Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/mschuchard/vault-raft-backup
A lean tool for creating Vault Raft snapshots and transferring them to AWS S3
https://github.com/mschuchard/vault-raft-backup
aws-s3 backup raft s3 vault vault-plugin vault-plugins
Last synced: 1 day ago
JSON representation
A lean tool for creating Vault Raft snapshots and transferring them to AWS S3
- Host: GitHub
- URL: https://github.com/mschuchard/vault-raft-backup
- Owner: mschuchard
- License: bsd-3-clause
- Created: 2023-12-26T21:00:34.000Z (9 months ago)
- Default Branch: master
- Last Pushed: 2024-09-08T11:00:37.000Z (25 days ago)
- Last Synced: 2024-09-08T12:47:41.706Z (25 days ago)
- Topics: aws-s3, backup, raft, s3, vault, vault-plugin, vault-plugins
- Language: Go
- Homepage:
- Size: 136 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
## vault-raft-backup
Vault Raft Backup is a lean tool for creating snapshots of the Raft integrated storage in [Hashicorp Vault](https://www.vaultproject.io), and transferring those backups to AWS S3. This plugin's code at `HEAD` is currently tested against Vault version 1.15.6. The most recent release was tested against 1.15.6.
This repository and project is based on the work performed for [MITODL](https://github.com/mitodl/vault-raft-backup), and now serves as an upstream for the project hosted within that organization. Although the original work is unlicensed, this repository maintains the BSD-3 license with copyright notice on good faith.
Statically linked binaries for various operating systems and processor architectures are available at the Github releases page (see link in the right column).
### Prerequsities
If executing as an ad-hoc compile and run (i.e. `go run`), then the dependencies and requirements can be viewed in the [go.mod](go.mod) file. Additional setup requirements are as follows:
- Connectivity to a functioning Vault server cluster with Raft integrated storage.
- Authentication and authorization against the Vault server cluster for executing Raft snapshots.
- Authentication can be input in general as a token.
- Authentication can also specified as AWS IAM. In this situation, the Vault server cluster must have a role configured and mapped to an AWS IAM role. This AWS IAM role authorization must also be accessible by the Vault Raft Backup tool somehow (e.g. tool executed on EC2 instance with appropriate IAM Instance profile corresponding to AWS IAM role).
- A local filesystem with permissions and storage capable of staging the snapshot.
- Authentication and authorization against an AWS account for listing, reading, and writing objects to a S3 bucket.
- A S3 bucket capable of storing the snapshot.
The Vault policy for authorizing the execution of Raft snapshots appears like:
```hcl
path "sys/storage/raft/snapshot" {
capabilities = ["read"]
}
```
### Usage
Vault Raft Backup can be configured with either environment variables or a HCL2 config file.
Additionally, AWS authentication and configuration must be provided with standard methods that do not require manual inputs. The AWS Golang SDK will automatically read authentication information as per normal (i.e. IAM instance profile, `AWS_SHARED_CREDENTIALS_FILE` credentials file, `AWS_PROFILE` config file, environment variables e.g. `AWS_SESSION_TOKEN` and `AWS_REGION`, etc.).
#### Environment Variables
The following environment variables are read for the configuration of the backup tool.
```
# equivalent to VAULT_ADDR with vault cli executable
# default: http://127.0.0.1:8200
export VAULT_ADDR=
# equivalent to VAULT_SKIP_VERIFY with vault cli executable
# default: false
export VAULT_SKIP_VERIFY=
# default: determined based on other inputs
export VAULT_AUTH_ENGINE=
# equivalent to VAULT_TOKEN with vault cli executable
# default: empty
export VAULT_TOKEN=
# default: aws
export VAULT_AWS_MOUNT=
# default: empty
export VAULT_AWS_ROLE=
# default: /vault<\d+>.bak
# NOTE: if this file does not exist it will be created with 0600; if it does exist it will be completely overwritten
export VAULT_SNAPSHOT_PATH=
# required
export S3_BUCKET=
# this is prepended to the base filename in VAULT_SNAPSHOT_PATH
# default: empty
export S3_PREFIX=
# determines whether or not the local snapshot file is removed after a successful transfer to the final storage location
# default: true
export SNAPSHOT_CLEANUP=
```
#### HCL2 Config File
The HCL2 config file path is passed to the `vault-raft-backup` executable via the `-c` command line argument (e.g. `vault-raft-backup -c config.hcl`). The schema can be viewed below.
```hcl2
vault_config {
# equivalent to VAULT_ADDR with vault cli executable
# default: http://127.0.0.1:8200
address =
# equivalent to VAULT_SKIP_VERIFY with vault cli executable
# default: false
insecure =
# default: determined based on other inputs
auth_engine =
# equivalent to VAULT_TOKEN with vault cli executable
# default: empty
token =
# default: aws
aws_mount_path =
# default: empty
aws_role =
# default: /vault<\d+>.bak
# NOTE: if this file does not exist it will be created with 0600; if it does exist it will be completely overwritten
snapshot_path =
}
aws_config {
# required
s3_bucket =
# this is prepended to the base filename in VAULT_SNAPSHOT_PATH
# default: empty
s3_prefix =
}
# determines whether or not the local snapshot file is removed after a successful transfer to the final storage location
# default: true
snapshot_cleanup =
```
## Contributing
Code should pass all unit and acceptance tests. New features should involve new unit tests.
Please consult the GitHub Project for the current development roadmap.