[jira] [Closed] (FLINK-33803) Add observedGeneration to Operator's status spec

["Gyula Fora (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FLINK-33803?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Gyula Fora closed FLINK-33803.
------------------------------
    Fix Version/s: kubernetes-operator-1.8.0
       Resolution: Fixed

merged to main 0fa78d350bd4a6cf2ea5e5bca0880092104355d8

> Add observedGeneration to Operator's status spec
> ------------------------------------------------
>
>                 Key: FLINK-33803
>                 URL: https://issues.apache.org/jira/browse/FLINK-33803
>             Project: Flink
>          Issue Type: Improvement
>          Components: Kubernetes Operator
>            Reporter: Ryan van Huuksloot
>            Assignee: Ryan van Huuksloot
>            Priority: Minor
>              Labels: pull-request-available
>             Fix For: kubernetes-operator-1.8.0
>
>
> Thoughts on adding the observedGeneration status field to the Kubernetes Operator?
>  
> I saw this issue https://issues.apache.org/jira/browse/FLINK-30858 which is useful but the traditional thing would be to just update the observedGeneration when the reconciliation is complete. There is [tooling]([https://github.com/Shopify/krane/blob/main/README.md#specifying-passfail-conditions]) out there that requires this spec in order to determine if the reconciliation passed for CRDs. Not only that, but it is the expected flow: [https://alenkacz.medium.com/kubernetes-operator-best-practices-implementing-observedgeneration-250728868792]
>  
> I think we keep the current implementation of the reconciliation spec for now but at the same time, we can update the `status.observedGeneration`?
>  
> Mailing List Discussion: https://lists.apache.org/thread/1dv33dyqqh18qncot1c6lxgn02do6xnw



--
This message was sent by Atlassian Jira
(v8.20.10#820010)