$ kubectl get po --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
[...]
amazon-cloudwatch fluentd-cloudwatch-7fncs 0/1 CreateContainerConfigError 0 4m26s
[...]
$ kubectl -n amazon-cloudwatch describe po fluentd-cloudwatch-7fncs
[...]
Warning Failed 2m40s (x8 over 3m53s) kubelet, ip-192-168-95-53.ap-northeast-1.compute.internal Error: Couldn't find key logs.region in ConfigMap amazon-cloudwatch/cluster-info
[...]
After
$ kubectl get po -n amazon-cloudwatch
NAME READY STATUS RESTARTS AGE
[...]
fluentd-cloudwatch-7fncs 1/1 Running 0 31m
Follows up on #3.
Why?
Fixes:
Besides, this is what is generated by the command mentioned in the Amazon documentation: https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/Container-Insights-setup-logs.html#ContainerInsights-install-FluentD
See also: https://github.com/weaveworks/eks-gitops-example/pull/3/files#r317047875
Before
After