TABLE OF CONTENTS
With this release we officially launch Instana’s support for Google Cloud Run in Technical Preview.
Google Cloud Run is a serverless compute engine for containers that removes the need to provision and manage servers.
You can now seamlessly trace all calls in your applications running on Cloud Run and collect metrics for Google Cloud Run service revisions with Instana, without any code change. This gives you full visibility into your Cloud Run resources and their performance. Tracing is currently supported for the Node.js and Go runtime, more runtimes will follow very soon.
Check out our Google Cloud Run documentation to get started.
Instana now supports monitoring of Google Cloud Pub/Sub. This includes a built-in dashboard that displays key metrics about topics and subscriptions. For more information, see the Google Cloud Pub/Sub docs.
- Python: as of version 1.26.0 the in-app sensor instruments Google Cloud Storage clients that use the
google-cloud-storagepackage v1.24.0 and above.
- .NET Core: as of version 1.1.60 of
Instana.Tracing.Core, instrumentation for Google Cloud Storage is provided by instrumenting the Google.Cloud.Storage.V1 package.
Following the deprecation of Helm Hub, we launched a new, shiny, Instana-specific Helm repository at https://agents.instana.io/helm. We will make the sources available soon, so that you can open PRs again. (And you do open cool PRs!) Documentation and in-product guidance is also going to be updated soon.
By the way, if you snoop into that repository, you might find something else very interesting to play with ;-)
Go to Settings > Account & Billing to find your basic account info, the Billing Usage for the entire account and for each unit and the technologies monitored.
- Remote download of log-files from the Agent dashboard
- Infrastructure & Agent view: Added DFQ fields making it possible to search for specific Agents / Hosts based on Agent details such as mode or bootstrap version
- Application dashboards: The Time Shift feature is made available for latency distribution.
- Smart Alerts: Added 15min and 20min evaluation granularity options to allow more tuning options when configuring an alert.
- CRI-O Sensor: The infra (or 'pause') container that runs in each k8s pod is now excluded from monitoring.
- The recently-released PHP extension version 1.12 additionally traces
php-amqp-liband Kohana logger
- Go: as of
cloud.google.com/goinstrumentation package supports tracing calls via the Google Cloud Pub/Sub.
- Application Perspectives: HTTP status code chart becomes available on Application and Service dashboards if the Application or Service has HTTP endpoints.
- HTTPd: HTTPd instance name/label now includes PID and listen address to distinguish multiple instances running on a single host.
- Agent list: Fixed issue where 'offline' Agents were no longer correctly included in the list. Also, correctly show the timestamp when Agent started reporting.
- Elasticsearch: Fixed grouping nodes into a cluster using UUID when it's available.
- JBoss service naming: Fixed service mapping from JBoss deployment names containing special characters.
- Java: Tracing support for Redisson
- Python: Tracing support for Boto3
- Custom Events: The EQUAL condition operator has been changed from
=in order to be compliant with the default code generator of OpenAPI, which prevously generated an identifier that was invalid when using Java 9+. The operator
==will still be accepted by the backend for compatibility.
Python 3.9 has been recently released.
We are aware of regressions in a number of asynchronous instrumentations, like
asyncio, and thus we do not currently support Python 3.9.
Please open a support ticket to inform us of your plans to move to Python 3.9 (and we also would love to hear why), so that we can prioritize support accordingly.