After on-prem installation of 24.1, dxi-grafana pod in CrashLoopBackoff
Due to connectivity issues on the environment, the database scripts for the grafana pod failed,
As a result the data setup/migration of grafana tables did not complete successfully hence the grafana POD is in CrashLoopBackoff
Verification:
Open postgres pod terminal and run the below queries
psql grafana_db aopuser;
select count(*) from dashboard_version;
select count(*) from migration_log;
select * from migration_log where migration_id like 'save existing dashboard %';
IMPORTANT : the expected # of rows for dashoard_version is 239 and for migration_log is 305
1) Scale down to 0 grafana pods: dxi-grafana and dxi-grafana-reporting-engine
If openshift, go to the console > deployments
2) Delete data in dashboard_versions:
delete from dashboard_version;
3) Scale up to 1 grafana pods: dxi-grafana and dxi-grafana-reporting-engine
4) Wait for 2 minutes, check there is no exceptions in dx-grafana log
5) Open postgres pod terminal again and run the below queries for verification:
psql grafana_db aopuser;
select count(*) from dashboard_version;
select count(*) from migration_log;
6) Next step is to fix the registration of the tenant because of the grafana pod issue.
When trying to access dx-dashboard you will see "Server Error" "organization not found"
a) Access http(s)://dxi-dashboards.<your-domain>/login
b) Go to Server > Orgs
c) Click New Org, add the CohortID of your tenant
You can obtain this ID from the connector parameter page
Click "Create" button
7) Wait for a couple of minutes
8) Open DX OI in a new browser window, open DX Dashboards and it should open without any error.