-
Notifications
You must be signed in to change notification settings - Fork 4
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
Merge with https://github.com/simonpasquier/prometheus-scaleway-sd #2
Comments
Hello @simonpasquier, Thanks for you review. What you have done is awesome.
As you already gave a lot of your time contributing to a service discovery for Scaleway, I'm not very confortable asking you to take some time again to prepare the MR. If you are OK with that, it would be perfect. What do you think ? Do you agree with option 2, or do you have some time to submit a MR ? Thanks again, |
Option 2 is fine for me. No problem for reviewing your PR and helping the project in the longer term. |
Hello @simonpasquier, can you review #3 ? We credited you on the Readme. If this is ok to merge, we will create a new release after merging. |
…us-scaleway-sd-#2 Merge with simonpasquier prometheus scaleway sd #2
This is a meta-issue describing how we could merge https://github.com/simonpasquier/prometheus-scaleway-sd into this repository. I'm more than happy to see this happening!
I had a quick look at the code and here are my initials remarks.
Dealing with stopped/deleted servers
IIUC this project doesn't update the inventory when servers are stopped, deleted, etc. This could leave stale targets in the output file. In
prometheus-scaleway-sd
, I'm keeping track of the servers at each run so the SD application can notify servers that don't exist anymore.https://github.com/simonpasquier/prometheus-scaleway-sd/blob/master/main.go#L217-L224
Vendoring
I see that this repository doesn't vendor any dependency delegating it to the developer (cf. README.md).
I see a few problems here:
github.com/prometheus/prometheus/discovery/manager
is copied in the repository. My guess is that it is to avoid pulling all SD dependencies (eg AWS, GCP, ... packages) but it also means that any fix to themanager
package will have to be manually backported here.Logging
prometheus-scaleway-sd
has a wrapper aroundgo-kit/log
to grab logs from the Scaleway API client. This may be useful for troubleshooting.https://github.com/simonpasquier/prometheus-scaleway-sd/blob/master/main.go#L117-L146
Instrumentation
prometheus-scaleway-sd
exposes 2 metrics:prometheus_scaleway_sd_request_duration_seconds
, histogram of latencies for requests to the Scaleway API.prometheus_scaleway_sd_request_failures_total
, total number of failed requests to the Scaleway API.Thanks to the
prometheus/client_golang
, it also expose process metrics.Security
As noted here, the application shouldn't get the token from the command line, reading it from a file is fine.
Release tooling
I've integrated with promu because it eases the generation of binaries across multiple platforms.
Labels
They look more or less the same. No issue there.
The text was updated successfully, but these errors were encountered: