You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@bookkeeper.apache.org by Yong Zhang <zh...@gmail.com> on 2023/02/21 01:14:37 UTC

CI shows successfully but the tests actually has errors

Hi folks,

I found our CI may have some problems.

In the master code, the tests failed, but the CI was successful.
The following is the latest commit CI action:
https://github.com/apache/bookkeeper/actions/runs/4219246330/jobs/7324432945
Error:  Tests run: 633, Failures: 0, Errors: 29, Skipped: 0, Flakes: 1

You can see there have errors but the CI passed.

Also, the PR's check has the same issue.
https://github.com/apache/bookkeeper/actions/runs/4217320319/jobs/7321051845

We may need to fail the CI if there have errors.

Should we stop merging PRs until we fix the CI? But looks like most
failures are
about bookie shell commands. I am not sure whether we need to stop merging.

Best,
Yong

Re: CI shows successfully but the tests actually has errors

Posted by Yong Zhang <zh...@gmail.com>.

Here is the PR for fixing this
https://github.com/apache/bookkeeper/pull/3790

I am still trying to get the tests passed
https://github.com/zymap/bookkeeper/pull/20

Thanks,
Yong

On Tue, 21 Feb 2023 at 10:10, Yong Zhang <zh...@gmail.com> wrote:

> I took a look at the issue and found the root cause.
>
> I will push a fix later.
>
> Yong
>
> On Tue, 21 Feb 2023 at 10:00, Yan Zhao <ho...@apache.org> wrote:
>
>> >  Hi Yong,
>> >      Thanks for finding this critical issue. I think it's dangerous to
>> > merge new PRs before the CI is fixed, because we can't ensure whether
>> > the new PRs introduces bugs.
>>
>> +1
>>
>

Re: CI shows successfully but the tests actually has errors

Posted by Yong Zhang <zh...@gmail.com>.

I took a look at the issue and found the root cause.

I will push a fix later.

Yong

On Tue, 21 Feb 2023 at 10:00, Yan Zhao <ho...@apache.org> wrote:

> >  Hi Yong,
> >      Thanks for finding this critical issue. I think it's dangerous to
> > merge new PRs before the CI is fixed, because we can't ensure whether
> > the new PRs introduces bugs.
>
> +1
>

Re: CI shows successfully but the tests actually has errors

Posted by Yan Zhao <ho...@apache.org>.

>  Hi Yong,
>      Thanks for finding this critical issue. I think it's dangerous to
> merge new PRs before the CI is fixed, because we can't ensure whether
> the new PRs introduces bugs.

+1

Re: CI shows successfully but the tests actually has errors

Posted by Hang Chen <ch...@apache.org>.

 Hi Yong,
     Thanks for finding this critical issue. I think it's dangerous to
merge new PRs before the CI is fixed, because we can't ensure whether
the new PRs introduces bugs.

Thanks,
Hang

Yong Zhang <zh...@gmail.com> 于2023年2月21日周二 09:15写道：
>
> Hi folks,
>
> I found our CI may have some problems.
>
> In the master code, the tests failed, but the CI was successful.
> The following is the latest commit CI action:
> https://github.com/apache/bookkeeper/actions/runs/4219246330/jobs/7324432945
> Error:  Tests run: 633, Failures: 0, Errors: 29, Skipped: 0, Flakes: 1
>
> You can see there have errors but the CI passed.
>
> Also, the PR's check has the same issue.
> https://github.com/apache/bookkeeper/actions/runs/4217320319/jobs/7321051845
>
> We may need to fail the CI if there have errors.
>
> Should we stop merging PRs until we fix the CI? But looks like most
> failures are
> about bookie shell commands. I am not sure whether we need to stop merging.
>
> Best,
> Yong