7.4 KiB
layout | title | parent |
---|---|---|
page | How to recover a destroyed deployment | How to ... |
How to recover a destroyed deployment
Overview
Let's consider a situation where we had an ArangoDeployment in Cluster mode (3 DbServers, 3 Coordinators, 3 Agents) with Local storage attached (only one K8s Node in the K8s cluster).
Due to some reason the ArangoDeployment was deleted (e.g. ETCD storage has been wiped out) and we want to recover it. Fortunately, we have a backup of the data on the disk.
To recover the ArangoDeployment we need to:
- Create PV and PVC for each member with persistent storage (agent, dbservers, single)
- Create a new ArangoDeployment with the same members IDs
Local storage data
We have a members (Agents & DbServers) data in the following directories:
> ls -1 /var/data/
f9rs2htwc9e0bzme
fepwdnnbf0keylgx
gqnkahucthoaityt
vka6ic19qcl1y3ec
rhlf8vixbsbewefo
rlzl467vfgsdpofu
To find out the name of the members to which data should be attached,
we need to check the UUID
file content in each directory:
> cat /var/data/*/UUID
AGNT-pntg5yc8
AGNT-kfyuj8ow
AGNT-bv5rofcz
PRMR-9xztmg4t
PRMR-l1pp19yl
PRMR-31akmzrp
Initial ArangoDeployment
Here is an example of the initial ArangoDeployment before deletion:
cat <<EOF | kubectl apply -f -
apiVersion: "database.arangodb.com/v1"
kind: "ArangoDeployment"
metadata:
name: "cluster"
spec:
externalAccess:
type: NodePort
mode: Cluster
agents:
volumeClaimTemplate:
spec:
storageClassName: my-local-ssd
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
volumeMode: Filesystem
dbservers:
volumeClaimTemplate:
spec:
storageClassName: my-local-ssd
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
volumeMode: Filesystem
EOF
Create PV and PVC
-
We need to create ArangoLocalStorage first:
cat <<EOF | kubectl apply -f - apiVersion: "storage.arangodb.com/v1alpha" kind: "ArangoLocalStorage" metadata: name: "local-storage" spec: storageClass: name: my-local-ssd isDefault: true localPath: - /mnt/data EOF
-
Now create PV and PVC for every directory listed above
-
Agents - here is an example for
AGNT-pntg5yc8
(f9rs2htwc9e0bzme
directory)-
PV
cat <<EOF | kubectl apply -f - apiVersion: "v1" kind: PersistentVolume metadata: labels: arango_deployment: cluster role: agent name: agent-pntg5yc8-f9rs2htwc9e0bzme spec: accessModes: - ReadWriteOnce capacity: storage: 1Gi local: path: /mnt/data/f9rs2htwc9e0bzme persistentVolumeReclaimPolicy: Retain storageClassName: my-local-ssd volumeMode: Filesystem nodeAffinity: required: nodeSelectorTerms: - matchExpressions: - key: kubernetes.io/hostname operator: In values: - minikube EOF
-
PVC
cat <<EOF | kubectl apply -f - apiVersion: v1 kind: PersistentVolumeClaim metadata: labels: app: arangodb arango_deployment: cluster role: agent name: agent-pntg5yc8 spec: accessModes: - ReadWriteOnce resources: requests: storage: 1Gi storageClassName: my-local-ssd volumeMode: Filesystem volumeName: agent-pntg5yc8-f9rs2htwc9e0bzme EOF
-
-
DbServers - here is an example for
PRMR-9xztmg4t
(vka6ic19qcl1y3ec
directory)-
PV
cat <<EOF | kubectl apply -f - apiVersion: "v1" kind: PersistentVolume metadata: labels: arango_deployment: cluster role: dbserver name: dbserver-9xztmg4t-vka6ic19qcl1y3ec spec: accessModes: - ReadWriteOnce capacity: storage: 1Gi local: path: /mnt/data/vka6ic19qcl1y3ec persistentVolumeReclaimPolicy: Retain storageClassName: my-local-ssd volumeMode: Filesystem nodeAffinity: required: nodeSelectorTerms: - matchExpressions: - key: kubernetes.io/hostname operator: In values: - minikube EOF
-
PVC
cat <<EOF | kubectl apply -f - apiVersion: v1 kind: PersistentVolumeClaim metadata: labels: app: arangodb arango_deployment: cluster role: dbserver name: dbserver-9xztmg4t spec: accessModes: - ReadWriteOnce resources: requests: storage: 1Gi storageClassName: my-local-ssd volumeMode: Filesystem volumeName: dbserver-9xztmg4t-vka6ic19qcl1y3ec EOF
-
-
Create ArangoDeployment with previously created PVC
Now we can create ArangoDeployment with previously created PVCs:
cat <<EOF | kubectl apply -f -
apiVersion: "database.arangodb.com/v1"
kind: "ArangoDeployment"
metadata:
name: "cluster"
spec:
externalAccess:
type: NodePort
mode: Cluster
agents:
volumeClaimTemplate:
spec:
storageClassName: my-local-ssd
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
volumeMode: Filesystem
dbservers:
volumeClaimTemplate:
spec:
storageClassName: my-local-ssd
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
volumeMode: Filesystem
status:
agency:
ids:
- AGNT-pntg5yc8
- AGNT-kfyuj8ow
- AGNT-bv5rofcz
size: 3
members:
agents:
- id: AGNT-pntg5yc8
persistentVolumeClaim:
name: agent-pntg5yc8
persistentVolumeClaimName: agent-pntg5yc8
- id: AGNT-kfyuj8ow
persistentVolumeClaim:
name: agent-kfyuj8ow
persistentVolumeClaimName: agent-kfyuj8ow
- id: AGNT-bv5rofcz
persistentVolumeClaim:
name: agent-bv5rofcz
persistentVolumeClaimName: agent-bv5rofcz
dbservers:
- id: PRMR-9xztmg4t
persistentVolumeClaim:
name: cluster-dbserver-9xztmg4t
persistentVolumeClaimName: cluster-dbserver-9xztmg4t
- id: PRMR-l1pp19yl
persistentVolumeClaim:
name: cluster-dbserver-l1pp19yl
persistentVolumeClaimName: cluster-dbserver-l1pp19yl
- id: PRMR-31akmzrp
persistentVolumeClaim:
name: cluster-dbserver-31akmzrp
persistentVolumeClaimName: cluster-dbserver-31akmzrp
EOF
That's it! Now you can use ArangoDB with restored data.