I’m trying to create a simple docker elk stack to run in Rancher.
All of the containers should run on the same host, so that’s what I’m
setting the host_label.
The reason for my post here is that I can run this with docker-compose, on the same VM Host, and
kibana comes up and connects to elasticsearch. But in Rancher, on the same VM Host,
I get the following error:
kibana_1 |
log [16:11:16.888]
[error][status][plugin:elasticsearch] Status changed from yellow to red -
Request Timeout after 1500ms
I’ve tried running this docker-ocmpose file with the rancher-compose CLI and as a custom catalog entry and the results are the same.
I did try a little guessing from looking around in your ELK catalog entries, but they are much more complex (and I’m sure very well thought out), but I just want something very simple.
Please, advise…
====== docker-compose.yml =========
elasticsearch:
image: elasticsearch:latest
ports:
- "9200:9200"
- "9300:9300"
expose:
- "9200"
- "9300"
volumes:
- "/var/esdata:/usr/share/elasticsearch/data"
labels:
io.rancher.container.pull_image: always
io.rancher.scheduler.affinity:host_label: ${HOST_LABEL}
io.rancher.container.hostname_override: container_name
kibana:
image: kibana:latest
command:
- kibana
ports:
- "5601:5601"
expose:
- "5601"
links:
- elasticsearch:elasticsearch
environment:
ELASTICSEARCH_URL: "http://elasticsearch:9200"
labels:
io.rancher.container.pull_image: always
io.rancher.scheduler.affinity:host_label: ${HOST_LABEL}
io.rancher.container.hostname_override: container_name
logstash:
image: rcdn6-vm67-9.cisco.com/dot-logstash
command:
- /docker-entrypoint.sh
- -f
- /etc/logstash/conf.d/*.conf
ports:
- "5000:5000"
expose:
- "5000"
links:
- elasticsearch:elasticsearch
labels:
io.rancher.container.pull_image: always
io.rancher.scheduler.affinity:host_label: ${HOST_LABEL}
io.rancher.container.hostname_override: container_name