Hi, we have an instance of ELK on version 7.17.9, orchestrated on Openshift 4.11 using ECK Operator 1.9.0.
For some reason the Certified ECK Operator for Openshift stopped upgrading itself. When Operator 2.0.0 came out I remember was needed some steps to upgrade it to this version, because the CustomResourceDefinitions (CRD) changed, or because previous Operator was not using a bundled version with CRDs.
When we migrated the Development, we got some problems, we unlink the elasticsearch and kibana from the operator, removed the operator 1.9.0, and installed the 2.1.0, which was the newest back there, and linked the elasticsearch and kibana back to the operator. Doing this I think the elasticsearch user and group have changed inside the Openshift, and to put the cluster back online I was needed to insert a initContainers inside the elasticsearch yaml code:
initContainers:
- command:
- sh
- ' -c'
- chown -R elasticsearch:elasticsearch /usr/share/elasticsearch/data
name: chown
securityContext:
privileged: true
runAsUser: 0
I don't know if this was needed because the Operator or because something related to Openshift.
Well, back to my main question, how to upgrade from 1.9.0 to 2.7.0 on OpenShift 4.11(in a few weeks we will upgrade 4.12), what steps to take.
PS: I also like to ask if the Certificated ECK Operator for Openshift is bundled with all CRDs or if is needed to install the CRDs, as described in the documentation of ECK
Thanks in advance!