OpsCenter 5.1.4 Release Notes
Release notes for the OpsCenter version 5.1.4 release.
Backup Service
Backported from 5.2.0:
- Fixed restore from S3 using Other Location. (OPSC-5915)
- S3 information added to cluster_name.conf for ad hoc backups is now cleaned up (removed) when the backup is completed. (OPSC-4570)
- Fixed issue with Create Backup dialog positioning when using Advanced Options. (OPSC-3951)
New in 5.1.4; will also be included in a future 5.2.x patch release:
- Dramatically improved transfer speeds to/from S3. (OPSC-5067)
- Fixed validation for stream-throttling integer to not allow decimals. (OPSC-4390)
- Fixed issues with backups to S3 falsely appearing to be stalled in some edge cases; this is a logical bug found in code that has never been reported in the wild. (OPSC-5935)
Diagnostic tarball
Backported from 5.2.0:
- Users will be prompted to enter the location of system.log when downloading a diagnostics tarball if it cannot be found in a standard location. (OPSC-1854)
- Sample of cfhistograms are now included. (OPSC-3954)
- Spark configuration and log files are now included. (OPSC-3653)
- solrconfig.xml and schema.xml are now included. (OPSC-2450)
Provisioning and configuration
Backported from 5.2.0:
- Java class name suggestions are now given for various cassandra.yaml properties when configuring the cluster. (OPSC-3853)
- Added a warning when adding nodes with a different DSE version than existing nodes. (OPSC-2493)
- Improved stability in Cloud provisioning. (OPSC-4508)
- Improved error handling during provisioning when Java is not installed on target machines. (OPSC-3443)
Other improvements
Backported from 5.2.0:
- Added a best practice rule to encourage users to encrypt sensitive configuration values. (OPSC-4462)
- Ability to see Best Practice Service rule description by hovering over the rule. (OPSC-3137)
- Improved error messages on login page. (OPSC-5021)
- Added Search: Core Size metric to monitor and alert on disk space used by Solr cores. (OPSC-3523)
- Number of connected clients added to Node dialog. (OPSC-2115)
- Added warning when cloning schema-related graphs. (OPSC-3649)
New in 5.1.4; will also be included in a future 5.2.x patch release:
- Added the dse_security keyspace to the list of system keyspaces ignored in: backup/restore, rebalance calculations, metrics collection, and repair service. (OPSC-4633)
Other resolved issues
Backported from 5.2.0:
- Improved handling of metrics overflow queue on agent. (OPSC-4618)
- Fixed issues with agent OOM when storing metrics for large numbers of tables. (OPSC-5934)
- Fixed the install_agent script trying to run "yum install" on some ubuntu instances with yum installed (uses apt-get instead). (OPSC-5184)
- Rebalance menu option now disabled for clusters using vnodes. (OPSC-3957)
- Removed unnecessary validation of settings when deleting a managed cluster. (OPSC-5055)
- Fixed issue with rolling restart not failing properly when DSE fails to startup. (OPSC-4875)
- Fixed Cluster Report and Diagnostics tarball failing in some disk configurations. (OPSC-3153)
- Fixed cloning Dashboard configurations after first load of UI. (OPSC-5284)
- Fixed Alerts link when managing multiple clusters. (OPSC-5857)
- Fixed sorting by Status in List View. (OPSC-4935)
- Fixed issue where the Select All check box in the List View would choose all nodes in the cluster rather than only the visible nodes. (OPSC-4097)
New in 5.1.4; will also be included in a future 5.2.x patch release:
- Upgraded DataStax Java Driver using in the agent to 2.1.6, which fixes many connection and reconnection issues to the node used for storing OpsCenter data. (OPSC-5657)
- The reconnect back-off ceiling is now 15 seconds instead of ten minutes, and is now configurable via max_reconnect_time. (OPSC-6268)
- Added proper HTML escaping to Dashboard tab names. (OPSC-4586)
- Fixed divide-by-zero errors seen in agent.log when collecting metrics in some cases. (OPSC-3293)
- Removed deprecated "Streams Pending" metric. (OPSC-4666)
- Improved error handling when calling the "df" command on the agent, which is used when collecting information and metrics about the disks on each node. (OPSC-3075)
Compatibility
To see which versions of OpsCenter are compatible with the various Cassandra and DataStax Enterprise versions, see the OpsCenter Compatibility chart.