Release Guide

Following are details about the current and planned releases and updates of Stackdriver Kubernetes Monitoring during the Beta period.

Release versions

The following releases and updates of Stackdriver Kubernetes Monitoring exist or are planned. The releases are named by the Kubernetes version that supports them:

  • 1.10.2: This is the original and current Beta release. It also supports Kubernetes 1.11.0, when available. This release has a list of known issues.

  • 1.10.2 agents update: As of late July 2018, updated Stackdriver agents are available to address some of the known issues in the 1.10.2 release.

    You must install the updated agents on newly created Kubernetes clusters using the Manual Installation instructions. For this reason, we have temporarily withdrawn the installation instructions that use the Kubernetes Engine console for release 1.10.2.

  • 1.10.7 / 1.11.2: This planned release should resolve problems with using the Kubernetes Engine console and incorporate the latest agents.

Known issues

This section lists the known issues in the current Beta release versions.

Beta release 1.10.2

Note. Because of the following issues, we recommend that all users install updated Stackdriver agents in newly created clusters by following the instructions in Manual Installation of Stackdriver Kubernetes Monitoring.

These issues are current as of August 10, 2018.

  1. Missing resource entries

    You might not see all your namespaces, workloads, pods, or other entities in Stackdriver Kubernetes Monitoring. This is due to a problem in gathering metadata. It can take up to 30 minutes for the missing entities to be found. They appear after refreshing the display.

    Schedule: Fixed in Kubernetes versions 1.10.7 and 1.11.2..

    Workarounds: Choose either of the following:

    • Use the kubectl get command in a Cloud Shell to list your entities.
    • Use the Manual Installation steps to install newer versions of Heapster and the Stackdriver agents in a new cluster.
  2. Regional and multi-zonal clusters

    Entities belonging to regional and multi-zonal clusters are not displayed correctly. The Infrastructure, Workload, and Service tabs do not expand to the node/pod level for these clusters.

    Schedule: Fixed in Kubernetes versions 1.10.7 and 1.11.2..

    Workarounds: Choose either of the following:

    • Use the kubectl get command in a Cloud Shell to list your entities.
    • Use the Manual Installation steps to install newer versions of Heapster and the Stackdriver agents in a new cluster.
  3. Severity missing in log entries

    Log entries ingested by the logging agent are missing the severity field, which causes all log entries to show up with default severity (a gray star), breaking Error Reporting integration.

    Schedule: Fixed in Kubernetes versions 1.10.7 and 1.11.2..

    Workaround: Use the Manual Installation steps to install newer versions of the agents in a newly created cluster.

  4. Exception stack trace detection is broken

    Multi-line exception stack traces are split into multiple separate log entries, breaking Error Reporting integration.

    Schedule: Fixed in Kubernetes versions 1.10.7 and 1.11.2..

    Workaround: Use the Manual Installation steps to install newer versions of the agents in a newly created cluster.

  5. Irrelevant error messages from agents

    The metadata agent can write irrelevant error log messages after a previous instance of the agent shuts down incorrectly. The messages are:

    terminate called after throwing an instance of 'boost::exception_ ...'
    what():  Error listening on provided port.
    terminate called recursively
    Aborted (core dumped)
    

    Also, the logging agent will log Error calling Metadata Agent after the metadata agent terminates.

    Schedule: Fixed in Kubernetes versions 1.10.7 and 1.11.2..

    Workaround: Use the Manual Installation steps to install newer versions of the agents in a newly created cluster.

  6. Stackdriver agents don't work behind Compute Engine firewalls

    Stackdriver Kubernetes Monitoring does not work with clusters that are behind certain Compute Engine firewalls. In those situations, your metadata agent issues the following error message:

    Metadata request unsuccessful: Address family not supported by protocol.
    

    Schedule: Fixed in the latest agent versions, which will be incorporated into Kubernetes versions 1.10.7 and 1.11.2..

    Workaround: Use the Manual Installation steps to install newer versions of the agents in a newly created cluster.

  7. "Unscheduled pods" are not displayed.

    The Containers UI does not display unscheduled pods. Unscheduled pods can be identified by the following metadata label:

    "node_name" = ""
    

    Schedule: Not yet available. This is a feature addition.

  8. Stackdriver metadata agent restarts

    In the Beta release, the metadata agent can restart once per hour in normal operation, and might restart several times in succession.

    Schedule: Not yet available.

    Workaround: Do not worry about restart counts for the metadata agent.

Was this page helpful? Let us know how we did:

Send feedback about...

Stackdriver Monitoring