Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Hello. I Am Looking For Some Input On How I Could Deal With A Triggerscheduler That Should Run All The Time As A Service. I Run Clearml Inside Kubernetes (Server, Agents And Workers). The Problem I Want To Overcome Is Making Sure The Triggerscheduler Tas

Hello. I am looking for some input on how I could deal with a TriggerScheduler that should run all the time as a service.

I run ClearML inside kubernetes (server, agents and workers). The problem I want to overcome is making sure the TriggerScheduler task is able to self-heal so to say. At the moment if the k8s worker pod running the TriggerScheduler task dies for any reason, the task will show as "running" for a while until it gets aborted because of timeout. Since this task is supposed to run as a service I should not have to manually restart it every time something happens with the pod running it.

Is there any option to automatically restart the task if it fails/gets aborted?

Posted one year ago
Votes Newest