I will pull the helm command in a moment… we have modified this slightly because we are deploying dgraph to two namespaces on the same cluster.
were seeing the following in the control plane logs:
6m24s Warning Evicted pod/dgraph-dev-dgraph-alpha-1 The node was low on resource: memory. Container dgraph-dev-dgraph-alpha was using 4484260Ki, which exceeds its request of 100Mi.
As you can see in the chart above, Dgraph is consuming ~7gb and then appears to be getting evicted. We’re going to try setting a hard limit of 6gb which perhaps will help kubernetes properly schedule the pod and prevent evictions.