search

canonical / s3-integrator

An integrator charm for handling s3 credentials
https://charmhub.io/s3-integrator
Apache License 2.0
2 stars 2 forks source link

[Enhancement Proposal] Enable setting multiple bucket names for each `S3 integrator` app #48

Open michaeldmitry opened 2 months ago

michaeldmitry commented 2 months ago

Steps to reproduce

  1. Deploy s3-integrator, loki-coordinator-k8s, mimir-coordinator-k8s
  2. juju run s3-integrator/leader sync-s3-credentials access-key=<accesskey> secret-key=<secretkey>
  3. juju config s3-integrator endpoint=<minIO_endpoint>
  4. juju integrate s3-integrator loki-coordinator-k8s:s3 first (Note that: the loki charm passes loki as a bucket name in requirer app data)
  5. Then, juju integrate s3-integrator mimir-coordinator-k8s:s3 (Note that: the mimir charm passes mimir as a bucket name in requirer app data)
  6. jhack show-relation s3-integrator loki-coordinator-k8s
  7. jhack show-relation s3-integrator mimir-coordinator-k8s

Expected behavior

When multiple applications relate to the same s3-integrator while providing different bucket names from their requirer side (without setting bucket config option on the S3 integrator provider side), calling get_s3_connection_info from loki-coordinator-k8s should return {"bucket": "loki"} and calling get_s3_connection_info from mimi-coordinator-k8s should return {"bucket": "mimir"}

Actual behavior

get_s3_connection_info from loki or mimir returns {"bucket": "loki"}

Versions

Operating system: Ubuntu 22.04.4 LTS

Juju CLI: 3.4.5-genericlinux-amd64

Juju agent: 3.4.5

Charm revision: latest/edge rev: 41

Log output

Juju debug log:

Additional context

https://github.com/canonical/s3-integrator/issues/49

syncronize-issues-to-jira[bot] commented 2 months ago

Thank you for reporting us your feedback!

The internal ticket has been created: https://warthogs.atlassian.net/browse/DPE-5389.

This message was autogenerated