Skip to content
This repository has been archived by the owner on Jun 30, 2022. It is now read-only.

WebSocket provider timeout when running 0_test_all_combinations.js against Devnet #49

Closed
peterjgilbert opened this issue Mar 26, 2019 · 6 comments

Comments

@peterjgilbert
Copy link
Member

Example output:

    4) should have a starting count of 0 for a confidential Rust contract with [provider=ws]
    > No events were emitted
connection not open on send()

This is due to Cloudflare's 100 second inactivity timeout for WebSocket connections.

@peterjgilbert
Copy link
Member Author

Also happens when running the full set of tests against staging. This is presumably due to the 10 minute timeout set in nginx.

@ptrus
Copy link

ptrus commented Mar 28, 2019

Staging is also behind Cloudflare so the same 100 second WS inactivity timeout would apply (unless we're bypassing it and not using web3.oasiscloud-staging.net endpoint)

@peterjgilbert
Copy link
Member Author

We're running against staging in-cluster using the ws://\$(WEB3_GATEWAY_WEBSOCKET_SERVICE_HOST):\$(WEB3_GATEWAY_WEBSOCKET_SERVICE_PORT_WEB3_WEBSOCKET) endpoint, so we should be bypassing cloudflare.

@ptrus
Copy link

ptrus commented Mar 28, 2019

That is also bypassing the ingress (=nginx), so it's not the 10 minute nginx timeout in that case

@peterjgilbert
Copy link
Member Author

That is also bypassing the ingress (=nginx), so it's not the 10 minute nginx timeout in that case

Hmm, in that case the cause when running in-cluster must be different. Will file a separate issue.

@armaniferrante
Copy link
Contributor

This test suite has been removed with the new oasis-client. Closing.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants