6 releases

0.3.4	Sep 8, 2024
0.3.3	Sep 8, 2024
0.3.0	Aug 26, 2024
0.2.1	Aug 26, 2024

#22 in Caching

697 downloads per month

MIT license

61KB
935 lines

turbo-repo-remote-cache-rs

Fast turbo remote cache server written in Rust.

if you are using turbo and you want to have a self-hosted remote cache server this is for you.

It is fast ⚡️.
It supports multiple object storage providers (S3, GCP, Azure, and file).

Installation

Using Cargo (Linux/macOS/Windows)

cargo install turbo-remote-cache-rs

Using Npm (Linux/macOS/Windows)

npm install -g turbo-remote-cache-rs

Download the latest binary from release page

Usage

Copy .env.example to same directory as the binary and rename it to .env.
Edit .env file to your needs (TURBO_TOKENS, BUCKET_NAME are required) please refer to .env.example for more information about required vars for each cloud storage provider.
Run the binary.

  turbo-repo-remote-cache-rs

Provide the url to turbo cli

turbo build --api="http://localhost:4000" --token=<token> --team=<team>

you can also set teamId and apiUrl .turbo/config.json in the root of your project.

Enjoy your self-hosted remote cache and faster builds 🚀.

Docker

Create a docker file.

FROM salamaashoush/turbo-remote-cache-rs
ENV PORT=4000
ENV FS_PATH=/tmp
ENV BUCKET_NAME=cache
ENV TURBO_TOKENS="token1,token2,token3"
EXPOSE 4000
CMD ["turbo-remote-cache-rs"]

Build the image.

docker build -t your-image .

Run the image.

docker run -p 4000:4000 -v ./cache:/tmp/cache your-image

Provide the URL to turbo cli

turbo build --api="http://localhost:4000" --token=<token> --team=<team>

you can also set teamId and apiUrl .turbo/config.json in the root of your project.

Kubernetes

See example in examples/k8s, Don't forget to change the spec and env vars for your needs before applying it (NOTE that it is just an example and it is not production ready).

Environment Variables

Required

Name	Description	Default
`TURBO_TOKENS`	Comma separated list of turbo tokens that are allowed to access the cache.	`""`
`BUCKET_NAME`	Name of the bucket to store the cache in.	`"cache"`
`STORAGE_PROVIDER`	Storage provider to use. `s3`, `azure`, `gcs`, `file` or `memory`	`"memory"`

File Storage Provider

Name	Description	Default
`FS_PATH`	Path to store the cache in.	os temp dir

S3 Storage Provider

Name	Description	Default
`AWS_ACCESS_KEY_ID`	AWS access key id.	`""`
`AWS_SECRET_ACCESS_KEY`	AWS secret access key.	`""`
`AWS_DEFAULT_REGION`	AWS region.	`""`
`AWS_ENDPOINT`	AWS endpoint.	`""`
`AWS_SESSION_TOKEN`	AWS session token	`""`
`AWS_CONTAINER_CREDENTIALS_RELATIVE_URI`	AWS bucket endpoint.	`""`
`AWS_ALLOW_HTTP`	set to “true” to permit HTTP connections without TLS.	`false`
`AWS_PROFILE`	set profile name, requires aws_profile feature enabled	`""`

Azure Storage Provider

Name	Description	Default
`AZURE_STORAGE_ACCOUNT_NAME`	storage account name.	`""`
`AZURE_STORAGE_ACCOUNT_KEY`	storage account master key	`""`
`AZURE_STORAGE_ACCESS_KEY`	alias for `AZURE_STORAGE_ACCOUNT_KEY`	`""`
`AZURE_STORAGE_CLIENT_ID`	client id for service principal authorization	`""`
`AZURE_STORAGE_CLIENT_SECRET`	client secret for service principal authorization	`""`
`AZURE_STORAGE_TENANT_ID`	tenant id used in oauth flows	`""`

GCS Storage Provider

Name	Description	Default
`GOOGLE_SERVICE_ACCOUNT`	location of service account file	`""`
`GOOGLE_SERVICE_ACCOUNT_PATH`	(alias) location of service account file	`""`
`SERVICE_ACCOUNT`	(alias) location of service account file	`""`
`GOOGLE_SERVICE_ACCOUNT_KEY`	JSON serialized service account key	`""`
`GOOGLE_BUCKET`	bucket name	`""`
`GOOGLE_BUCKET_NAME`	(alias) bucket name	`""`

Todo

Support turbo headers x-artifact-duration, x-artifact-tag, x-artifact-client-ci and x-artifact-client-interactive right now those are ignored and they don't affect the cache.
Add nx cloud support.
Add more advanced authentication support.
Maybe having a dashboard to manage teams and projects would be nice.

Dependencies

~17–28MB
~510K SLoC