You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@shardingsphere.apache.org by Robert Scholte <rf...@apache.org> on 2020/03/28 08:58:42 UTC

shardingsphere-ci at builds.a.o

Hi, yesterday I was waiting for our Apache Maven jobs to finish. It took a long time before the ubuntu runs were picked up. I noticed quite a lot of shardingsphere-ci hanging (still at this moment...). I've asked one of my teammembers to have a look at your Jenkinsfile. He came with a couple of advices: - https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L59-L71 using script where you don't need to. There a per-stage post support. The try{...}catch is likely why the build is hanging. - https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L33-L46 is unnecessary with new syntax - https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L63 should be wrapped in a withMaven - https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L79-L81 is not required as you wipe anyway in their checkout (which is also unnecessary as you could configure the job to have the SCM plugin do the wipe for you and the submodule update) The Apache Maven project has a lot of subproject, which are all tested with a matrix of OS x JDK ( x Maven version in case of plugins). If nodes are not released, these jobs pile up and it looks like our project is doing a bad thing. I hope you can improve the pipeline script. best, Robert

ps. could you kill the current hanging jobs?

Re: shardingsphere-ci at builds.a.o

Posted by "zhangliang@apache.org" <zh...@apache.org>.
Hi Robert,

Sorry to hear that. I will stop the job ASAP, and then check it carefully.
We will restart the job after find the reason.

------------------

Liang Zhang (John)
Apache ShardingSphere & Dubbo


Robert Scholte <rf...@apache.org> 于2020年3月28日周六 下午4:58写道:

> Hi, yesterday I was waiting for our Apache Maven jobs to finish. It took a
> long time before the ubuntu runs were picked up. I noticed quite a lot of
> shardingsphere-ci hanging (still at this moment...). I've asked one of my
> teammembers to have a look at your Jenkinsfile. He came with a couple of
> advices: -
> https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L59-L71
> using script where you don't need to. There a per-stage post support. The
> try{...}catch is likely why the build is hanging. -
> https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L33-L46
> is unnecessary with new syntax -
> https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L63
> should be wrapped in a withMaven -
> https://github.com/apache/incubator-shardingsphere/blob/master/jenkinsfile#L79-L81
> is not required as you wipe anyway in their checkout (which is also
> unnecessary as you could configure the job to have the SCM plugin do the
> wipe for you and the submodule update) The Apache Maven project has a lot
> of subproject, which are all tested with a matrix of OS x JDK ( x Maven
> version in case of plugins). If nodes are not released, these jobs pile up
> and it looks like our project is doing a bad thing. I hope you can improve
> the pipeline script. best, Robert
>
> ps. could you kill the current hanging jobs?