system hang after 'bfa: abort cmnd' error messages in the logs

Issue

• System becomes unresponsive after following error messages in the logs:

kernel: bfa 0000:24:00.1: scsi8: abort cmnd ffff81042dc95200 iotag 387
kernel: bfa 0000:24:00.1: scsi8: complete abort 0xffff81042dc95200 iotag 0x387 
kernel: lin_tape 8:0:0:0: timing out command, waited 2000s 
kernel: lin_tape: IBMtape4-----26213 driver_byte 06, host_byte 07, msg_byte 00, status_byte 00
[...]
lin_tape: IBMChgr0-----0_LLA changer_check_result sensekey: 6 asc: 29 ascq: 0
INFO: task dsmserv:848 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
dsmserv       D ffff81000900caa0     0   848      1          1958 30389 (NOTLB)
 ffff81030c09fc28 0000000000000082 ffff81042a18a670 0000000000000000
 0000000000000000 0000000000000009 ffff81030d940040 ffff81010ebaf100
 00003d5892b122e8 0000000000009563 ffff81030d940228 0000000100000080
Call Trace:
 [<ffffffff80063171>] wait_for_completion+0x79/0xa2
 [<ffffffff8008ee78>] default_wake_function+0x0/0xe
 [<ffffffff887bce7d>] :lin_tape:lin_tape_execute_async+0xc5/0xfc
 [<ffffffff887bcf2d>] :lin_tape:tape_execute_scsi_command+0x79/0xa6
 [<ffffffff887c059f>] :lin_tape:tape_send_scsi_io+0x191/0x1fa
 [<ffffffff887c069f>] :lin_tape:tape_send_scsi_cmd+0x97/0x220
 [<ffffffff887aef87>] :lin_tape:lin_tape_perform_write+0x175/0x20a
 [<ffffffff887b8553>] :lin_tape:lin_tape_drive_write+0x169/0x268
 [<ffffffff887aad45>] :lin_tape:lin_tape_write+0x220/0x273
 [<ffffffff80016b9e>] vfs_write+0xce/0x174
 [<ffffffff80017467>] sys_write+0x45/0x6e
 [<ffffffff8005d116>] system_call+0x7e/0x83