Howie

@hagen1778

if this is a bug, i could propose a pr

/assign

if we have multiple vmstorage nodes and one unavaliable, vmselect still query all nodes, which will lead to long query(refer barrel effect)

any plan to suppose distributed trace?

any method like this? `kubectl - wait - --for=condition=Ready `

> Hi @Howie59! Did you manage to fetch a trace for your query? sry, vm is deployed in Intranet

ref: https://github.com/containerd/nerdctl/issues/1274 i changed this code and encountered another problem!! hope for reply