To clean up an unplanned PDL:
- Power off the relevant VMs
- Verify the process of the VM is gone
esxcli vm process list
- Restart Management Agents (If needed): If the PDL state persists, or if hostd (the main host management daemon) is behaving erratically, a restart can clear the kernel's reference.
/etc/init.d/hostd restart
/etc/init.d/vpxa restart
- Rescan All from UI or cli :
esxcli storage core adapter rescan --all
- In rare cases, especially when the device is still stuck in a persistent Dead state despite the LUN being available, a reboot of the ESXi host may be required to clear the kernel's persistent memory of the PDL state.
- Place Host in Maintenance Mode.
- Reboot the Host.
- Exit Maintenance Mode and verify the RDM is fully functional.