Skip to content

Monitoring is showing '?' for counters when there is one-or-more inactive instance #2854

Closed

Description

@RemyDuijkeren

Describe the bug

Description

I scale up and down my endpoints depending on the amount of msg. Every time there is an inactive instance for an endpoint then Monitoring is not showing numbers for the counters but '?' .

Only when inactive endpoints are auto cleaned up in the INSTANCES tab, the counters show numbers again.

I disable 'Track Instances' for all my endpoints in the Endpoint Heartbeats Configuration.

Expected behavior

It should ignore inactive endpoints and show numbers for the counters in the Monitoring overview.

Actual behavior

It shows '?' for all the ccounters:
Image

When going to the INSTANCEs tab it shows the number for counters correctly for the active instance. I expect this number also in the overview.
Image

Versions

ServicePulse v2.3.2 (but this bug is already for some time, I think when the Track Instances feature was introduced)

Steps to reproduce

Scale an endpoint up with one or more instances and then scale it back to 1.

Relevant log output

no logs

Additional Information

Workarounds

It used to be possible to remove inactive endpoints under the INSTANCES tab, so I could 'fix' it in that way, but this is not possible anymore.

Possible solutions

Additional information

Activity

  1. danielmarbach commented on Feb 26, 2026

    @danielmarbach
    Contributor

    Thanks for reporting @RemyDuijkeren

    We are going to try to reproduce it as part of our bug triage process. We'll keep you informed

  2. danielmarbach commented on Feb 27, 2026

    @danielmarbach
    Contributor

    Hi @RemyDuijkeren . We have reviewed this bug and added it to our bug fix queue. However, there is no ETA for when we will get to this bug as it may be prioritized behind other, more urgent, bugs. We will update this issue when it has a team assigned and work begins.

  3. PhilBastian commented on Feb 27, 2026

    @PhilBastian
    Contributor

    IsStale is being set in ServiceControl based on any of the instances, rather than all of the instances. This setting is used in many places in ServicePulse, so any fix will need to verify all usage points, and also verify behaviour against endpoints of type Track Instances/Do Not Track Instances

  4. jasontaylordev commented on May 29, 2026

    @jasontaylordev
    Contributor

    Hey @RemyDuijkeren. This has been fixed with the latest release of Service Control (6.15). Additional details here; Particular/ServiceControl#5474.
    Thanks for reporting this issue.

  5. RemyDuijkeren commented on Jun 12, 2026

    @RemyDuijkeren
    Author

    @jasontaylordev I installed it yesterdays and looks good so far! Many thanks you guys!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions