You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@yetus.apache.org by Nick Dimiduk <nd...@apache.org> on 2020/04/11 00:12:49 UTC

Help making sense of build output

Hi folks,

I think i'm looking at a Yetus bug, but I really don't know what to make of
it. The build in question is [0]. I'm looking at the "yetus general check,"
summary table in the Blue Ocean console output. All four of `compile`,
`cc`, `golang`, and `javac` checks are pointing at the same output file.
Are all the checks processed from the same command invocation? It seems to
me that somehow the association between the check and it's output has
gotten tangled here, resulting in the same output file being flagged for
all of them.

Thanks,
Nick

```
[2020-04-10T10:59:44.509Z] | compile |
https://builds.apache.org/job/HBase%20Nightly/job/master/1694/artifact/output-general/patch-compile-root.txt
 |
[2020-04-10T10:59:44.509Z] | cc |
https://builds.apache.org/job/HBase%20Nightly/job/master/1694/artifact/output-general/patch-compile-root.txt
 |
[2020-04-10T10:59:44.509Z] | golang |
https://builds.apache.org/job/HBase%20Nightly/job/master/1694/artifact/output-general/patch-compile-root.txt
 |
[2020-04-10T10:59:44.509Z] | javac |
https://builds.apache.org/job/HBase%20Nightly/job/master/1694/artifact/output-general/patch-compile-root.txt
 |
```

[0]:
https://builds.apache.org/blue/organizations/jenkins/HBase%20Nightly/detail/master/1694/pipeline/98/

Re: Help making sense of build output

Posted by Allen Wittenauer <aw...@effectivemachines.com.INVALID>.


> On Apr 10, 2020, at 5:12 PM, Nick Dimiduk <nd...@apache.org> wrote:
> I think i'm looking at a Yetus bug, but I really don't know what to make of
> it. The build in question is [0]. I'm looking at the "yetus general check,"
> summary table in the Blue Ocean console output. All four of `compile`,
> `cc`, `golang`, and `javac` checks are pointing at the same output file.
> Are all the checks processed from the same command invocation? It seems to
> me that somehow the association between the check and it's output has
> gotten tangled here, resulting in the same output file being flagged for
> all of them.


This all looks correct. 

compile phase is really "build tool was told to go compile something."  The expectation is that the build tool is going to trigger the necessary invocations of javac, cc, go, etc, and the messages from those compilers will be in the build tool's output.  In this particular case, the build tool--maven--failed so the language tests will all point to the compile phase's failure as to why they are also considered failed.  There's not much point in doing any deeper level of language analysis if equivalent of 'mvn compile' failed since it's going to be incomplete/wrong/etc.