The following issues are resolved in Juniper Networks CSO Release 3.2.2.
If the device connection status was not updated correctly when a regional infrastructure node was shut down, config push jobs for such devices remain in the in-progress state indefinitely. [CXU-21996]
Because of an auto-delete condition for reply queues in oslo.messaging 4.0.0, the queue was deleted when a connection was lost, and that was causing a not found error when a reconnection was attempted. [CXU-21853]
Because the Cassandra database was not restored correctly during the database backup and restore, the Cassandra database failed to start on the node. [CXU-21802]
Elasticsearch restore process abruptly ends and that causes the indices that were closed as part of the restore to remain in the closed state. [CXU-21497] )
ArangoDB backup in a high-availability deployment fails because of a race condition. ArangoDB restore restores the database to system database instead of topology database. [CXU-21493]
When a stage-2 template has initial configuration that is set to be deployed in device-profile during ZTP, the configuration fails to deploy even though the auto-deployment job is successful. [CXU-19389]
When the infra VM is down, the keystone token for the MySQL database fails to update. This causes sync issues for the MySQL database and jobs to fail with an authentication required error. [CXU-21076]