Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Question] What happened to headscale_last_update_seconds ? #1935

Closed
2 tasks done
n0valis opened this issue May 12, 2024 · 3 comments
Closed
2 tasks done

[Question] What happened to headscale_last_update_seconds ? #1935

n0valis opened this issue May 12, 2024 · 3 comments
Labels
enhancement New feature or request

Comments

@n0valis
Copy link

n0valis commented May 12, 2024

Use case

In previous version this was a metric.

Description

I had this metric integrated in my monitoring. After update to alpha version it disappeared.
Is this on purpose?

Contribution

  • I can write the design doc for this feature
  • I can contribute this feature

How can it be implemented?

No response

@n0valis n0valis added the enhancement New feature or request label May 12, 2024
@kradalby
Copy link
Collaborator

Yes! That metric was very deeply integrated to our old and very naive way of updating the client.

It would essentially check if anything had happened every X seconds and then send an update to every node. That metric was a reflection of when the last update had been sent and I originally added it while I was debuting a deadlock issue where it would stop sending updates.

In the new alpha, we have rewritten the entire update distribution and updates are now sent directly to a node when the update comes in, and it's sent on a per node basis.

This metric therefore does not have any meaning since the whole mechanism setting it is gone. There is now a bunch of new (and in my opinion more meaningful) metrics available.

@n0valis
Copy link
Author

n0valis commented May 27, 2024

I used the metric to check on the server that handshakes from clients came in on a regular basis. So I would knew - only by checking this value - that the whole network was intact. Had a nice Prometheus Grafana visualisation for this.

You say you have more meaningful) metrics available.
Which would that be to accomplish this task?

@kradalby
Copy link
Collaborator

I used the metric to check on the server that handshakes from clients came in on a regular basis. So I would knew - only by checking this value - that the whole network was intact. Had a nice Prometheus Grafana visualisation for this.

That assumption is not correct, you could not tell from that metric if the network was intact, it would only indicated that some update was sent to at least one node.

The most relevant would probably be https://github.com/juanfont/headscale/blob/main/hscontrol/notifier/metrics.go#L50, which indicate how many nodes are connected and able to receive updates.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants