After upgrade FC latency increased 10 times
Applies to
ONTAP
Issue
- After a node was rebooted during the upgrade process Fibre channel (FC) Host experiences extremely degraded IO performance with Brocade FC switches.
- SRAM dump autosupport is generated
Brocade:
porterrshow- Shows error counters increasing
nsshow- Shows no device registered on port
sfpshow- RX Power is very low
ONTAP:
fct_tpd_thread_1: ems.engine.suppressed:debug]: Event 'fcp.io.status' suppressed 1 times in last 5 seconds.
fct_tpd_thread_1: fcp.io.status:debug]: STIO Adapter:1b, found hung cmd:0xfffff8093c5d12f0(state=10, flags=0x2, ctio_sent=2/3,RecvExAddr=0x212b, OX_ID=0x867, RX_ID=0xffff,SID=0x10800, Cmd[28], req_q_free:0)
fct_tpd_thread_1: scsitarget.fcp.dump:debug]: FCP target SRAM dump generated for adapter 1b, fct_tpd_check_hung_commands: Command termination hung. cmd:0xfffff8093c5d12f0 (state=0xa, flags=0x2,ctio_sent=2/3, RecvExAddr=0x212b, OX_ID=0x867, RX_ID=0xffff, SID=0x10800)
fct_tpd_thread_1: scsitarget.hwpfct.dump.saved:notice]: A dump for adapter 1b was stored in /etc/log/fctsli_1b/fct_fw_1b.dmp.gz.
fct_tpd_thread_1: callhome.fcp.sram.dump:error]: Call home for FCP SRAM DUMP.
fct_tpd_thread_1: scsitarget.fct.reset:notice]: Resetting Fibre Channel target adapter 1b.