OpsCenter 6.1.1 Release Notes

Release notes for the OpsCenter version 6.1.1 release.

Highlights

  • Many helpful content additions to the diagnostic tarball for better troubleshooting, along with an improved customer experience.
  • Support for HTTP Strict Transport Security (HSTS).
  • Added SSL truststore configuration options for the agent.
  • Summary panels for Services and cluster health have been added to the Nodes page for a quick view of status and fast access to each Services section.

Upgrade Information

Important: Please be sure to review the 6.1 changes pertinent to the release as noted in the DataStax OpsCenter Upgrade Guide. Configuration and other notable changes are provided in detail.

Compatibility

To see which versions of DataStax Enterprise are supported with OpsCenter 6.x, see the OpsCenter Compatibility chart.

Known Issues

Important: Please be sure to review the list of known issues before running a new OpsCenter version on a production DSE cluster.

Core

  • Added some additional files (logback.xml files to /opscenterd and /dse, log4j.properties to /agent, jvm.options and commitlog_archiving.properties to /cassandra, and /etc/hosts) to diagnostic tarball contents. Also fixed an issue with a missing trailing slash that caused agent diagnostic files to be filed directly under the /conf folder instead of the /agent subfolder. (OPSC-11611)
  • Agent status information is now included in the diagnostic tarball. (OPSC-7277)
  • When reporting a failure to generate a diagnostic tarball due to a lack of available disk space, the amount of space required, the amount of space available, and the working directory for diagnostic tarball generation is now clearly logged. (OPSC-5959)
  • The diagnostic tarball no longer creates a file containing an error message rather than a directory if a node's diagnostic tarball could not be downloaded. A file name agent_requests.json is now included that indicates the success or failure of retrieving diagnostic tarballs from each node. (OPSC-6258)
  • Whenever possible, any current information from the Best Practice Service is now included in the diagnostic tarball. (OPSC-4956)
  • The agent now re-uses an existing connection to DSE when gathering some diagnostic tarball output rather than forking additional cqlsh processes. (OPSC-7081)
  • The agent will retry hung connection attempts to DSE nodes. (OPSC-11935)
  • Summary panels with status and links to the Services have been added to the Nodes page. (OPSC-12138)
  • Selecting a keyspace is now optional when performing a cleanup, compact, or flush node operation. This feature is not backwards-compatible with agent versions earlier than 6.1.1. (OPSC-510)
  • Added truststore settings to agent configuration for both monitored and storage clusters. (OPSC-8003)
  • Cleaned up logging on opscenterd shutdown. (OPSC-8278)
  • Selecting NetworkTopologyStrategy while editing a keyspace in the OpsCenter Data section now populates the datacenter replication factor parameters with the datacenters in the cluster topology. Any previously specified replication factor values for a given datacenter parameter will be reused. Assigning a replication factor of 0 to a datacenter parameter excludes replication on the datacenter. (OPSC-10680)
  • Included change details in audit log messages for updates to role permissions. (OPSC-10929)
  • Added summary information about running Services to the cluster landing page. (OPSC-11247)
  • Removed repetitive log message 'INFO: Node 172.31.11.1 has multiple tokens (vnodes). Only one picked for display. (MainThread)'. Only affects vnode clusters. (OPSC-11557)
  • Added the X-XSS-Protection to content responses to enable the XSS auditor in various supported browsers. (OPSC-11845)
  • Fixed a bug where opscenterd would automatically encrypt values in its configs if a system key was present, whether config encryption was enabled or not. (OPSC-7152)
  • Allow access to the left navigation, including LCM, when there's only one cluster configured and OpsCenter cannot connect to it. (OPSC-11101)
  • Provide clear messages in the diagnostic tarball when using DSE 5.0 or greater that the removal of Thrift has also removed the use of cassandra-cli. (OPSC-11239)
  • Improve messaging around timeouts when logging in via LDAP. (OPSC-11396)
  • New OpsCenter roles now default to having no permissions for any clusters. (OPSC-11685)
  • The DataStax Agent may now use different passwords for the keystore and truststore. (OPSC-11767)
  • Fixed a bug when resuming monitoring after a rolling restart. (OPSC-12095)
  • Optimized requests for Service status while viewing a service detail page. (OPSC-12136)
  • Improved error handling logging around HTTP calls to the agent. (OPSC-12143)
  • Improved handling of separate storage cluster. (OPSC-8235)
  • Attempting to log in without specifying a username now gives a proper HTTP Unauthorized response instead of a 500 error response. (OPSC-9469)
  • If OpsCenter authentication is enabled, users who do not have View Schema permission in their assigned role are not allowed to view schema data in the Data area of OpsCenter monitoring. OpsCenter presents an informational message that instructs users to contact their OpsCenter administrator for access. (OPSC-11319)
  • Redacted passwords from debug log statements in WrappedDriver. (OPSC-11448)
  • Fixed a bug during agent install that is triggered when no events have been reported. (OPSC-11640)
  • Updated solr-cores and range-list routes in agent to address intermittent errors. (OPSC-11674)
  • Fixed agent processing of KMIP errors for alerts. This was causing log spam in the agent and was preventing other real-time information from being reported to OpsCenter. (OPSC-11972)
  • Fixed a bug where the dashboard would not display when non-ASCII characters are present in a username. (OPSC-12064)
  • Fixed an issue where requests would timeout when using AD 2012 with SSL. (OPSC-11995)

Monitoring

  • Added Speculative Retries metrics to OpsCenter monitoring. (OPSC-11503)
  • Added Coordinator Read Latency table metric. (OPSC-11509)
  • Clarify the units in several 'TP: Dropped X' metrics. (OPSC-11714)

Backup Service

  • Improved performance when displaying updates in the restore status dialog. (OPSC-1498)
  • Added support for folders and subfolders to Amazon S3 buckets used for backups and restores. (OPSC-6845)
  • Agent configuration now validates that backup_staging_dir is set to an absolute path. (OPSC-9393)
  • Fixed sorting of scheduled backups. (OPSC-11007)
  • Horizontal scroll has been added to the OpsCenter Services section when the browser window is too small. (OPSC-11144)
  • Improved the error message during certain backup failure cases. (OPSC-11733)
  • Fixed the display of the list of backups presented after adding a backup location. (OPSC-11933)
  • Fixed an issue where backups to S3 would stop without logging an error. (OPSC-12188)
  • Fixed an issue that caused the destination dialog to appear multiple times when restoring from an ad hoc S3 backup. (OPSC-11569)
  • Removed checks preventing users from restoring backups across major versions. This restriction was in place because earlier but no longer supported versions of Cassandra did not allow this. It is now possible to restore backups from any version of Cassandra that is compatible for upgrading. (OPSC-11766)

Repair Service

  • Increased accuracy when computing repair service job throughput. (OPSC-11171)
  • Fixed a NullPointerException in agents when repair service was running and the cluster was rebalanced. (OPSC-11668)
  • Repair Service settings shows calculated time to completion separately from the setting's input field. (OPSC-11985)
  • Clarified display of total repair attempts in the repair status page. (OPSC-12004)
  • Fixed repair to not use keyspaces in the ignored_keyspace property when computing the maximum number of repairs that can run in parallel. (OPSC-12045)
  • Fixed an issue in the repair service where restarting OpsCenter would disable the repair service. (OPSC-12237)

Performance Service

  • Fixed column sorting in the Slow Queries page of the Performance Service. (OPSC-11704)
  • Performance Service in OpsCenter can only be configured when at least one agent is connected. The OpsCenter api now returns an empty array instead of an empty object for /perf/config when no agents are connected. (OPSC-11267)

Best Practice Service

  • Best Practice Rule for Secondary Index Cardinality no longer fails for system keyspaces. (OPSC-6913)
  • Fixed column sorting in the Slow Queries page of the Performance Service. (OPSC-11704)

Lifecycle Manager (LCM) Provisioning

  • LCM now warns users that config-encryption is not supported. (OPSC-7616)
  • LCM now tries to assign one seed per rack, two seeds per datacenter, and three seeds for each cluster. (OPSC-11160)
  • LCM proceeds with jobs when a become-password is set in the machine-credential even if the target does not prompt for a sudo password, rather than generating a misleading meld-not-started error. (OPSC-11216)
  • Updated the Java download UI in LCM to reflect new Oracle URL structure. Also added tooltips to assist users with completing Java Setup fields for Config Profiles. (OPSC-11418)