Kibana 7.8.1. does not start. Stuck on migration notice while the cluster is newly created. #74225
Labels
Team:Core
Core services & architecture: plugins, logging, config, saved objects, http, ES client, i18n, etc
Kibana version:
7.8.1
Elasticsearch version:
7.8.1
Server OS version:
Azure Kubernetes Service 1.16.8 - Ubuntu 16.04
Browser version:
FF 79.0 64bit
Browser OS version:
Windows 10 2004 (19041.329)
Original install method (e.g. download page, yum, from source, etc.):
Kustomize deployment of ELK stack on Azure Kubernetes instance. All 7.8.1 images are default, except Elasticsearch which has the
repository-azure
installed as extra component.Describe the bug:
When deploying an entire new and empty ES cluster via kustomize, we get the following behaviour: Kibana is started together with Logstash and Elasticsearch in one go. Kibana logs will show it is not able to contact the Elasticsearch cluster yet as it's still spinning up. After Kibana is able to contact the ES cluster, it creates both the
.kibana_1
and the.kibana_task_manager_1
index. a few steps later it tells usAnother Kibana instance appears to be migrating the index
. Manual removal of the mentioned index files and deleting the pod (resulting in the replicasets.apps to restart a new one) makes the Kibana pod work as intended.Steps to reproduce:
Another Kibana instance appears to be migrating the index
Expected behavior:
Kibana will start in one go without human intervention.
Screenshots (if relevant):
N/A
Errors in browser console (if relevant):
N/A
Provide logs and/or server output (if relevant):
Any additional context:
The text was updated successfully, but these errors were encountered: