You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Port/Duplicate async_hooks tests to use AsyncLocalStore #55712
Short after #55552 was merged I notices that this resulted in loosing a lot test for AsyncLocalStore.
Before AsyncLocalStore was based on async hooks. There are plenty of tests for async hooks which implict verified that AsyncLocalStore does what it is expected.
But since #55552AsyncLocalStore and async hooks are independent.
I don't think we have to dup them all, but there are some special cases at least in HTTP area (e.g. here) which might require followups.
changed the title [-]Port/Duplicte async_hooks tests to AsyncLocalStore[/-][+]Port/Duplicate async_hooks tests to use AsyncLocalStore[/+]on Nov 6, 2024
#54879 got merged? It looks like it was just closed to wait for v24.
In any case, I've added the drag-agenda label. Let's discuss this in more detail in the next diagnostics working group call. (Not the one today...probably a bit short notice for that.)
Anyhow, just noticed that I linked the wrong PR. It is about yours which was merged a while ago: #55552 That one changes internal how ALS works and relying on AsyncHooks tests is no longer applicable.
It swaps the default for the flag, but all tests with the test-async-local-storage- prefix are run both with and without he changes, so test coverage should remain the same.
Yes, they run with both. I'm talking more about the tests which don't have the test-async-local-storage- prefix.
e.g. test-crypto-randomBytes verifies init/before/after/destroy for async-hooks and as long as this test works an async hooks based ALS will likely work fine.
or test-async-exec-resource-http-32060 which has also no variant using test-async-local-storage- prefix therefore ALS coverage for that case is lost.
Ah, well if there's things which are testing ALS-related behaviour that aren't actually marked as such then there should probably be some tests specific to ALS for those scenarios.
This issue has been marked as stale due to 210 days of inactivity.
It will be automatically closed in 30 days if no further activity occurs. If this is still relevant, please leave a comment or update it to keep it open.
This issue has been automatically closed after 30 days of inactivity following its stale status (no activity for a total of 240 days).
If this is still relevant, feel free to reopen it or leave a comment with additional details so we can continue the discussion.
Activity
[-]Port/Duplicte async_hooks tests to AsyncLocalStore[/-][+]Port/Duplicate async_hooks tests to use AsyncLocalStore[/+]Qard commentedon Nov 21, 2024
#54879 got merged? It looks like it was just closed to wait for v24.
In any case, I've added the drag-agenda label. Let's discuss this in more detail in the next diagnostics working group call. (Not the one today...probably a bit short notice for that.)
Flarna commentedon Nov 21, 2024
No it wasn't. Will reopen for 24 at some time.
Anyhow, just noticed that I linked the wrong PR. It is about yours which was merged a while ago: #55552 That one changes internal how ALS works and relying on AsyncHooks tests is no longer applicable.
Qard commentedon Nov 25, 2024
It swaps the default for the flag, but all tests with the
test-async-local-storage-prefix are run both with and without he changes, so test coverage should remain the same.Flarna commentedon Nov 26, 2024
Yes, they run with both. I'm talking more about the tests which don't have the
test-async-local-storage-prefix.e.g.
test-crypto-randomBytesverifies init/before/after/destroy for async-hooks and as long as this test works an async hooks based ALS will likely work fine.or
test-async-exec-resource-http-32060which has also no variant usingtest-async-local-storage-prefix therefore ALS coverage for that case is lost.Qard commentedon Nov 29, 2024
Ah, well if there's things which are testing ALS-related behaviour that aren't actually marked as such then there should probably be some tests specific to ALS for those scenarios.
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
test: add ALS test using http agent keep alive
github-actions commentedon Apr 26, 2026
This issue has been marked as stale due to 210 days of inactivity.
It will be automatically closed in 30 days if no further activity occurs. If this is still relevant, please leave a comment or update it to keep it open.
github-actions commentedon May 26, 2026
This issue has been automatically closed after 30 days of inactivity following its stale status (no activity for a total of 240 days).
If this is still relevant, feel free to reopen it or leave a comment with additional details so we can continue the discussion.