Had an issue not long ago when adding a Windows Cluster, it grabbed all available LUNs presented to the server and put a reservation lock on them. The LUNs were also presented to VMware, rendering VMware unavailable.

Offlining and removing the LUNs from the Windows Cluster didn’t remove the reservation lock and the VMware host kernel log shows the below.

2020-02-06T12:18:15.540Z cpu15:33069)lpfc: lpfc_scsi_cmd_iocb_cmpl:2145: 1:(0):3271: FCP cmd x0 failed <0/9> sid x580b01, did x580300, oxid x573 SCSI Reservation Conflict -

2020-02-06T12:20:04.540Z cpu21:33554)lpfc: lpfc_scsi_cmd_iocb_cmpl:2145: 0:(0):3271: FCP cmd x0 failed <0/10> sid x7f0a02, did x7f0300, oxid x237 SCSI Reservation Conflict -

Because all Windows options had been exhausted, turned to the SAN itself which was a HP 3PAR V400 & 8200 to see what locks were present.

SSH to your 3PAR Storage System (with admin access), hostname or IP and run the command showrsv to show current reservations.

HPE-SAN-8200 cli% showrsv
             VVname        Host            Owner  Port   ReservationType
Vv_8200_Vmfs_DATA.01    s1-sql2 5001438021E07AC0 0:0:1       SCSI-3(5)
Vv_8200_Vmfs_DATA.02    s1-sql2 5001438021E07AC0 0:0:1       SCSI-3(5) 

This returned the host that had grabbed all available LUNs and presented to Windows Cluster (this should not be there!) time to remove the locked reservations.

HPE-SAN-8200 cli% setvv -clrrsv Vv_8200_Vmfs_DATA.01

WARNING: Active hosts may be disrupted.

Are you sure you want to run setvv -clrrsv Vv_8200_Vmfs_DATA.01
 select q=quit y=yes n=no: y 

Re-run showrsv and setvv -clrrsv {VVname} until all the reservations are gone.

Restart the OS and see if the disks are back online!