Upgrading a Cluster |
Upgrading cluster nodes can be done while the Cluster Manager is running, and does not require that the Cluster Manager itself is also upgraded.
You can confirm that the nodes are being upgraded by watching their replacement and checking the version of Enterprise Server that each one is running. From the Clusters page of the Cluster Manager UI, click the number in the Nodes column to open the Monitor Nodes page. The Current Pool column shows the replacment status (REPLACEMENT IN PROGRESS for new nodes, and IDLE or RUNNING if the node is pending replacement or is replaced. Click each node to view the version of Enterprise Server and confirm it is the one associated with your upgrade.
If the node is listed as DEGRADED after an upgrade and the NODE service is shown with an ERROR status, continue to the next step.
<server> ... <activity_bandwidth_logging>true</activity_bandwidth_logging> ... </server>This setting is required by Cluster Manager to use the Node API for querying the status of transfers. As of version 1.2.2, cluster templates include this setting.
# /opt/aspera/bin/asnodeadmin --db-updateThis command updates the Redis database that is used by the Node API to the latest format. To activate the database update, you must trigger a restart of asperanoded on the nodes. Go to the Transfer Configuration tab for the cluster and click Edit. Add a space after a setting (not within a tag) and save your changes.