Closed basilzuberi closed 1 month ago
The solution was to remove the resources section from helm, and just not have it. I'm not entirely sure why, but it clearly works now.
the following lines were omitted from my helm chart, and it works perfectly now:
resources:
requests:
memory: "3Gi"
cpu: "8"
limits:
memory: "3Gi"
cpu: "8"
any insights to this are appreciated, but at least this MIGHT help someone.
@basilzuberi thanks. But clearly DB creation has failed because of your resource limits. 3Gi memory is too less for 8 cpus. It needs 16Gi at least.
Hi, I have found the solution already to this, but wanted to document it here for anyone else that may be looking for it.
When attempting to run the 21.3.0 EE or XE image within kubernetes using the YAML files, or even within Helm, I was getting an error stating that the control01.ctl file could not be found in the /opt/oracle/oradata/ORCLCDB/ folder, with an SPFile error.
I even logged into the pod, and could verify that nothing existed in the SID folder.
here is the Log I got:
I stripped down to the following basic values of my kubernetes yaml file (deconstructed a helm chart to the basics to figure out what was wrong)
The solution will be in the first comment below