New to KubeDB? Please start here.
Running MongoDB Branch Post-Actions
Post-actions prepare a branch before it becomes ready for use. For MongoDB, they can mask personal fields, remove sensitive collections, or create test fixtures. This guide masks five customer documents and removes a sensitive collection without changing the source.
Execution and readiness
Each postActions entry creates a Job in the target namespace. This example uses one entry and one Job. Courier waits for target database readiness, then runs the action; the Branch becomes Ready only after it succeeds. A configured password reset would run before the action.
Before You Begin
- Read the branching overview.
- Complete Courier setup with KubeDB and the Branch CRD. Use branching-capable Courier, provisioner, and database images.
- Ensure
MongoDBVersion7.0.21is available in your KubeDB catalog. - Use Ceph RBD storage supporting copy-on-write snapshots and clones. These three-member examples use compatible VolumeSnapshot and VolumeGroupSnapshot classes, with the corresponding CSI APIs, controllers, and driver support. See snapshot configuration.
- Allow capacity for three source and three target database pods, each requesting 1Gi of memory, plus operators and any action Jobs.
- Your account must be able to create databases and Branches, read auth Secrets, inspect snapshots and Jobs, and open pod shells.
The source uses namespace demo; the target uses dev. Replace ceph-rbd consistently if your Ceph StorageClass has another name. Reserve the names in this tutorial and create namespaces if they do not already exist.
kubectl create namespace demo
kubectl create namespace dev
Deploy the source database
Save the following as mongodb-pa-source.yaml. It creates a three-member replica set:
apiVersion: kubedb.com/v1
kind: MongoDB
metadata:
name: mongodb-pa-source
namespace: demo
spec:
version: 7.0.21
replicas: 3
replicaSet:
name: rs0
podTemplate:
spec:
containers:
- name: mongodb
resources:
requests:
cpu: 500m
memory: 1Gi
limits:
memory: 1Gi
storageType: Durable
storage:
storageClassName: ceph-rbd
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 2Gi
deletionPolicy: WipeOut
kubectl -n demo apply -f mongodb-pa-source.yaml
kubectl -n demo wait mongodb/mongodb-pa-source --for=jsonpath='{.status.phase}'=Ready --timeout=30m
Seed and inspect the source data
Discover the auth Secret referenced by mongodb-pa-source:
kubectl -n demo get mongodb mongodb-pa-source -o jsonpath='{.spec.authSecret.name}{"\n"}'
mongodb-pa-source-auth
Use the returned Secret name below if it differs. Decode its username and password:
kubectl -n demo get secret mongodb-pa-source-auth -o jsonpath='{.data.username}' | base64 -d
kubectl -n demo get secret mongodb-pa-source-auth -o jsonpath='{.data.password}' | base64 -d
The generated username in this example is root. Enter the decoded password when the client prompts.
From your terminal, open a pod shell:
kubectl -n demo exec -it mongodb-pa-source-0 -c mongodb -- bash
Inside the pod, start the database client through the primary Service. This also works when pod -0 is a secondary:
mongosh --host mongodb-pa-source.demo.svc --port 27017 --username root --authenticationDatabase admin --password
Enter the password decoded above. Run JavaScript at the mongosh prompt.
Result examples below are formatted for readability. Insert five customers. Majority acknowledgment and journaling are requested for the replica set writes; the final command flushes the primary’s pending writes.
use branchdb
db.customers.insertMany([
{"_id": 1, "full_name": "Customer 1", "email": "[email protected]", "phone": "+1-555-0001"},
{"_id": 2, "full_name": "Customer 2", "email": "[email protected]", "phone": "+1-555-0002"},
{"_id": 3, "full_name": "Customer 3", "email": "[email protected]", "phone": "+1-555-0003"},
{"_id": 4, "full_name": "Customer 4", "email": "[email protected]", "phone": "+1-555-0004"},
{"_id": 5, "full_name": "Customer 5", "email": "[email protected]", "phone": "+1-555-0005"}
], {writeConcern: {w: "majority", j: true}})
db.sensitive_tokens.insertMany([{_id: 1, token: "sample-token-1"}, {_id: 2, token: "sample-token-2"}], {writeConcern: {w: "majority", j: true}})
db.getSiblingDB("admin").runCommand({fsync: 1})
use branchdb
db.customers.find().sort({_id: 1}).toArray()
[
{"_id": 1, "full_name": "Customer 1", "email": "[email protected]", "phone": "+1-555-0001"},
{"_id": 2, "full_name": "Customer 2", "email": "[email protected]", "phone": "+1-555-0002"},
{"_id": 3, "full_name": "Customer 3", "email": "[email protected]", "phone": "+1-555-0003"},
{"_id": 4, "full_name": "Customer 4", "email": "[email protected]", "phone": "+1-555-0004"},
{"_id": 5, "full_name": "Customer 5", "email": "[email protected]", "phone": "+1-555-0005"}
]
The source also contains two sensitive tokens:
db.sensitive_tokens.find().sort({_id: 1}).toArray()
[{ _id: 1, token: "sample-token-1" }, { _id: 2, token: "sample-token-2" }]
Run exit in mongosh, then exit in the pod shell to return to your terminal.
Create a Branch with post-actions
Save the following as mongodb-pa-branch.yaml:
apiVersion: courier.kubedb.com/v1alpha1
kind: Branch
metadata:
name: mongodb-pa-branch
namespace: demo
spec:
source:
databaseRef:
apiGroup: kubedb.com
kind: MongoDB
name: mongodb-pa-source
namespace: demo
target:
namespace: dev
name: mongodb-pa-branch
storageClassName: ceph-rbd
resources:
requests:
cpu: 500m
memory: 1Gi
limits:
memory: 1Gi
postActions:
- image: ghcr.io/appscode-images/mongo:7.0.21
jobDefaults:
imagePullPolicy: IfNotPresent
backoffLimit: 3
activeDeadlineSeconds: 900
ttlSecondsAfterFinished: 3600
jobTemplate:
metadata:
labels:
branch-action: mongodb-pa-branch-prepare
spec:
# First operation: mask customer data.
# Second operation: drop the sensitive collection after masking succeeds.
args: [mongosh, --quiet, "mongodb://$(MONGO_HOST):$(MONGO_PORT)/branchdb",
--username, "$(MONGO_USERNAME)", --password, "$(MONGO_PASSWORD)",
--authenticationDatabase, admin, --eval,
"db.customers.updateMany({}, {$set: {full_name: \"Masked Customer\", email: \"[email protected]\", phone: \"+1-555-0000\"}}); db.sensitive_tokens.drop(); print(\"Prepared customers: \" + db.customers.countDocuments({}));"]
resources:
requests:
cpu: 100m
memory: 256Mi
deletionPolicy: Delete
kubectl -n demo apply -f mongodb-pa-branch.yaml
kubectl -n demo wait branch/mongodb-pa-branch --for=condition=Ready --timeout=60m
kubectl -n demo get branch mongodb-pa-branch
Courier creates mongodb/mongodb-pa-branch and its cloned PVCs in dev; KubeDB provisions the target database.
The image entrypoint executes mongosh. Kubernetes expands the $(MONGO_...) references from Courier’s injected connection environment. The JavaScript updates customers first, then drops the collection; an uncaught command error stops evaluation and exits unsuccessfully. Repeated updates produce the same masked values, and dropping an absent collection is harmless.
Inspect the action Job
The Branch passes through ActionsRunning; the phase can finish before you observe it. Find its Job by Courier’s labels:
kubectl -n dev get jobs -l courier.kubedb.com/branch=mongodb-pa-branch,courier.kubedb.com/branch-namespace=demo
kubectl -n dev logs job/mongodb-pa-branch-post-action -c actions
kubectl -n dev get pods -l branch-action=mongodb-pa-branch-prepare
Expect Job status Complete with 1/1 completions. The log includes Prepared customers: 5.
kubectl -n demo get branch mongodb-pa-branch -o jsonpath='{range .status.conditions[*]}{.type}{"\t"}{.status}{"\n"}{end}'
kubectl -n demo get branch mongodb-pa-branch -o jsonpath='{.status.resources}{"\n"}'
Expect PostActionsCompleted=True and Ready=True. The resource references identify the target auth Secret, cloned PVCs, and the action Job.
Verify that only the branch was changed
Inspect the branch
Discover the auth Secret referenced by mongodb-pa-branch:
kubectl -n dev get mongodb mongodb-pa-branch -o jsonpath='{.spec.authSecret.name}{"\n"}'
mongodb-pa-branch-auth
Use the returned Secret name below if it differs. Decode its username and password:
kubectl -n dev get secret mongodb-pa-branch-auth -o jsonpath='{.data.username}' | base64 -d
kubectl -n dev get secret mongodb-pa-branch-auth -o jsonpath='{.data.password}' | base64 -d
The generated username in this example is root. Enter the decoded password when the client prompts.
From your terminal, open a pod shell:
kubectl -n dev exec -it mongodb-pa-branch-0 -c mongodb -- bash
Inside the pod, start the database client through the primary Service. This also works when pod -0 is a secondary:
mongosh --host mongodb-pa-branch.dev.svc --port 27017 --username root --authenticationDatabase admin --password
Enter the password decoded above. Run JavaScript at the mongosh prompt.
use branchdb
db.customers.find().sort({_id: 1}).toArray()
[
{"_id": 1, "full_name": "Masked Customer", "email": "[email protected]", "phone": "+1-555-0000"},
{"_id": 2, "full_name": "Masked Customer", "email": "[email protected]", "phone": "+1-555-0000"},
{"_id": 3, "full_name": "Masked Customer", "email": "[email protected]", "phone": "+1-555-0000"},
{"_id": 4, "full_name": "Masked Customer", "email": "[email protected]", "phone": "+1-555-0000"},
{"_id": 5, "full_name": "Masked Customer", "email": "[email protected]", "phone": "+1-555-0000"}
]
db.getCollectionNames().includes("sensitive_tokens")
false
Run exit in mongosh, then exit in the pod shell to return to your terminal.
Inspect the source
Discover the auth Secret referenced by mongodb-pa-source:
kubectl -n demo get mongodb mongodb-pa-source -o jsonpath='{.spec.authSecret.name}{"\n"}'
mongodb-pa-source-auth
Use the returned Secret name below if it differs. Decode its username and password:
kubectl -n demo get secret mongodb-pa-source-auth -o jsonpath='{.data.username}' | base64 -d
kubectl -n demo get secret mongodb-pa-source-auth -o jsonpath='{.data.password}' | base64 -d
The generated username in this example is root. Enter the decoded password when the client prompts.
From your terminal, open a pod shell:
kubectl -n demo exec -it mongodb-pa-source-0 -c mongodb -- bash
Inside the pod, start the database client through the primary Service. This also works when pod -0 is a secondary:
mongosh --host mongodb-pa-source.demo.svc --port 27017 --username root --authenticationDatabase admin --password
Enter the password decoded above. Run JavaScript at the mongosh prompt.
use branchdb
db.customers.find().sort({_id: 1}).toArray()
[
{"_id": 1, "full_name": "Customer 1", "email": "[email protected]", "phone": "+1-555-0001"},
{"_id": 2, "full_name": "Customer 2", "email": "[email protected]", "phone": "+1-555-0002"},
{"_id": 3, "full_name": "Customer 3", "email": "[email protected]", "phone": "+1-555-0003"},
{"_id": 4, "full_name": "Customer 4", "email": "[email protected]", "phone": "+1-555-0004"},
{"_id": 5, "full_name": "Customer 5", "email": "[email protected]", "phone": "+1-555-0005"}
]
db.sensitive_tokens.find().sort({_id: 1}).toArray()
[{ _id: 1, token: "sample-token-1" }, { _id: 2, token: "sample-token-2" }]
Run exit in mongosh, then exit in the pod shell to return to your terminal.
Connection environment
| Variable | Value used by the action |
|---|---|
MONGO_HOST | Target database Service |
MONGO_PORT | 27017 |
MONGO_USERNAME | Username from the target auth Secret |
MONGO_PASSWORD | Password from the target auth Secret |
See the common action configuration and Job defaults for the feature fields. The image must execute its supplied arguments; the Branch action template does not expose a container command override.
Use post-actions with refresh or another cluster
With refresh, actions run again against each new copy before readiness. Generation 1 uses a Job named mongodb-pa-branch-post-action-1; older generation Jobs can be removed by Courier. Design actions for repetition and use logs before cleanup. For cross-cluster branching, the Job runs on the target cluster; inspect it there and use the target namespace in any explicit connection address.
Cleanup
Delete the Branch on the source cluster first and let Courier finish removing its target and snapshots. Then delete the source database:
kubectl -n demo delete branch mongodb-pa-branch --wait=true --timeout=30m
kubectl -n demo delete mongodb mongodb-pa-source --wait=true --timeout=30m
Keep Courier, the storage backend, and the OCM control plane when used available until Branch deletion completes. The commands leave the namespaces in place.

































