You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
Helm repository management features like adding, removing or updating repositories do not work in HA setup. The node that gets the request can only perform the required repo management action locally on its own filesystem. Unfortunately there is no way to use another backend (like a database) that would work in a HA scenario seamlessly.
Describe the solution you'd like to see
I would like to disable repo management options in a HA scenario for the short term until figure out whether we want to keep it around in the long term or not.
Approach:
create a configuration flag that would disable the API endpoints in question (or make them raise an error if called)
add a feature flag for clients to understand that the repo management features are disabled
Describe alternatives you've considered
We've considered creating proxy endpoints that would know about all the running pipeline instances and would fanout an incoming request to all backends. We are hesitant to implement it until we get feedback on whether anybody uses this combination or not (HA + Helm repo management features)
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
Helm repository management features like adding, removing or updating repositories do not work in HA setup. The node that gets the request can only perform the required repo management action locally on its own filesystem. Unfortunately there is no way to use another backend (like a database) that would work in a HA scenario seamlessly.
Describe the solution you'd like to see
I would like to disable repo management options in a HA scenario for the short term until figure out whether we want to keep it around in the long term or not.
Approach:
Describe alternatives you've considered
We've considered creating proxy endpoints that would know about all the running pipeline instances and would fanout an incoming request to all backends. We are hesitant to implement it until we get feedback on whether anybody uses this combination or not (HA + Helm repo management features)
The text was updated successfully, but these errors were encountered: