Today I had a problem with a host that could not restart the management agent, it just hung at the starting agent in the console. I tried to restart the /etc/init.d/hostd restart and /etc/init.d/vpxa restart commands form a ssh prompt, but no dice. I looked in the system logs and found that there was some problems with connectivity to some LUN from an old storage box that was detatched some time ago. Using SHH the only way to communicate with the host I ran this command : esxcfg-mpath -L | more the output showed that some of the LUNs connections was in the state of dead! Then I ran esxcfg-rescan vmhba5 the storage services will now refresh its connections on each LUN. This can take some time, depending on how many LUN you have. I hope this can help when you are discarting LUNs 🙂