search

mittwald / kube-httpcache

Varnish Reverse Proxy on Kubernetes
MIT License
294 stars 71 forks source link

Is it possible to share the memcache data between multiple pods to run Varnish in HA mode? #245

Closed vickylance closed 2 months ago

vickylance commented 2 months ago

I want to run varnish in HA mode for horizontal scaling. But I want the cache to be persistent across the pods so that if the request goes to 1 varnish pod or the other it will be able to still serve the cache if its available. And the cache wont be on per pod level. To achieve this is it possible to do it with memcache or some other backend storage?

martin-helmich commented 2 months ago

AFAIK this is not something that Varnish supports (see the list of supported storage backends in the documentation). To achieve redundancy, you'd need individual cache instances with independent (per-pod) storage backends.