On 9/26/23 8:51 PM, Khem Raj wrote:
On Tue, Sep 26, 2023 at 7:55 PM Robert Joslyn
<[email protected]> wrote:
On 9/26/23 7:55 AM, Khem Raj wrote:
I am seeing a ptest failure on qemx86-64/glibc
Failed ptests:
{'curl': ['test_1474', 'curl']}
In looking at the test, it is marked as flaky with the comment:
# Because of the timing sensitivity (scheduling delays of 500 msec can cause
# the test to fail), this test is marked flaky to avoid it being run in
the CI
# builds which are often run on overloaded servers.
https://github.com/curl/curl/blob/curl-8_3_0/tests/data/test1474#L21
My server is pretty lightly loaded while running the tests, so that
could explain why I haven't seen it fail normally. I was able to induce
a failure by running a very heavy load while running that test on my
desktop (which is older and slower than my build server). If you'd like,
I can send a follow up patch to skip tests marked as flaky.
another run resulted in yet another failure
test 0587...
FAIL: 587: protoc!
There was no content at all in the file log/2/server.input.
Server glitch? Total curl failure? Returned: 42
prior to this upgrade my test setup never saw curl failures and it has
same amount of load etc.
Test 587 is also marked flaky by upstream. There are 12 tests marked as
flaky total, and since you've just hit two of them, I'm much more
inclined to just skip them. Right before this version bump there was
also an update to run the curl ptests in parallel, which could be
contributing to the change in test behavior.
Thanks,
Robert
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#188321):
https://lists.openembedded.org/g/openembedded-core/message/188321
Mute This Topic: https://lists.openembedded.org/mt/101543494/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
