Management database changes from V10.0.7

Management database operator changes from V10.0.7 that impact upgrade from earlier releases.

Note: If you are upgrading to V10.0.8, then SFTP and local backups are supported, and your V10.0.5 management database backup configuration is automatically migrated. You do not need to take any other actions. The rest of this topic applies to upgrades to V10.0.7 only.

In V10.0.7.0, the Crunchy Postgres operator is replaced with a different operator called EDB. The Postgres database itself remains the same, but the Kubernetes pods and jobs that manage the Postgres database are new.

The change in Postgres operator results in new CR status outputs, and new backup requirements and settings, which are explained in this topic.

Management upgrade warning

When you upgrade the API Connect operator to V10.0.7.0, the management CR reports a Warning state until the management subsystem is also upgraded to V10.0.7.0. The reason for the warning is that due to the database operator change in V10.0.7.0 the management subsystem must not be left on V10.0.6.0 while the operator is on V10.0.7.0.

S3 backup settings automatically updated

If your management subsystem is already configured for S3 backups, then no extra pre-upgrade steps are required. The upgrade process converts the existing S3 backup settings for the new management database in V10.0.7.0.

When you upgrade to V10.0.7.0 with S3 backups configured, an extra backup path folder /edb is appended to your backup path. The /edb path is to distinguish new V10.0.7.0 backups, from your previous V10.0.6.0 backups.

Update backup settings before upgrade

V10.0.7.0 supports only backups to an S3 object-store for the management database. If your V10.0.6.0 deployment uses SFTP or local backups, then you must reconfigure your management backup settings to use an S3 object-store.

Upgrade to V10.0.7.0 is blocked if management database backups are not configured to use an S3 object-store. To update your backup settings, see the topic for your platform:

If you do not require management database backups, you can enable an opt-out parameter to override the upgrade block:

Kubernetes and OpenShift individual subsystem installations: Add the following annotation to your management CR:
```
metadata:
  annotations: 
    apiconnect-operator/backups-not-configured: "true"
```
Cloud Pak for Integration and OpenShift top-level CR installations: Add the following annotation to your APIConnectCluster CR:
```
metadata:
  annotations: 
    apiconnect-operator/backups-not-configured: "true"
```