For information about new features and resolved issues in the following releases, see:
Before upgrading, check the compatibility of the OpsCenter version you want to run against your version of Cassandra or DataStax Enterprise.
OpsCenter Enterprise Edition 2.1.3 supports Cassandra 1.2 when virtual nodes are disabled. This Edition also includes bug fixes.
OpsCenter Enterprise Edition 2.1.3 can monitor and administer Cassandra 1.2 as described in the documentation unless you enable virtual nodes. When you enable virtual nodes, OpsCenter chooses a single token for each node for operations such as collecting metrics. Attempting to move nodes, rebalance nodes, and perform other tasks involving token ranges is not supported.
- Fixed issue processing keyspaces that only exist in 1 datacenter.
- Fixed data collection issues during compactions on secondary indexes.
OpsCenter Enterprise Edition 2.1.2 changes include a new feature and bug fixes.
- Better parsing of some device types when collecting metrics.
- Fixed ability to configure user access on tarball installations.
- Fixed Cluster Report feature on tarball installations.
OpsCenter Enterprise Edition 2.1 key enhancements include data modeling and data browsing, report generation, and additional metrics for performance monitoring.
- Browsing data in the Cassandra database.
- Generation of a PDF report about the cluster.
- Capability to handle thousands of column families efficiently.
- Online feedback form.
- Optional HTTPS support.
- Installation of agents to RHEL and Debian clusters regardless of the location of the configuration file (opscenterd.conf).
- Truncation of column families. Truncation removes the data but not the column family itself.
OpsCenter Enterprise Edition 2.0 is a major release that introduces backup and cluster enhancements.
- Multi-cluster management. You can manage multiple cluster from the same OpsCenter instance.
- New Overview page for monitoring multiple clusters. It shows a condensed view of each cluster's Dashboard. It is visible only when multiple clusters exist.
- Perform and schedule operations from OpsCenter.
- Point OpsCenter at existing Cassandra clusters, where you can add, modify, or delete clusters from OpsCenter.
- Adds support and metrics for DataStax Enterprise 2.0 Search nodes.
- No longer compatible with Cassandra 0.7.
- Fix opscenter-agent package not starting on machine restart.
- Allow node move operation when a keyspace is set to replicate to a non-existent data center.
- Fix agent listen address when SSL is disabled.
- Fix data explorer error when searching for row key containing "#".
OpsCenter 1.4.1 is a patch release to OpsCenter 1.4. For upgrading instructions, follow
the instructions for Upgrading OpsCenter.
- Fix free-space disk check before moving node.
- Add additional support for Linux device types.
- Improvements to OpsCenter keyspace creation logic.
- Include heap options in tarball agent startup script.
OpsCenter 1.4 adds support for two major operating systems and widens support for Python.
For upgrading to 1.4, follow the instructions for Upgrading OpsCenter.
- Support for OpsCenter and Cassandra on Windows. Using a GUI installer, you can
easily be up and running on a 32- or 64-bit Windows 7 or Windows Server 2008 in
a few minutes. OpsCenter and OpsCenter agents run as Windows services.
- Support for OpsCenter on Mac OSX. You can install OpsCenter in a single directory
using a tar package. Root permissions are not required.
- Extended support of Python. OpsCenter supports Python 2.5, 2.6, and 2.7
on Linux, Mac OSX, and Windows.
- Detection of OpenSSL fails on RedHat Enterprise Linux (RHEL) 5. x.
- Configuring user access using the set_passwd.py utility fails.
- When you replace a node at position 0-1, OpsCenter obscures the new node in the ring
- When viewing a table of a column family data in Data Explorer, the Row Key heading of
the table is incorrectly labeled Super Column when you click the search button.
- When keys and super column names include spaces, the Data Explorer does not display
data when you view all columns.
- The Performance graph of data over a period of time momentarily shows a false, steep
drop in reads and writes.
- OpsCenter 1.2.3 on Red Hat Enterprise Linux behaves unpredictably when started with
OpsCenter 1.3.1 is a patch release to OpsCenter 1.3. For upgrades to 1.3.1, follow
the instructions for Upgrading OpsCenter.
This release introduces support for OpsCenter on RedHat Enterprise Linux
- (DataStax Enterprise Only) Creating a new Keyspace while a rebalance is in
progress causes the rebalance operation to error out.
- "Unable to import needed OpsCenter packages" error when running
set_passwd.py to set up OpsCenter authentication.
- Data center and version information not returned from agents when iostat
is missing on monitored nodes.
OpsCenter 1.3 introduces the DataStax Enterprise (DSE) and DataStax Community
(DSC) versions of OpsCenter. DSE contains enhanced features and functionality
and is available to registered DataStax customers. DSE is a free version of
OpsCenter and no longer requires registration with DataStax to download and
- (Enterprise-only) Auto-Rebalancing for Randomly Partitioned Clusters -
OpsCenter Enterprise now has the ability to check if nodes are evenly
distributed in each data center, and if not automatically moves tokens and
data to balance the ring.
- (Enterprise-only) Proactive Alerts - OpsCenter Enterprise now has the ability
to configure alert thresholds for commonly watched performance metrics and to
send alerts when those thresholds are triggered.
- (Enterprise-only) The ability to track the progress of MapReduce jobs running
on DSE Analytics nodes.
- More Reliable JMX Operations on Agent - Improved channels for pushing JMX
operations to the agents from opscenterd, getting success or failure
responses, and polling the status of an operation.
- Self-Contained Distribution - New tarball distribution of OpsCenter to allow
for self-contained, non-root installations.
- Respect the user's JAVA_HOME and PATH environment when starting up
- Sort nodes in correct token order in Cluster list view
- Cluster list view usability improvements
- Limit how many files are rotated for agent.log
- Use more precise timestamps for Events list
- OS memory graph not graphing total correctly
- Data Modeling - Fixed Create Keyspace using NetworkTopologyStrategy