Hello, I’m using Loki 1.6.1 and getting the error “failed to flush user” and “err=timeout”. I’m also getting “error querying storage”. I read on slack, that this means that the chuck storage save failed. Is this right? I’m using NFS storage provided by Kubernetes. I’m running Loki in Kubernetes. Can any one provide some direction to debug and resolve this problem?
I figured it out. I was using a filesystem storage with a ReadWriteMany Kubernetes PV accross three replicas. When I reduce the replicas to 1, the problems go away. If want to use more than 1 replica, I need to use another storage method. Cassandra is the only non cloud based solution.