Build 161

2019-07-19

On-Premises Release Notes

For an upgrade to the new version please follow the Packaged Based Operations manual.

Note: We changed the required users and groups:

+ - instana-accountant:instana-accountant

Cumulative Features from Release #158, #159, #160, #161

Pipeline Feedback: Release Markers

The new Pipeline Feedback feature, which includes the release markers capability, enables you to visualize, identify, and correlate issues that may occur when a new release is made available to your customers.

release marker charts

A release marker represents a single release that defines a moment in time when a change in code, infrastructure, or architecture becomes generally available. Releases are displayed as annotations on every chart and help you quickly spot the impact of deployed changes concerning health or application performance.

release marker incidents

Further information can be found in the Documentation.

Resource timing support for XHR / Fetch

The JavaScript agent now supports collecting resource timing data, e.g. DNS / TCP / SSL timings, whenever the browser provides them. These new timings are displayed on the website dashboard, analyze, and the page load view.

Improved Backend Correlation for Resources

Improved backend correlation to now support retrieved resources whenever possible. Previously, the backend correlation was restricted to website page loads and usages of the XMLHttpRequest and fetch API.

Configurable Table columns

Within the Kubernetes Pods table, selecting the gear icon on the last table column enables you to customize each table view by showing or hiding the table columns that you consider essential or not. The configuration is stored per-user level.

configurable pod table

Node.js support for NATS and NATS streaming tracing

Tracing messages published to and consumed by NATS and NATS streaming, are now supported for the Node.js runtime.

Cloud Foundry improvements

When working with Cloud Foundry environments, the following improvements have been made:

  • Grouping containers on the map based on Application or Space.
  • Searching for containers using dynamic focus queries based on CF Application, Space, or Org.
  • AP Service naming based on the CF Application name tag.

New call tags

  • In the Analyze view, traces and calls can now be grouped and filtered by runtime and process.name.

VictorOps integration

We have added an out-of-the-box alerting integration for VictorOps.

Redis cluster

Support for Redis running in cluster mode. Read more.

Custom events

An additional system rule has been added in order to ensure the presence (and be notified in the absence) of entities on a host. This rule can be found under “System Rules” with the name “Hosts that do not have matching entities running on them”.

Micrometer custom metrics health events

  • Instana now supports health events based on Micrometer custom metrics.
  • For existing Grafana dashboards using Micrometer builtin metrics - please switch to Infrastructure custom metrics in order to retain dashboard functionality.

Solr cloud

  • Instana now supports discovery and monitoring of Solr cloud. Read more.

AWS Elastic MapReduce (AWS EMR)

New discovery and monitoring of AWS Elastic MapReduce (AWS EMR). See more information here.

Linux Containers (LXC)

Instana now supports discovery and monitoring of Linux Containers (LXC). See more information here.

LXC dashboard

Windows services

New monitoring of Windows Services and its child processes. How to add services to be monitored is described here.

Expose all metrics via the API

Expose all metrics via the API for the following technologies:

  • Azure SQL Database
  • Azure SQL Elastic Pool
  • Finagle
  • JBoss AS - Wildfly
  • Jenkins
  • Microsoft IIS
  • Mule ESB

Other

  • Python sensor: New mysqlclient package instrumentation & visibility
  • Python sensor: New PyMySQL package instrumentation & visibility

Improvements

  • Service mapping: improved the mapping between calls to Oracle and monitored Oracle instances, leading to better service names.
  • Service mapping: use header SOAPAction as service name when present. Please note the (required agent configuration to capture this header)
  • ContainerD and Garden containers are now supported in the application infrastructure tab.
  • Trace Detail View: Display more specific names for Ruby render calls.
  • Sybase SQL Anywhere health rules for server connection and database count. Read more about our Sybase SQL Anywhere Health rules
  • Activity timelines within the website monitoring page load view will now adjust end-user timestamps when the end-user timestamp combinations are impossible, e.g. activity before the page is even loaded.
  • Sudden drop detection: improved the detector to not create false positives in case the Application, Service or Endpoint receives periodic infrequent calls.
  • It is now possible to link multiple Kubernetes services to multiple logical Instana services. See Kubernetes for more details.
  • For Jaeger and Zipkin, there is increased support for spans of the type “Messaging”.

Fixes

  • PHP Trace Sensor: properly resolve _default_ host names in Apache configs to the host IP address
  • PHP Trace Sensor: performance improvements when parsing Sql queries
  • Dynamic Focus: fixes a potential bug that a DFQ containing an application/service/endpoint term (e.g. entity.service.name) could contain results from other entities outside of the defined scope which are running on the same host.
  • Alerting: alerts for an Incident don’t create a redundant alert for the underlying Issue anymore that triggered this Incident.
  • Kubernetes Cluster ComponentStatuses: Sometimes Kubernetes reports wrong health status for components. This results in incorrect issues + alerts. We are now trying to filter out these ComponentStatuses from creating incorrect alerts. For more information see Kubernetes Health Rules.
  • Application Perspectives: Fix possible linking to infrastructure for SDK spans.
  • Application Perspectives: unmonitored destinations (e.g. databases) that are called by their Kubernetes Pod IP are now linked to infrastructure.
  • Filtering and grouping by beacon metadata keys are now case insensitive.
  • The analyze filters is present and is not present are now case insensitive for HTTP header keys and beacon metadata keys.
  • Fixes the problem when Incidents were not created when there was a spike in the number of Kubernetes events.
  • Kubernetes events are now not displayed as Change events in the Events Timeline anymore.
  • Docker Agent image now correctly allows version pinning via the INSTANA_AGENT_UPDATES_VERSION ENV variable.
  • PostgreSQL database list was updated as soon as a database was added or removed.

Deprecations

  • Analyze: port tags (mysql.port, redis.port, etc.) are removed