OpenShift: Recovering the V2018 deployment

Use the disaster recovery process to revert a failed upgrade to the IBM® API Connect V2018 deployment on OpenShift.

Before you begin

A recovery can only be performed if you completed the following tasks before beginning the upgrade:

You saved a backup copy of the apicup project that was created when you installed API Connect V2018.
You upgraded the subsystems in the correct sequence to ensure that dependencies were satisfied:
1. Management
2. Portal
3. Analytics
4. Gateway

About this task

To revert to V2018, remove artifacts from the failed upgrade, install V2018, and then restore the backups that you created for the V2018 deployment before you started the upgrade.

You can also use this process if you find a need to revert to V2018 after a successful upgrade to V10.

Procedure

Delete the upgrade tasks by running the following commands:
1. oc delete up2018g -n <APIC_namespace> --all
2. oc delete up2018a -n <APIC_namespace> --all
3. oc delete up2018p -n <APIC_namespace> --all
4. oc delete up2018m -n <APIC_namespace> --all
where APIC_namespace is the namespace where you installed API Connect.
Delete all of the V10 subsystems by running the following commands in the sequence shown:
1. Delete the Gateway subsystem: oc delete gw -n <APIC_namespace> --all
2. Delete the Analytics subsystem: oc delete a7s -n <APIC_namespace> --all
3. Delete the Portal subsystem: oc delete ptl -n <APIC_namespace> --all
4. Delete the Management subsystem: oc delete mgmt -n <APIC_namespace> --all
where APIC_namespace is the namespace where you installed API Connect.

Note: To revert to 2018 FP24, you remove all of the 10.0.1.12-eus subsystems.

Delete the Helm releases that were used in 2018 FP24.

List the Helm releases in the 2018 FP24 API Connect deployment by running the following command:

helm ls

The response looks like the following example, with the name of each Helm release displaying in the CHART column:

NAME            NAMESPACE       REVISION        UPDATED                                 STATUS          CHART                           APP VERSION
r19b9d4b2eb     e2e             1               2021-02-05 17:22:11.050619495 -0800 PST deployed        apiconnect-2.0.0
r5d63b7fc92     e2e             1               2021-02-05 17:22:24.115569226 -0800 PST deployed        apic-portal-2.0.0
r6c7b1188e0     e2e             1               2021-02-05 17:22:50.169539831 -0800 PST deployed        apic-analytics-2.0.0
r8dacbcffda     e2e             1               2021-02-05 17:22:36.248182901 -0800 PST deployed        dynamic-gateway-service-1.0.63  1.0
rff8ba771c8     e2e             1               2021-02-05 17:22:10.329984775 -0800 PST deployed        cassandra-operator-1.0.0        1.0.1

For each Helm release, run the following command to delete it:
```
helm uninstall <helm_release>
```
where <helm_release> is the name of the Helm release from the CHART column.
For example, in the sample response, there are five Helm releases, so you must run the command to delete each of those releases:
- apiconnect-2.0.0
- apic-portal-2.0.0
- apic-analytics-2.0.0
- dynamic-gateway-service-1.0.63
- cassandra-operator-1.0.0

Delete all PVCs related to the API Connect V2018 and V10 deployments.
- If there is no PVC in the namespace other than those created by API Connect, you can delete them all at once by running the following command:
```
oc delete pvc --all -n <APIC_namespace>
```
- If there are other PVCs in the same namespace, then delete the API Connect PVCs individually by running the following commands:
  1. Get the list of PVCs in the namespace:
```
oc get pvc
```
  2. Delete one PVC:
```
oc delete pvc <PVC_name> -n <APIC_namespace>
```
where:
- PVC_name is the name of a particular PVC that you want to delete
- APIC_namespace is the namespace where you installed. API Connect
If you do not intend to attempt the V10 upgrade again, use Operator Hub to uninstall the operators and catalog sources related to the upgrade.
Install API Connect V2018.
- Use the apicup project that you created in V2018 and backed up before starting the upgrade (this is necessary to ensure that settings such as certificates and encryption secrets are retained).
- Grant all of the security permissions required for the V2018 installation.
Configure backup and restore settings for each subsystem, and then restore the subsystems with the data you backed up previously in V2018, in the following sequence:
1. Management
2. Portal
3. Analytics
For instructions, see Backing up and restoring in the V2018 documentation.