Hi all,
Due to a storage-presentation mistake I made, I have a 5.1 Host that is now unmanagable and disconnected from vCenter.
All the VM's are still happily running, but I would like to avoid having to shut them all down if possible.
I have been trying my hardest to breath some life back into the management agents.
No matter what I try, hostd doesn't seem to want to shutdown or restart properly, its status is always 'running' while being simultaneously completely unresponsive.
~ # /etc/init.d/hostd stop
watchdog-hostd: PID file /var/run/vmware/watchdog-hostd.PID does not exist
watchdog-hostd: Unable to terminate watchdog: No running watchdog process for hostd
hostd stopped.
~ # /etc/init.d/hostd status
hostd is running.
Stuff I have tried so far:
KB1003490 - /etc/init.d/hostd restart, /etc/init.d/vpxa restart, services.sh restart
KB1005566 - deleted vmware-hostd.PID watchdog-hostd.PID and tried to forcefully kill hostd with kill -9
KB1003494
KB1002849 - No other process seems to be keeping hostd busy as far as I can see, and the host is very quiet. Hostd itself doesnt seem to be doing anything.
Also tried:
Johnny's Random Tips: Restart hostd - "The real way" - This involves deleting the files in /var/lib/vmware/hostd/stats/, Didn't help.
services.sh restart hangs when it gets to /etc/init.d/vpxa , so I generated the list of services myself and manually tried to restart all of the rest of them. That didnt really help either.
~ # /sbin/chkconfig -o | sed -n -e '1!G' -e 'h' -e '$p'
/etc/opt/init.d/vmware-fdm
/etc/init.d/xorg
/etc/init.d/wsman
/etc/init.d/snmpd
/etc/init.d/sfcbd
/etc/init.d/sfcbd-watchdog
/etc/init.d/vpxa
/etc/init.d/vobd
/etc/init.d/memscrubd
/etc/init.d/cdp
/etc/init.d/lacp
/etc/init.d/smartd
/etc/init.d/dcbd
/etc/init.d/slpd
/etc/init.d/vprobed
/etc/init.d/hostd
/etc/init.d/rhttpproxy
/etc/init.d/lbtd
/etc/init.d/storageRM
/etc/init.d/sensord
/etc/init.d/usbarbitrator
/etc/init.d/ntpd
/etc/init.d/lsassd
/etc/init.d/netlogond
/etc/init.d/DCUI
/etc/init.d/lwiod
So anyway, I am plain out of ideas.
My next step is to go and arrange an organised shutdown of the VM's running on the host, so I can reboot it.
But perhaps someone here has any more ideas, things I can try.