Brocade HBA resets 'dev_loss_tmo' from 'infinity' to 31 seconds

Issue

• Brocade HBA resets 'dev_loss_tmo' from 'infinity' to 31 seconds. Due to this, failed ports are removed from system:
  For example, before the failure of 2 FC remote ports multipath device had following 4 paths:

  mpathg (268688d68fd6786df67666681111111888) dm-2 Nimble  ,Server          
  size=500G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
  |-+- policy='round-robin 0' prio=50 status=active
  | |- 5:0:5:200 sdf 8:80 active ready running
  | `- 6:0:3:200 sdg 8:96 active ready running
  `-+- policy='round-robin 0' prio=1 status=enabled
    |- 6:0:2:200 sde 8:64 active ghost running
    `- 5:0:4:200 sdd 8:48 active ghost running
  

• The dev_loss_tmo value for rports connecting above paths was set to infinity (2147483647):

  $ for f in /sys/class/fc_remote_ports/rport-*/dev_loss_tmo; do d=$(dirname $f); echo $(basename $d):$(cat $d/port_name):$(cat $f); done             
  rport-5:0-4:0xaac9bb1011c11103:2147483647
  rport-5:0-5:0xaac9bb1011c11101:2147483647
  rport-6:0-2:0xaac9bb1011c11104:2147483647
  rport-6:0-3:0xaac9bb1011c11102:2147483647
  

• But after the failure of two rports, dev_loss_tmo was reset to 31 seconds and paths were deleted from system:

  $ for f in /sys/class/fc_remote_ports/rport-*/dev_loss_tmo; do d=$(dirname $f); echo $(basename $d):$(cat $d/port_name):$(cat $f); done             
  rport-5:0-4:0xaac9bb1011c11103:31                 <---
  rport-5:0-5:0xaac9bb1011c11101:31                 <---
  rport-6:0-2:0xaac9bb1011c11104:2147483647
  rport-6:0-3:0xaac9bb1011c11102:2147483647
  
  $ multipath -ll
  mpathg (268688d68fd6786df6766668s6f877888) dm-2 Nimble  ,Server          
  size=500G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
  |-+- policy='round-robin 0' prio=50 status=active
  | `- 6:0:3:200 sdg 8:96 active ready running
  `-+- policy='round-robin 0' prio=1 status=enabled
    `- 6:0:2:200 sde 8:64 active ghost running