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
I'd like to contribute another backend to do authentication based on IPs that I assemble from our cloud setup. Happy to write the code, I am guessing this would need some kind of "pluggable" auth mechanism.
And then integrate that would some code to e.g. walk through a domain/tenant, assemble IPs based on tags/metadata and populate auth-gateway with the necessary settings. The rules to assign resources from OpenStack to tenants could/should be left for the downstream implementation when people use this. I wouldn't make assumptions about how people organize their setups or what kind of flexibility they need.
We currently have a service in-house which is very similar to auth-gateway. The service acts as an IP-firewall and sits in front of our cortext setup. On the side we walk through different tenants on openstack (on an interval) and assign them to tenants for cortex based on instance metadata or networks resources are attached to.
The data is then used to allow remote write to Cortex.
The text was updated successfully, but these errors were encountered:
I'd like to contribute another backend to do authentication based on IPs that I assemble from our cloud setup. Happy to write the code, I am guessing this would need some kind of "pluggable" auth mechanism.
My goal would be to do something like:
And then integrate that would some code to e.g. walk through a domain/tenant, assemble IPs based on tags/metadata and populate auth-gateway with the necessary settings. The rules to assign resources from OpenStack to tenants could/should be left for the downstream implementation when people use this. I wouldn't make assumptions about how people organize their setups or what kind of flexibility they need.
We currently have a service in-house which is very similar to auth-gateway. The service acts as an IP-firewall and sits in front of our cortext setup. On the side we walk through different tenants on openstack (on an interval) and assign them to tenants for cortex based on instance metadata or networks resources are attached to.
The data is then used to allow remote write to Cortex.
The text was updated successfully, but these errors were encountered: