ZFS can't migrate to other node (cannot open pool: no such pool)

Cluster pacemaker+corosync, Centos 9 Stream VMs with quorum.
[root@vnas-centos-1 labadmin]# pcs resource config
Group: group-pool1
Resource: zfs-pool1 (class=ocf provider=heartbeat type=ZFS)
Attributes: zfs-pool1-instance_attributes
importargs="-d /dev/disk/by-id/"
pool=pool0
Operations:
monitor: zfs-pool1-monitor-interval-5s
interval=5s
timeout=30s
start: zfs-pool1-start-interval-0s
interval=0s
timeout=15s
stop: zfs-pool1-stop-interval-0s
interval=0s
timeout=15s
Resource: vip-scsi (class=ocf provider=heartbeat type=IPaddr2)
Attributes: vip-scsi-instance_attributes
cidr_netmask=24
ip=192.168.90.50
Operations:
monitor: vip-scsi-monitor-interval-10s
interval=10s
timeout=20s
start: vip-scsi-start-interval-0s
interval=0s
timeout=20s
stop: vip-scsi-stop-interval-0s
interval=0s
timeout=20s
Resource: target-pool1 (class=ocf provider=heartbeat type=iSCSITarget)
Attributes: target-pool1-instance_attributes
additional_parameters="DefaultTime2Retain=0 DefaultTime2Wait=0"
implementation=lio-t
iqn=iqn.2004-10.com.ubuntu:01:84de25ddfc37
portals=192.168.90.50
Operations:
monitor: target-pool1-monitor-interval-5s
interval=5s
timeout=1s
on-fail=restart
start: target-pool1-start-interval-0s
interval=0s
timeout=10s
stop: target-pool1-stop-interval-0s
interval=0s
timeout=10s
Resource: lun1-pool1 (class=ocf provider=heartbeat type=iSCSILogicalUnit)
Attributes: lun1-pool1-instance_attributes
implementation=lio-t
lun=0
path=/dev/pool0/vol1
target_iqn=iqn.2004-10.com.ubuntu:01:84de25ddfc37
Operations:
monitor: lun1-pool1-monitor-interval-10s
interval=10s
timeout=10s
start: lun1-pool1-start-interval-0s
interval=0s
timeout=10s
stop: lun1-pool1-stop-interval-0s
interval=0s
timeout=10s

Moving resource by pcs resource move resource_id nodename is OK, but if I try to hard stop node (reset VM in VMware), in log view this:

Mar 30 15:23:17.738 vnas-centos-1 pacemaker-execd [1242] (log_op_output) info: zfs-pool1_start_0[2266] error output [ cannot open 'pool0': no such pool ]
Mar 30 15:23:17.739 vnas-centos-1 pacemaker-execd [1242] (log_op_output) info: zfs-pool1_start_0[2266] error output [ /usr/lib/ocf/resource.d/heartbeat/ZFS: line 35: [: : integer expression expected ]
Mar 30 15:23:17.739 vnas-centos-1 pacemaker-execd [1242] (log_op_output) info: zfs-pool1_start_0[2266] error output [ cannot import 'pool0': one or more devices is currently unavailable ]
Mar 30 15:23:17.739 vnas-centos-1 pacemaker-execd [1242] (log_op_output) info: zfs-pool1_start_0[2266] error output [ /usr/lib/ocf/resource.d/heartbeat/ZFS: line 35: [: : integer expression expected ]
Mar 30 15:23:17.739 vnas-centos-1 pacemaker-execd [1242] (log_finished) info: zfs-pool1 start (call 26, PID 2266) exited with status 1 (execution time 237ms)
Mar 30 15:23:17.739 vnas-centos-1 pacemaker-controld [1245] (log_executor_event) notice: Result of start operation for zfs-pool1 on vnas-centos-1: error | CIB update 23, graph action confirmed; call=26 key=zfs-pool1_start_0 rc=1
Mar 30 15:23:17.739 vnas-centos-1 pacemaker-controld [1245] (log_executor_event) notice: zfs-pool1_start_0@vnas-centos-1 output [ cannot open 'pool0': no such pool\n/usr/lib/ocf/resource.d/heartbeat/ZFS: line 35: [: : integer expression expected\ncannot import 'pool0': one or more devices is currently unavailable\n/usr/lib/ocf/resource.d/heartbeat/ZFS: line 35: [: : integer expression expected\n ]

Start by inspecting /usr/lib/ocf/resource.d/heartbeat/ZFS, especially line 35, and compare its start behavior with the logged pcs resource configuration. Reproduce the hard-stop or VM-reset migration scenario and trace the ZFS import output. Done means the resource agent handles failover to the other node without the reported pool and integer-expression errors.

Written by the indexing model from the issue text.