Hello guys.
We have 2 Rancher clusters (one for test/stage and one for production) for a few systems in my job.
Both clusters are configured to send logs to an ElasticSearch instance on AWS.
The problem is that on the test cluster, the logs works perfectly, and both containers and rancher logs are sent to the ES instance, but on the cluster for production there is one workload that is not sending logs to ES, and also the rancher logs are not being sent.
The logs are generated in /var/log/containers on the workers without any issue.
Anyone knows what could be happening?
Best regards
Hello Guys, sorry for reactivating this Thread.
I re-installed the cluster, and it was working until now, same issue: no logs sent to Elasticsearch.
But now, I noticed that the pod “Log aggregator” in the namespace “Cattle-logging” was terminated, and in fact, the namespace is stuck in a removing state.
No one in my work touched that (in fact, I’m the only one who has access to the kubernetes cluster here), so I don’t know how it could happen.
Is there a way to recreate it? I looked up for a yaml to hope to recreate that pod, but had no luck.
Best regards!
did u find any solution for this ?