[BACKPORT][v1.6.1][BUG] Failed to restore a backup to file by the scripts/restore-backup-to-file.sh with a CIFS backup target.

[github-actions]

backport #8126

[longhorn-io-github-bot]

Pre Ready-For-Testing Checklist

• Where is the reproduce steps/test steps documented?
  The reproduce steps/test steps are at:
  [BUG] Failed to restore a backup to file by the scripts/restore-backup-to-file.sh with a CIFS backup target. #8126 (comment)

• Is there a workaround for the issue? If so, where is it documented?
  The workaround is at:

• Does the PR include the explanation for the fix or the feature?

• Does the PR include deployment change (YAML/Chart)? If so, where are the PRs for both YAML file and Chart?
  The PR for the YAML change is at:
  The PR for the chart change is at:
  fix(scripts): Add CIFS credentials options to restore-backup-to-file (backport #8110) #8137

[roger-ryao]

Verified on v1.6.1-rc1 20240314

• longhorn v1.6.1-rc1 6e96e27

The test steps

#8126 (comment)

1. Deploy the CIFS service.

   kubectl create -f https://raw.githubusercontent.com/longhorn/longhorn/master/deploy/backupstores/cifs-backupstore.yaml`

2. Set Backup Target to cifs://longhorn-test-cifs-svc.default/backupstore
3. Set Backup Target Credential Secret to cifs-secret
4. Create a volume and backup
5. Check which node longhorn-test-cifs-xxx-xxxx pod was applied to and confirm its IP:

   > kubectl get pod -w -owide 
   NAME                                    READY   STATUS    RESTARTS   AGE     IP             NODE                         NOMINATED NODE       READINESS GATES
   longhorn-test-cifs-6968778858-bx7n5     1/1     Running   0          28m     10.42.3.58     ryao-161-w2-50978cc1-ltj2q   <none>               <none>
   ubuntu-mountvol0                        1/1     Running   0          57m     10.42.2.50     ryao-161-w3-8b7eea2f-4d776   <none>               <none>

6. get the backup file url

   > k -n longhorn-system get backup backup-bc3e9c2e3d3b49f4 -oyaml |grep url
   url: cifs://longhorn-test-cifs-svc.default/backupstore?backup=backup-bc3e9c2e3d3b49f4&volume=vol-0

7. Access the node to which longhorn-test-cifs-xxx-xxxx was applied.
8. Clone the longhorn repository git clone -b v1.6.1-rc1 https://github.com/longhorn/longhorn.git
9. Execute the script located at longhorn/scripts

   ./restore-backup-to-file.sh --cifs-username "longhorn-cifs-username"  \
             --cifs-username "longhorn-cifs-password" \
             --backup-url "cifs://10.42.3.58/backupstore?backup=backup-bc3e9c2e3d3b49f4&volume=vol-0" \
             --output-file "volume.raw" \
             --output-format raw \
             -v v1.6.1-rc1

10. mount volume.raw and confirm data consistency in the mount point.

    mount /tmp/restore/volume.raw /mnt
    

Result Passed

[roger-ryao]

To provide additional context
To enable access to our CIFS server via an AWS public IP, we can modify the cifs-backupstore.yaml file as follows:

diff --git a/deploy/backupstores/cifs-backupstore.yaml b/deploy/backupstores/cifs-backupstore.yaml
index 8ac586d..6739d4c 100644
--- a/deploy/backupstores/cifs-backupstore.yaml
+++ b/deploy/backupstores/cifs-backupstore.yaml
@@ -74,10 +74,13 @@ apiVersion: v1
 metadata:
   name: longhorn-test-cifs-svc
   namespace: default
+  annotations:
+    service.beta.kubernetes.io/aws-load-balancer-internal: "false" # For AWS ELB, set this to "false" for public access
 spec:
+  type: LoadBalancer  # Change this to LoadBalancer
   selector:
     app: longhorn-test-cifs
-  clusterIP: None
+#  clusterIP: None
   ports:
   - name: netbios-port
     port: 139

After applying this change, you can verify the service's external IP using the following command:

> kubectl get svc longhorn-test-cifs-svc -n default
NAME                     TYPE           CLUSTER-IP      EXTERNAL-IP                                PORT(S)                       AGE
longhorn-test-cifs-svc   LoadBalancer   10.43.111.236   34.223.148.37,52.122.15.136,54.170.247.242   139:31674/TCP,445:31087/TCP   30m

This should display a LoadBalancer service with an external IP assigned, allowing access to the CIFS server. You can then update the domain from cifs://longhorn-test-cifs-svc.default/backupstore to cifs://<external_ip>/backupstore for access using the node's public IP.