0d7d14b0c2
Changed names of kubernetes and openstack subdirs, and reorganized index files for consistency with other docs. Signed-off-by: Stone <ronald.stone@windriver.com> Change-Id: I234dd88e500f12e7c68b0e8bc5b68ca6e0875aba
1.8 KiB
Host Status and Alarms During System Configuration Changes
For all types of configuration changes, alarms and status messages appear while the system is in transition. You can use the information provided by these messages to help guide the transition successfully.
Configuration changes may require multiple hosts to be reconfigured, during which the settings across the cluster are not consistent. This causes alarms to be raised for the system.
- Changes to the DNS server configuration cause transitory alarms. These alarms are cleared automatically when the configuration change is applied.
- Changes to the External network IP addresses or server addresses, and in particular to the controller storage allotments, cause persistent alarms. These alarms must be cleared manually, by locking and unlocking the affected hosts or performing other administrative actions.
Alarms appear on the Fault Management page of the Horizon web interface, and related status messages appear on the Hosts tab on the Host Inventory page. A variety of alarms can be reported on the Fault Management page, depending on the configuration change.
Caution
To help identify alarms raised during a configuration change, ensure that any existing system alarms are cleared before you begin.
On the Hosts tab of the Host Inventory page, the status Config out-of-date is shown for hosts affected by a configuration change. Each host with this status must be locked and then unlocked to update its configuration and clear the alarm.