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
Hi,
thanks for this really nice project. It solves some hard issues in an elegant way.
Currently its only possible to specify a single backend via -backend-service=backend-service. Yes this backend service can have multiple replicas. But what If you have a single varnish in front of lets say 2 (or more) backends, each with several replicas.
I'd need exactly that: Support for multiple backends.
With "backends", I mean distinct backend applications.
Request --> [ Varnish ] -->
|---> [ Backend Service A]
[Backend Service A POD 1]
[Backend Service A POD 2]
|---> [ Backend Service B]
[Backend Service B POD 1]
[Backend Service B POD 2]
Currently I'd need to launch two kube-httpcache deployments. One for each backend. However, that leads to more overhead and less efficient memory utilization because I cannot share a given memory space for all the backend applications.
The text was updated successfully, but these errors were encountered:
Hey David -- thanks for your feedback; happy to hear that it's useful to you. 🙂
This is a very reasonable suggestion, that I'd very much like to see implemented as well. However, this would probably require some major changes in the internal design of the controller, so I cannot make any promises as to when we'd actually get around to implementing this. PRs are always welcome, though. 🙂
Hi,
thanks for this really nice project. It solves some hard issues in an elegant way.
Currently its only possible to specify a single backend via
-backend-service=backend-service
. Yes this backend service can have multiple replicas. But what If you have a single varnish in front of lets say 2 (or more) backends, each with several replicas.I'd need exactly that: Support for multiple backends.
With "backends", I mean distinct backend applications.
Currently I'd need to launch two kube-httpcache deployments. One for each backend. However, that leads to more overhead and less efficient memory utilization because I cannot share a given memory space for all the backend applications.
The text was updated successfully, but these errors were encountered: