RATIS-2169. mvnw test: Corrupted STDOUT by directly writing to native stream in forked JVM 1 #1162
Conversation
… stream in forked JVM 1.
|
The failed test is a known problem but not related this; see RATIS-2151. |
There was a problem hiding this comment.
Thanks @szetszwo for the patch.
Did it happen only locally or in some PR? I don't see such issue in logs collected from runs on master.
pom.xml
Outdated
| @@ -176,7 +176,7 @@ | |||
| <maven-checkstyle-plugin.version>3.3.0</maven-checkstyle-plugin.version> | |||
| <maven-clover2-plugin.version>4.0.6</maven-clover2-plugin.version> | |||
| <maven-pdf-plugin.version>1.6.1</maven-pdf-plugin.version> | |||
| <maven-surefire-plugin.version>3.0.0-M4</maven-surefire-plugin.version> | |||
| <maven-surefire-plugin.version>3.5.1</maven-surefire-plugin.version> | |||
There was a problem hiding this comment.
Versions newer than 3.0.0-M4 are affected by https://issues.apache.org/jira/browse/SUREFIRE-1722
Maybe we have to decide which bug to live with ;)
There was a problem hiding this comment.
Good to know. Let's remain in 3.0.0-M4 then.
| <junit.jupiter.version>5.10.1</junit.jupiter.version> | ||
| <jacoco.version>0.8.11</jacoco.version> | ||
| <junit-bom.version>5.11.2</junit-bom.version> | ||
| <jacoco.version>0.8.12</jacoco.version> |
There was a problem hiding this comment.
According to the comments on stack overflow:
upgrading jacoco-plugin from 0.8.1 to 0.8.4 did the job
But we already have 0.8.11, which is newer than both, so I guess this won't fix the problem.
We can upgrade anyway, Ozone has been using 0.8.12 for 6 months now (HDDS-10660).
There was a problem hiding this comment.
I am also aware that. Just wonder they might have a better fix in a newer version.
|
@adoroszlai , thanks for reviewing this! Tested this change in #1156 . It was able to pass everything (with a few retries). Since this is just a build change, I would ignore the known test failures and merge this. |
See RATIS-2169