FAQ

Frequently Asked Questions

Got a question?

We’re here to help.

Is there a difference between RepoFlow Cloud and RepoFlow Self-Hosted?

No, there is no difference. We deploy RepoFlow Self-Hosted on the cloud and provide it as a service. Cloud and self-hosted versions are always kept in sync, we believe in self-hosting.

I'm getting a 413 error. How do I resolve it?

By default, RepoFlow limits request sizes to help protect smaller environments from very large uploads.

You can adjust these limits as follows:

1. Server: Set the environment variables MAX_UPLOAD_FILE_SIZE_IN_BYTES and MAX_FILES_AMOUNT_IN_UPLOAD_REQUEST.
2. NGINX:
   • Helm Chart: Modify nginx.maxBodySize in values.yaml.
   • Docker Compose: Update all client_max_body_size values in your NGINX configuration.

RepoFlow feels slow. How can I improve performance?

1. Reduce network latency
   Ensure that your database, Hasura, and object storage are deployed as close as possible to the RepoFlow server.

2. Check compute resources and scale if needed
   Monitor CPU and memory usage for both the RepoFlow server and Hasura. If either service is under heavy load, scale it vertically or horizontally to improve performance.

3. Verify database connection limits
   Each Hasura instance can use up to 100 database connections. When scaling Hasura horizontally, make sure the database max_connections setting is increased accordingly to prevent connection exhaustion and performance degradation.

4. Test object storage performance
   RepoFlow includes a built-in object storage performance test. Enable it using the environment variables S3_TEST_PERFORMANCE_ON_START and TMP_FOLDER_TEST_PERFORMANCE_ON_START. More details can be found in the server environment variables.

I can't upload huge packages to RepoFlow

To allow huge package uploads, you need to configure a few things:

1. NGINX:
   • Helm Chart: Set nginx.maxBodySize in your values.yaml.
   • Docker Compose: Update the NGINX conf.d file and change client_max_body_size.
2. RepoFlow Server:
   • Set the environment variable MAX_UPLOAD_FILE_SIZE_IN_BYTES to a larger value.
   • Make sure the temporary folder has enough space to handle large uploads.
     For Docker Compose, this is under /tmp in the container.
     For Helm Chart, adjust the value under server.tempVolumeSize.

That’s it. If you still have issues uploading large packages, contact us for help.

RepoFlow takes too much RAM

RepoFlow RAM usage may increase when not configured properly. This often happens when using your own object storage and checksum or connection settings are not optimized. Make sure all storage-related environment variables are set correctly, especially the following defaults:

S3_CHECKSUM_ALGORITHM=NONE
S3_REQUEST_CHECKSUM_CALCULATION="WHEN_REQUIRED"
S3_RESPONSE_CHECKSUM_CALCULATION="WHEN_REQUIRED"

If you’re unsure, contact us for help, we can review your setup to ensure it’s optimal.

Why aren't my environment variables working correctly?

Ensure that all environment variable values are wrapped in double quotes (""), even if the value is a boolean or a number. This helps avoid parsing issues.

Is RepoFlow production ready even though it’s not 1.0.0?

Yes, RepoFlow is production ready and is being used every day in real environments.

How does RepoFlow licensing work for self-hosted?

RepoFlow self-hosted licenses are provided as a Docker image, which you’ll receive by email after purchase. Each license is valid for one year. Without a valid license, package uploads and downloads are disabled.

Does RepoFlow support air-gapped environments?

Yes, RepoFlow supports air-gapped setups. We provide dedicated Docker images for air-gapped deployments. Read more here.

I see 401 errors in the logs even though authentication is configured correctly. Is this a problem?

In many cases, this is completely normal behavior.

Some package managers, especially Docker and NuGet, intentionally make an initial request without credentials. If the server responds with 401 Unauthorized, the client then retries the same request with the appropriate authentication headers.

Because of this flow, seeing 401 responses in the logs does not necessarily indicate a misconfiguration or a failed authentication.

If you see repeated 401 responses without a successful follow-up request, or if package downloads fail, please contact us and we’ll help investigate your setup.

How can I find documentation for earlier versions of RepoFlow?

RepoFlow comes with documentation bundled as part of your instance. To find the most relevant docs for your version, just go to your RepoFlow installation and open the built-in documentation.

I need an API that's not in the documentation. What should I do?

No worries! We're constantly adding more APIs. Just contact us and let us know what you're trying to do. We'll help you out.

The CLI can't migrate a package type I need. What should I do?

Contact us! We're actively improving the CLI and can help you with your migration process directly.

Can’t find something in the docs?

If you can’t find what you’re looking for in the documentation, just contact us. We’re happy to help!