Last modified: 2014-03-20 15:35:09 UTC
On the beta cluster, we have to play with lvm and mount some volume at /srv/vdb . We end up with instances being stalled after reboot with: The disk drive for /srv/vdb is not ready yet or not present. Continue to wait, or Press S to skip mounting or M for manual recovery The affected instances are: https://wikitech.wikimedia.org/wiki/Nova_Resource:I-00000080.eqiad.wmflabs https://wikitech.wikimedia.org/wiki/Nova_Resource:I-00000103.eqiad.wmflabs Looking at the OpenStack documentation at https://wikitech.wikimedia.org/wiki/OpenStack#Get_an_interactive_console_to_an_instance_from_a_host : virsh console --devname serial1 <instance id> Daniel Zahn attempted to get a console access to the instances from their compute nodes but when entering a key ends up with: error: internal error: character device (null) is not using a PTY The instance can not be rebooted from the wikitech manager either.
The locked instances got unlocked by shut downing them then mounting the root partition following the instructions at https://wikitech.wikimedia.org/wiki/OpenStack#Mounting_an_instance.27s_disk . Then the faulty entry in /etc/fstab got wiped out letting the instance boot properly. We still have a console issue though :(
