-
Notifications
You must be signed in to change notification settings - Fork 243
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
Proposal: Enhance echo component for better insights #346
Comments
Components such as
|
We have built a new feature called live debugging, which will cover this use-case very soon. Some of it is already available in our |
How to do that ? |
Background
The
loki.echo
component is useful, however, if you're using a bunch of modules as an example. You don't have a way to debug or get insights into that very easily, without downloading all of those modules and changing them. For example take say that https://github.com/grafana/agent-modules/blob/main/modules/kubernetes/logs/all.river is my config. To add aloki.echo
and filter the results to a specific pod/service while still sending my logs to loki. I would have to do something like:While I would see these in the console output from
loki.echo
filtered to the messages I want, I would still need to go to the console, view the logs, etc.Proposal
Add a new optional argument to
loki.echo
ofselector
that can execute a LogQL selector just like the componentloki.process -> stage.match
which will filter the logs that get echoed.Additionally, from the UI it is not very easy to get insights into what is going on within the
loki.process
, where at least forprometheus.scrape
components, you can see the arguments that came in from service discovery but also each target that was scraped, the set of labels associated to that scrape from service discovery, when it was last successful, etc.To further enhance the UI experience for logs, leverage web sockets, so that anytime a message is sent to
loki.echo
it will write to both the console and publish that same message to the web socket, so that in real-time from the browser, I can see the messages coming in. This would allow for easier debugging and insights, viewing of what labels are associated with the log message, etc.This would enhance metrics as well by introducing a
prometheus.echo
that allows you to see the exact metrics that are being written with the combined service discovery labels plus the scraped labels and be able to filter the results down based on a PromQL label selector of what is echoed directly from the console or the UI.The text was updated successfully, but these errors were encountered: