doc/operator: update formatting for backup section, add FAQ section (#3416)

* doc/operator: update formatting for backup section, add FAQ section

* doc/operator: address review feedback

* doc/operator: add note about difference between `VMRestore` and `VMBackupmanager` as init containers

* Update docs/operator/backups.MD

Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>

Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>
This commit is contained in:
Zakhar Bessarab 2022-11-29 22:19:46 +04:00 committed by GitHub
parent 0002de937b
commit 2f837c1b23
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 101 additions and 68 deletions

22
docs/operator/FAQ.md Normal file
View File

@ -0,0 +1,22 @@
---
sort: 15
---
# FAQ
## How to change VMStorage PVC storage class
With Helm chart deployment:
1. Update the PVCs manually
2. Run `kubectl delete statefulset --cascade=orphan {vmstorage-sts}` which will delete the sts but keep the pods
3. Update helm chart with the new storage class in the volumeClaimTemplate
4. Run the helm chart to recreate the sts with the updated value
With Operator deployment:
1. Update the PVCs manually
2. Run `kubectl delete vmcluster --cascade=orphan {cluster-name}`
3. Run `kubectl delete statefulset --cascade=orphan {vmstorage-sts}`
4. Update VMCluster spec to use new storage class
5. Apply cluster configuration

View File

@ -6,18 +6,21 @@ sort: 5
## vmbackupmanager
## vmbackupmanager is proprietary software.
You can check vmbackupmanager [documentation](https://docs.victoriametrics.com/vmbackupmanager.html). It contains a description of the service and its features. This documentation covers vmbackumanager integration in vmoperator
Before using it, you must have signed contract and accept EULA https://victoriametrics.com/assets/VM_EULA.pdf
### vmbackupmanager is a part of VictoriaMetrics Enterprise offer
*Before using it, you must have signed contract and accept EULA https://victoriametrics.com/assets/VM_EULA.pdf*
## Usage examples
`VMSingle` and `VMCluster` has built-in backup configuration, it uses `vmbackupmanager` - proprietary tool for backups.
It supports incremental backups (hours, daily, etc) with popular object storages (aws s3, google cloud storage).
`VMSingle` and `VMCluster` has built-in backup configuration, it uses `vmbackupmanager` - proprietary tool for backups.
It supports incremental backups (hourly, daily, weekly, monthly) with popular object storages (aws s3, google cloud storage).
You can enable it with the simple configuration, define secret
The configuration example is the following:
```yaml
```yaml
---
apiVersion: v1
kind: Secret
@ -63,46 +66,45 @@ spec:
credentialsSecret:
name: remote-storage-keys
key: credentials
```
NOTE: for cluster version operator adds suffix for `destination: "s3://your_bucket/folder"`, it becomes `"s3://your_bucket/folder/$(POD_NAME)"`.
NOTE: for cluster version operator adds suffix for destination: `"s3://your_bucket/folder"`, it becomes `"s3://your_bucket/folder/$(POD_NAME)"`.
It's needed to make consistent backups for each storage node.
You can read more about backup configuration options and mechanics [here](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/master/app/vmbackup)
You can read more about backup configuration options and mechanics [here](https://docs.victoriametrics.com/vmbackup.html)
Possible configuration options for backup crd can be found at [link](https://docs.victoriametrics.com/operator/api.html#vmbackup)
Possible configuration options for backup crd can be found at [link](https://docs.victoriametrics.com/operator/api.html#vmbackup)
## Restoring backups
There are several ways to restore with [vmrestore](https://docs.victoriametrics.com/vmrestore.html) or [vmbackupmanager](https://docs.victoriametrics.com/vmbackupmanager.html).
It can be done with [vmrestore](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/master/app/vmrestore)
There two ways:
### Manually mounting disk
You have to stop `VMSingle` by scaling it replicas to zero and manually restore data to the database directory.
First:
You have to stop `VMSingle` by scaling it replicas to zero and manually restore data to the database directory.
Steps:
1. edit `VMSingle` CRD, set replicaCount: 0
2. wait until database stops
3. ssh to some server, where you can mount `VMSingle` disk and mount it manually
4. restore files with `vmrestore`
5. umount disk
6. edit `VMSingle` CRD, set replicaCount: 1
7. wait database start
Steps:
1) edit `VMSingle` CRD, set replicaCount: 0
2) wait until database stops
3) ssh to some server, where you can mount `VMSingle` disk and mount it manually
4) restore files with `vmrestore`
5) umount disk
6) edit `VMSingle` CRD, set replicaCount: 1
7) wait database start
### Using VMRestore init container
Second:
1. add init container with vmrestore command to `VMSingle` CRD, example:
1) add init container with vmrestore command to `VMSingle` CRD, example:
```yaml
apiVersion: operator.victoriametrics.com/v1beta1
kind: VMSingle
metadata:
```yaml
apiVersion: operator.victoriametrics.com/v1beta1
kind: VMSingle
metadata:
name: vmsingle-restored
namespace: monitoring-system
spec:
spec:
initContainers:
- name: vmrestore
image: victoriametrics/vmrestore:latest
@ -125,12 +127,21 @@ spec:
runOnStart: "true"
image:
repository: victoriametrics/vmbackupmanager
tag: v1.67.0-enterprise
tag: v1.83.0-enterprise
credentialsSecret:
name: remote-storage-keys
key: credentials
```
2. apply it, and db will be restored from s3
```
2) apply it, and db will be restored from s3
3. remove initContainers and apply crd.
3) remove initContainers and apply crd.
Note that using `VMRestore` will require adjusting `src` for each pod because restore will be handled per-pod.
### Using VMBackupmanager init container
Using VMBackupmanager restore in Kubernetes environment is described [here](https://docs.victoriametrics.com/vmbackupmanager.html#how-to-restore-in-kubernetes).
Advantages of using `VMBackupmanager` include:
- automatic adjustment of `src` for each pod when backup is requested
- graceful handling of case when no restore is required - `VMBackupmanager` will exit with successful status code and won't prevent pod from starting