Closed melvin-suter closed 1 year ago
@melvin-suter It would be incredibly hard to troubleshoot why Elasticsearch isn't functioning in your specific environment without much more details, but the one thing that does stand out is storageClassName: nfs-rwo
. Are you using an NFS
backend to run Elasticsearch? What storage system is backing this?
If you could try and start a test cluster with nothing that couldn't be shared publicly, and run https://github.com/elastic/eck-diagnostics against it, and upload them here, we could try and assist further.
Also look through these old issues that could be relevant:
https://github.com/elastic/cloud-on-k8s/issues/3877 https://github.com/elastic/cloud-on-k8s/issues/3701 https://github.com/elastic/cloud-on-k8s/issues/3293
Do any of the initContainers
show any errors?
A few things odd in the logs:
Skipping security auto configuration because the configuration file [/usr/share/elasticsearch/config/elasticsearch.yml] is missing or is not a regular file
which seems to suggest that there is no Elasticsearch configuration file.
{"message":"WARNING: [...] To make full use of MMapDirectory [...]","elasticsearch.node.name":"elastic-es-default-0","elasticsearch.cluster.name":"elastic"}
while node.store.allow_mmap
is false
in your manifest.
I am closing this due to inactivity. If you want to reopen, could you verify that the manifest you provided is the one actually deployed? Thanks.
Bug Report
What did you do?
I installed ECK without any issues. (Kibana for instance works fine.) I used this config to create an elasticsearch instance:
What did you expect to see?
A running and ready elasticsearch pod.
What did you see instead? Under which circumstances?
The pod starts and is shown as "running", but I can't seem to get it to work. The logs of the pod get filled with this entry:
If I open a shell inside the pod I cant
curl 172.0.0.1:9200
.Environment
ECK version:
2.6
Kubernetes information:
I'm running a RKE2 on premise cluster.
v1.24.8+rke2r1