simplify integration tests#148
Conversation
e2a0e29 to
465ded9
Compare
465ded9 to
53c4a2e
Compare
faf1721 to
299ee90
Compare
|
@dicej there is an issue with the latest nightly. Is nightly still necessary for the runtime? In which case I will maybe either allow the lint or opt into the feature it is flagging |
|
Wait nevermind I just realized I didnt try the latest nightly locally. I'll do that later |
|
Ok it appears my wait on stderr doesn't work on windows... and this hangs forever... sorry... |
baef216 to
70f033b
Compare
70f033b to
f7dc207
Compare
|
@dicej This is ready (I think) |
dicej
left a comment
There was a problem hiding this comment.
Thanks for doing this!
LGTM overall; just wondering if we could drop the sleep in http_example and boost the number of retries, as well as switch to an exponential back-off (e.g. starting at 100ms and capped at 20 seconds or so).
| .spawn()?; | ||
|
|
||
| // Sleep a bit to give the server time to start | ||
| std::thread::sleep(Duration::from_secs(5)); |
There was a problem hiding this comment.
Is this necessary given the retry code below? I'd rather do some kind of retry with exponential back-off (e.g. no initial delay, then 100ms, then 200, then 400, etc.) to ensure that the tests run quickly on fast machines but also run reliably on slower machines.
There was a problem hiding this comment.
Good point. I got a little sidetracked just trying to get this to work 😅
I pushed up the change, let me know what you think
Finally getting around to replacing the need for all of these commands, and instead just uses rust crates/builtins.
I haven't replaced the wasmtime calls yet.