I've rebooted h1guardian0 as a requested test to see if it recovers gracefully after a restart, which it appears to do.  Also, I took the opportunity to configure the IPMI management port during the reboot.  Note that when I logged in to reboot the machine, there were 20 zombie processes reported.  That's probably not great since it was just rebooted yesterday.