alexcrichton assigned rvolosatovs to issue #11656.
alexcrichton opened issue #11656:
I'm about to disable this test as it's flaky on main, so this is a tracking issue for it. Example failures are here and here
alexcrichton added the wasm-proposal:component-model-async label to Issue #11656.
rvolosatovs commented on issue #11656:
So far I've been unable to reproduce this failure in almost 1.5h of running it continuously on a x86_64-linux and aarch64-darwin laptops
Looking at the CI failures, it appears that they happen on fairly beefy machines:
AMD EPYC 7763 64-Core ProcessorI'll try this on a GCP box as well I already have provisioned, but it looks like the only way for me to reproduce this is going to be using the CI box for it and iterate in #11658
@alexcrichton is there some BA-provisioned box I could perhaps use to replicate the CI environment? (I'm guessing no)
rvolosatovs edited a comment on issue #11656:
So far I've been unable to reproduce this failure in almost 1.5h of running it continuously on x86_64-linux and aarch64-darwin laptops
Looking at the CI failures, it appears that they happen on fairly beefy machines:
AMD EPYC 7763 64-Core ProcessorI'll try this on a GCP box as well I already have provisioned, but it looks like the only way for me to reproduce this is going to be using the CI box for it and iterate in #11658
@alexcrichton is there some BA-provisioned box I could perhaps use to replicate the CI environment? (I'm guessing no)
alexcrichton commented on issue #11656:
Unfortunately we don't have any such environment, no. I was able to reproduce locally with:
for i in `seq 1 100`; do taskset 0x55 cargo test -p wasmtime-wasi-http --test all --features p3 -- p3::|| break; done... eventually
it took quite awhile, but I didn't try to make it more reliable to reproduce
alexcrichton edited a comment on issue #11656:
Unfortunately we don't have any such environment, no. I was able to reproduce locally with:
for i in `seq 1 100`; do taskset 0x55 cargo test -p wasmtime-wasi-http --test all --features p3 -- p3:: || break; done... eventually
it took quite awhile, but I didn't try to make it more reliable to reproduce
rvolosatovs closed issue #11656:
I'm about to disable this test as it's flaky on main, so this is a tracking issue for it. Example failures are here and here
Last updated: Dec 06 2025 at 06:05 UTC