----- !!!!

[ 98.796184] rport-35:0-0: blocked FC remote port time out: removing rport

[ 98.797606] [30]: qla2x00t(35): session for port 21:01:00:1b:32:b1:ad:fa (loop_id 0) scheduled for deletion in 20 secs

[ 118.829329] [30]: qla2x00t(35): local session for loop_id 0 deleted

node2

[ 894.397119] [1961]: qla2x00t(39): LIP reset occurred

[ 894.398708] [1961]: qla2x00t(39): LIP occurred

[ 894.411109] [1961]: qla2x00t(39): Loop up occurred

[ 894.423097] qla2xxx [0000:13:00.0]-505f:39: Link is operational (1 Gbps).

[ 894.435029] [1961]: qla2x00t(39): LIP reset (loop 0xffff), subcode 2

[ 894.502087] [1721]: qla2x00t(39): RSCN registration failed: 0x2 (OK for non-fabric setups)

----- !!!!

[ 954.200232] rport-39:0-0: blocked FC remote port time out: removing rport

[ 954.203290] [1817]: qla2x00t(39): session for port 21:01:00:1b:32:b1:3c:8d (loop_id 0) scheduled for deletion in 20 secs

[ 954.205070] [0]: qla2x00t(39): CTIO with PORT LOGGED OUT (29) or PORT UNAVAILABLE (28) status 29 received (state 3, scst_cmd ffff8fdf76850a48, op READ(10))

[ 954.206880] [0]: qla2x00t(39): CTIO with PORT LOGGED OUT (29) or PORT UNAVAILABLE (28) status 29 received (state 3, scst_cmd ffff8fdf76852108, op READ(10))

[ 954.209056] [0]: qla2x00t(39): CTIO with PORT LOGGED OUT (29) or PORT UNAVAILABLE (28) status 29 received (state 3, scst_cmd ffff8fdf76851748, op READ(10))

[ 954.210878] [0]: qla2x00t(39): CTIO with PORT LOGGED OUT (29) or PORT UNAVAILABLE (28) status 29 received (state 3, scst_cmd ffff8fdf76850708, op READ(10))

[ 974.233191] [1817]: qla2x00t(39): local session for loop_id 0 deleted

[ 984.291824] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 984.297207] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 984.305134] [1721]: ***WARNING***: qla2x00t(39): Unable to find initiator with S_ID 0:0:e8

[ 984.307612] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 984.308588] [1721]: ***WARNING***: qla2x00t(39): Unable to find initiator with S_ID 0:0:e8

[ 984.312021] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 984.312927] [1721]: ***WARNING***: qla2x00t(39): Unable to find initiator with S_ID 0:0:e8

[ 984.318881] [1721]: ***WARNING***: qla2x00t(39): Unable to find initiator with S_ID 0:0:e8

[ 984.321171] [1721]: ***WARNING***: qla2x00t(39): Unable to find initiator with S_ID 0:0:e8

[ 984.323435] [1721]: ***WARNING***: qla2x00t(39): Unable to find initiator with S_ID 0:0:e8

[ 988.311984] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 988.312806] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 988.313568] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 988.314314] [0]: ***WARNING***: qla2x00t(39): ABTS_RESP_24XX failed 31 (subcode 19:a)

[ 992.343396] [1721]: qla2x00t(39): RSCN registration failed: 0x2 (OK for non-fabric setups)

[ 992.348011] [1721]: qla2x00t(39): local session for wwn 21:01:00:1b:32:b1:3c:8d (loop_id 0, s_id 0:0:e8, confirmed completion not supported) added

[ 992.349489] [1539]: scst: Using security group "testgrp" for initiator "21:01:00:1b:32:b1:3c:8d" (target 21:00:00:1b:32:91:ad:fa)

0 Kudos
0 Replies