Build 181

2020-07-14

Cumulative Features from Release #180, #181

AWS Fargate Technical Preview

With this release we officially launch Instana’s support for AWS Fargate on ECS as a technical preview.

AWS Fargate is a serverless compute engine for containers that removes the need to provision and manage servers. Without any code change, with Instana you can now seamlessly trace all calls in your applications running on Fargate and collect metrics for all containers in an ECS task, giving you full visibility into your Fargate resources and their performance. Check out our AWS Fargate documentation to get started.

NGINX Tracing is Generally Available

Instana native tracing for NGINX has graduated to General Availability status. Among other things, it greatly reduces the need of custom configuration in the nginx.conf file, down to the very minimum that the NGINX module system will allow us to do. Refer to the NGINX Tracing documentation to get started.

Cisco Webex Teams Alert Channel Integration

It is now possible to use Cisco Webex Teams as an alert channel for Instana alert notifications (see documentation).

Symfony Tracing

The PHP extension v1.8.0 uses the route name as endpoint name to vastly improve user experience in terms of the number of endpoints that will be created for a single route. Where we previously relied on the full URL for endpoint extraction and hence created duplicate endpoints for a single route, we're now using the routes' name to create just one.

Laravel Tracing

The PHP extension v1.8.0 adds support for tracing the Laravel framework and the underpinning Blade template engine.

Much like with our enhanced support for Symfony, we're relying on the route name (if set), or the route scheme for endpoint creation.

W3C TraceContext Support for the PHP Tracer

The PHP extension v1.8.0 adds support for W3C Trace Context.

2FA User List

A full list of all users and a marker if they have 2FA activated can be found under Settings > Authentication > Users.

Profiling Hotspots

The profiles view is now enriched with a summary page. On this page, users can find a list of all profiles which have a self time greater than 5% (at least 5 profiles are shown, even if they don't match this threshold). The profiles can be expanded to the the stack trace up to this call. This allows users to concentrate on the important profiles and easily find out, where the potential issues come from.

Authentication

What could earlier be configured under the sub-portal unit-company.instana.io/ump/company/unit/authentication/ can now be configured in the product itself. A link to your page can be found in the product under More > Management Portal of the main navigation. Please check out the new pages under Settings > Authentication > Google SSO / SAML / LDAP (onprem only).

Vault Secret Manager Integration

Instana host agent can now retrieve confidential data from secret managers such as credentials, that you do not want to put in clear text in the configuration.yaml file. For more information, see the Instana docs.

URL Shortener

We previously added support for a URL shortener behind a shortcut. Today we are making it easier to discover the URL shortener and to generate good shareable URLs via a button within our header.

URL Shortener Overlay

Custom tags

Custom tags are indexed also on spans generated by Autotrace instrumentation, rather than be limited to spans created via the various SDKs spans.

SDK spans

Added support for transforming SDK spans into BATCH spans.

AWS Multiple Accounts Monitoring

Instana now supports monitoring of multiple AWS accunts using a single agent. For more information, see the AWS docs.

Git-based Configuration Management

The Instana host based agent now supports git-based configuration management to streamline configuration rollouts across multiple hosts. For more information, see the Git-based Configuration Management documentation.

Improvements

  • Service dependency graph on AP dashboard will be shown in left to right flow layout by default.
  • The Go sensor as of v1.13.2 supports activation of AutoProfile™ via an env variable (see documentation).
  • The Go sensor v1.14.0 supports monitoring services running in AWS Fargate (see documentation for configuration details).
  • Kafka Connect: Connector list moved to Cluster level.
  • Kafka Cluster: Exposed built-in metrics for API usage. Exposed dynamic built-in metrics for custom events.
  • CPU / Memory and Wait Time profile views now come with a settings button to allow configuring a threshold to filter unimportant profiles. It is set to a default of 1%.
  • Profiles are now enriched with the context guide to allow a better navigation to the corresponding entities.
  • On infrastructure entity dashboards (host, process, container, etc.) users can now find a button to jump to connected profiles directly if available.
  • We have added DaemonSets to the list of supported Kubernetes resources. Please note that in order to see DaemonSet data on your dashboards, you will need to update to the latest version of your Instana Agent installation method of choice. Please see our Kubernetes or OpenShift documentation for further information on the latest versions of the Agent installation methods.
  • Changed priority of predefined service and endpoint rules, so that rules based on call.batch.*, call.database.* and call.messaging.* tags have higher priority than rules based on call.http.* tags. This will help to create more specific service and endpoint names in some scenarios, e.g., when accessing a messaging queue via the HTTP protocol.
  • Smart Alerts: We enabled further configuration in the Time Threshold section to control the metric granularity the detection is based on.
  • The Java sensor now supports naming JVMs using existing system properties.
  • Docker metrics: The total memory usage is shown in percentage.

Fixes

  • Source service labels were not always correctly created for Messaging type spans, resulting in "empty" labels.
  • Monitoring Issue events were sometimes not correctly shown on Host and Agent dashboards, this is now resolved.
  • For Linux hosts, don't stack memory metrics "Available", "Buffered" and "Cached" as this leads to incorrect totals. Also split up Swap metrics for clarity.
  • AWS EMR: Fixed HDFS utilization metric for custom events.
  • Fixed support for the does not start with and does not end with tag operators in Application Perspective and Synthetic Endpoints configurations.
  • Re-introduced support for the starts with, does not start with, ends with, does not end with, contains and does not contain operators in combination with the call.http.status tag, which was accidentally dropped when the tag type was changed from string to number.