Configurable Cluster Domain about kubernetes-elasticsearch-cluster HOT 8 CLOSED

Did you try this image and it didn't work? At first I thought this was a limitation but after doing a little research on the issue I am under the impression this just works!

from kubernetes-elasticsearch-cluster.

yeah I tried using this image and i haven't seen anywhere in the code where this is referenced so im not sure why it doesn't work but in the ES container logs its trying to connect to the wrong cluster_domain. This could possibly be a kubernetes bug? anyway i think we are just going to change the cluster domain back to default

from kubernetes-elasticsearch-cluster.

@jimmidyson is there any limitation with the discovery plug-in (inherited by fabric8 Kubernetes API)?

@feelobot let's keep this open until Jimmi gets back with an answer.

from kubernetes-elasticsearch-cluster.

There's nothing hard coded in the plugin. Can you share logs? Could be a kubernetes bug in DNS resolution. I'd always suggest using the short DNS name rather than FQDN as the configuration cluster domain will be added as a search suffix in all pods in the cluster if configured.

from kubernetes-elasticsearch-cluster.

@jimmidyson good catch, it may actually be the container DNS misbehaving. @feelobot care to copy the container /etc/resolv.conf contents and paste them here?

from kubernetes-elasticsearch-cluster.

resolv.conf

/ # cat /etc/resolv.conf
nameserver 127.0.0.1 # added by go-dnsmasq
# nameserver 10.100.0.10
# nameserver 172.16.0.2
search default.svc.cluster.local svc.cluster.local cluster.local ec2.internal
options ndots:5

from kubernetes-elasticsearch-cluster.

also getting this now that ive switched to using cluster.local

[2016-02-08 23:12:14,663][WARN ][io.fabric8.elasticsearch.discovery.kubernetes.KubernetesUnicastHostsProvider] [Forbush Man] Exception caught during discovery: Error executing: GET at: https://kubernetes.default.svc/api/v1/namespaces/default/endpoints/elasticsearch-discovery. Cause: sun.security.validator.ValidatorException: PKIX path building failed: sun.security.provider.certpath.SunCertPathBuilderException: unable to find valid certification path to requested target
io.fabric8.kubernetes.client.KubernetesClientException: Error executing: GET at: https://kubernetes.default.svc/api/v1/namespaces/default/endpoints/elasticsearch-discovery. Cause: sun.security.validator.ValidatorException: PKIX path building failed: sun.security.provider.certpath.SunCertPathBuilderException: unable to find valid certification path to requested target

from kubernetes-elasticsearch-cluster.

search default.svc.cluster.local svc.cluster.local cluster.local ec2.internal means you're using cluster.local. Try it with a different domain and let me know here.

Regarding the second error, it seems your Kubernetes install doesn't provide tokens to pods.

from kubernetes-elasticsearch-cluster.