Hello guys! I need some light on a problem. Thanks...
# troubleshooting
f
Hello guys! I need some light on a problem. Thanks in advance if you can share your 2 cents here. We're using Flink with FlinkOperator v1.11 in our K8s clusters (dev, qa, uat, prod). In dev cluster, the FlinkDeployments just stopped creating the configMaps with the configuration files (Flink config and pod template) that are mounted in the pods - then the deployment fails as it cannot mount the files. This started all of a sudden, the stack trace in FlinkOperator logs (in DEBUG) are insufficient and we couldn't pinpoint anything further. Tried to upgrade FlinkOperator to v.1.12 with no success. The codebase is the same for all K8s clusters, just changing some values regarding databases and resources to connect to. We rebuilt the dev K8s cluster and it worked fine for the extent of a morning, then started doing it again. Any ideas?
Hi folks, regarding this issue we just found that a third-party malfunctioning webhook was timeouting the creation of "service" resources thus breaking FlinkOperator. We fixed the webhook and things got back into track. Thanks again!