For an upgrade to the new version please follow the Packaged Based Operations manual.
The new EUM functionality relies on a completely new data format, so after the update you will not have historic EUM data. Be aware that with this release the old EUM data will not be accessible anymore.
With release 150 we introduce the posbility to configure your LDAP configuration in our UI. You can try the configuration out before saving it.
We now support anonymous binding for your LDAP configuration.
A previous SaaS only feature will be now available on-premises: We Support authentication via SAML compliant identity providers. Check out the full description
In release 146 we introduced advanced analytics capabilities on website monitoring data within the analyze product area. This allowed advanced queries and through that unbounded analytics. This is great for very specific questions and validation of hypothesis, but it is also non-trivial to do.
With this release we are introducing analytics capabilities to our website monitoring dashboards! This is done with the help of the quick filter bar found in the analyze area of the product. This filter bar is now also available in the dashboards! This super powerful feature can be used to easily slice and dice dashboards to support use cases such as…
- focusing on the performance of a single browser,
- limiting all data to that of a specific geographic location,
- looking into A/B test participants’ experience via meta data.
Release 146 introduced configurable metrics for the grouped website monitoring analysis views. This release extends the capabilities so that the selected metrics are shown in the ungrouped view as well. This enables closer inspection once an interesting group of data points as been identified.
Page views can be complicated to understand just from an activity list. This release adds a visualization to understand when there was a lot of activity on the page.
The Analyze grouped traces and calls views also support configurable metrics such as latency percentiles.
Instana now supports PingDirectory server. Documentation can be found here
Instana now supports AWS Elasticsearch service. Documentation can be found here
When creating an application, you can now automatically include services that are downstream from an entry point, and they will become part of the application. That way you do not have to manually update the definition of an application in case your infrastructure changes, instead it will maintain itself, based on the path that the calls take.
When End User Monitoring is active on Webpages that are generated by Java HTTP frameworks, as Servlet, Spring, Play, Jersey or any other supported library, we now automatically correlate that call with the browser activity via the
Server-Timing header we set.
The agent now supports containerD and LXC containers in addtion to the previously supported Docker, Garden runc and Mesosphere UCR containers.
Instana now supports monitoring of Azure SQL Logical Server. Documentation can be found here.
Instana now supports monitoring of Azure SQL Elastic Pool. Documentation can be found here.
With this release, log message of level WARN or higher are now reported to your Instana dashboard. You can even drill-down into the stack trace if one was reported with the log.
With recent changes to our monitoring of the AWS services you can now specify deeper which and how often you want to poll AWS’s API for metrics. To further configure the Instana Agent, just jump to the Agent configuration file and add the sections documented here.
Instana now automatically discovers and monitors Amazon MQ instances. Read more about our AWS support
With release 150, Instana provides a unified settings page for Events. The event settings define the conditions triggering issues and incidents. This replaces the old built-in/custom issues and custom rules settings.
Beside creating an Alert configuration that applies to specific event types, it is now possible to be alerted on specific events which can be selected individually. This reduces the need of having to specify a complex Dynamic Focus query to filter down to the events of interest.
The redesigned settings page to configure an Alert Channel, formerly known as an Integration, lists the configured Alerts where this Alert Channel is referenced. This helps to figure out which Events can be sent to each respective Alert Channel, such as email, OpsGenie or Slack.
More details about the new configuration pages for Events & Alerts can be found in the documentation here.
Intermediate spans sent to the backend via SDKs (Trace Web SDK, Java SDK, etc.) as well as Jaeger, Zipkin and OpenCensus spans that do not carry a
span.kind tag are used to create
internal calls. Internal calls inherit applications, service and endpoint from their parents, e.g., HTTP calls, to ease the search in context of particular usages. Also, tags of internal calls are searchable via the
IMPORTANT: Internal calls is currently been rolled out to a few customers selected on the basis of use-cases. If you are interested in trying the functionality, reach out to your customer success representative or open a support ticket. General rollout is expected with release 151.
aiohttp is an asynchronous HTTP client/server framework for asyncio and Python. With the latest Python sensor release, we now provide comprehensive tracing of aiohttp server and client calls.
Instana now automatically discovers and monitors ActiveMQ Artemis instances. Read more about our ActiveMQ Artemis support. By supporting ActiveMQ Artemis, Instana is supporting JBoss MQ version 7 and later.
Instana now automatically discovers and monitors Google Cloud SQL instances. Read more about on our Google Cloud SQL support
Instana now automatically extracts information about the most expensive queries on your Microsoft SQL-Server installations and displays them for further investigation on the dashboard. Read more about our Microsoft SQL Server support
Instana now automatically discovered and monitoring AWS Beanstalk applications (environment and its instances). Documentation can be found here
Instana now supports the backend version of Grafana 6.0
- We have introduced notifications that will alert you when there are updates to the user interface. The notifications will prompt you to refresh your page to get the latest version of Instana loaded to your browser.
- Performance counters are now available on Windows hosts. Please find detailed information here.
Sensitivity for deprecation markers in the website monitoring area was reduced. This is a response to deprecations still occurring in less than 0.01% of cases.
JBoss 4: Thread pools & DB connection pools
JBoss AS: Added flag to disable EJB deployments monitoring
The settings pages got a major overhaul.
Use “Inbound Calls” label on Application and Service dashboards to clarify that only calls coming from outside of the application or service are counted.
Statsd Sensor: Allows to configure a bind-ip instead of listening on all interfaces.
Java Tracing: Supports MongoDb 3.8+ drivers and Play Stream Handlers
Etcd Sensor: Improved disk latency calculation and SSL certificate support
Kubenetes Sensor: Less memory usage and no longer exclude the agent pod
Cassandra Sensor: Support DataStax Enterprise 6.x
HAProxy Sensor: Non-worker parent processes are no longer being shown
Azure SQL Database: Databases are shown as hosts. vCore-based resource monitoring is supported.
Prevent saving of custom issues and alerting configurations with invalid Dynamic Focus query
Agent Management: Display sensors list
DynamoDB: Calls to DynamoDB are mapped to the “dynamodb” service and to endpoints which get their names from table names
ElasticSearch: Calls to an ElasticSearch node are mapped to a service named after the cluster instead of the node.
Host: Added open files metrics
MongoDB: Replica set grouping by replica set name and primary member name
Python: Use /proc/net/route to determine default gateway
Python: Use unsigned base 16 IDs internally & update tests
Python: API Client improvements
Python: Send Server-Timing in response headers for EUM correlation
Python: Centralize boot logic and include singletons
Python: Only update working_set when loaded into live process
Process: Added open files metrics
Spark application: Added support for 2.3.x and 2.4.x versions of spark applications
AWS EBS: Links to EBS dashboard(s) can be found in ec2 section on host dashboard
Filtering of incidents with Dynamic Focus: now when incidents are filtered with Dynamic Focus queries only the entity affected by the triggering issue (labeled as “triggering event” in the event list of the incident) is being matched by the Dynamic Focus query. This is the same entity which is shown in the “on” column of the incidents table. With this change negating a term in a Dynamic Focus query will filter out all incidents for which the entity affected by the triggering event is matching the negated term regardless of other events and affected entities correlated into the incident.
Go: Span Kind cleanup and improvements
Go: Change version tagging schema to support Go modules
Python: Improved JSON encoding, Kind handling & SDK span support
Python: Asynqp instrumentation Improvements
Python: New and updated examples; Add another asynqp test
Improve Custom metrics formatting (StatsD, Micrometer, Dropwizard)
- AWS CloudWatch metric drops fixed.
- Fixed an issue which had caused event/issue/incident details to either take long to load or load not at all in some cases.
- Fixed an overlap in two identical events on the same entity when the events are triggered one after another
- Fixed false sudden increase/drop detections in certain edge cases of bimodal metric value distributions
- Python: Respect
INSTANA_SERVICE_NAMEenv var consistently
- Python: Capture and scrub outgoing http query params
- Python: Assure Response headers are returned under all conditions
- Python: Fix case when GC is disabled; Add more safeties
- Ruby: Respect custom service name in in traces
- Application configuration:
call.nameand tags prefixed with
log.are removed from the list in the “Add Tag” dialog because they do not have any effect. If they are used in an existing application configuration, they can be removed manually.
- Kubernetes / OpenShift Agent installation: Updates to the YAML file for creation of the Agent DaemonSet (and similarly the Instana Agent Helm Chart) to make leader-election more robust by basing it on Pod-names instead of the hostname. See the links with updated YAML files for Kubernetes or for OpenShift. Customers using the Helm Chart should pull in latest changes.