fix: make netsim run from examples #2959
Conversation
|
Documentation for this PR has been generated and is available at: https://n0-computer.github.io/iroh/pr/2959/docs/iroh/ Last updated: 2024-11-25T14:05:10Z |
|
Arqu
changed the title
|
Waiting for CI and will merge. |
| let duration = start.elapsed(); | ||
| println!( | ||
| "Received {} B in {:.4}s with ttfb {}s in {} chunks", | ||
| HumanBytes(len as u64), |
There was a problem hiding this comment.
Did you mean to use HumanBytes here? the B in the format string suggests you just want an integer number of bytes that are easy to parse? The next line seems to do the human-friendly printing.
There was a problem hiding this comment.
Yes, mostly this is a precursor for moving the current netsim parsing to a different format. This just shows we have more data now.
There was a problem hiding this comment.
Did fix the extra B though.
| duration.as_secs_f64(), | ||
| HumanBytes((len as f64 / duration.as_secs_f64()) as u64) | ||
| ); | ||
|
|
There was a problem hiding this comment.
You now need to explicitly call endpoint.close().await to make a best effort to let the other side know you closed the connection. If you don't do this and the datagram carrying the close was lost the other side will trigger a timeout. Awaiting the endpoint.close makes sure to wait until the datagram carrying the close is acknowledged (or times out on the acknowledgement).
Currently you probably also want to put this endpoint.close().await wrapped in a application-level timeout like you do on the other side because our API WRT timeouts is a bit crap right now...
There was a problem hiding this comment.
Or should I move the endpoint.close also into the timeout?
There was a problem hiding this comment.
Something like this here:
| tokio::time::timeout(Duration::from_secs(3), async move { | |
| endpoint.close(0u8.into(), b"closing").await | |
| })??; |
dignifiedquire
merged commit f4adb94
into
refactor-drop-external-protocols
## Description Moves our netsim tests to run from an example and only test the lib part of the code, not the full CLI & blobs. Things left to do: - [x] rename the example from `new` to anything that makes more sense `netsim` runs it under `iroh-transfer` - [x] clean up the example code - [x] add the option to provide either a relay config or pass in at least the relay url as an argument - [x] continue the CI adjustment and move all invocations of the netsim runner to run from `iroh_v2` and `integration_v2` sims (we want a less abrupt netsim switchover) - [x] convert the remaining sims in `chuck/netsim/sims` to the new format - [ ] after some time flip back the CI invocations to be regular `iroh` and `integration` sims which includes doing the same on `netsim` and removing the old ones ## Breaking Changes <!-- Optional, if there are any breaking changes document them, including how to migrate older code. --> ## Notes & open questions <!-- Any notes, remarks or open questions you have to make about the PR. --> ## Change checklist - [ ] Self-review. - [ ] Documentation updates following the [style guide](https://rust-lang.github.io/rfcs/1574-more-api-documentation-conventions.html#appendix-a-full-conventions-text), if relevant. - [ ] Tests if relevant. - [ ] All breaking changes documented.
Description
Moves our netsim tests to run from an example and only test the lib part of the code, not the full CLI & blobs.
Things left to do:
newto anything that makes more sensenetsimruns it underiroh-transferiroh_v2andintegration_v2sims (we want a less abrupt netsim switchover)chuck/netsim/simsto the new formatirohandintegrationsims which includes doing the same onnetsimand removing the old onesBreaking Changes
Notes & open questions
Change checklist