I discovered a problem. If the maven command has a module filter (-pl), the
build hangs in the very beginning. This is a reproducible problem on a
private project (about 700 modules) and obviously related to switch from
3.9.9 to 3.9.10-SNAPSHOT.
The gist with stack traces:
https://gist.github.com/
Seems related to this:
https://issues.apache.org/jira/browse/MNG-8256
Interestingly, this is not noticeable on master, we may have missed
some backport maybe?
Thanks
T
On Mon, May 26, 2025 at 5:45 PM Sergey Chernov wrote:
>
> I discovered a problem. If the maven command has a module filter (-pl
Yup,
this went into master only:
https://issues.apache.org/jira/browse/MNG-8396
https://github.com/cstamas/maven/commit/62f85a4233366064750079d50fd86ef25a97c6f3
On Mon, May 26, 2025 at 6:57 PM Tamás Cservenák wrote:
>
> Seems related to this:
> https://issues.apache.org/jira/browse/MNG-8256
>
> I
ran the NetBeans maven tooling integration tests against the snapshot and
everything worked fine.
https://github.com/apache/netbeans/pull/8534
best regards,
michael
On 5/26/25 16:03, Slawomir Jaranowski wrote:
> Hi,
>
> For easy testing I have published latest version at:
>
> https://dist.apach
Created https://github.com/apache/maven/pull/2393
@Slawomir Jaranowski @Guillaume Nodet pls eyeball
On Mon, May 26, 2025 at 6:59 PM Tamás Cservenák wrote:
>
> Yup,
> this went into master only:
> https://issues.apache.org/jira/browse/MNG-8396
> https://github.com/cstamas/maven/commit/62f85a42333
I built a commit from your fork and tried the build with filtering. It's
still hanging (the problem is not addressed).
```
commit 5b37f4eacfa25e81545650f292c8ea0d00035512 (HEAD,
cstamas/maven-3.9.x-MNG-8396)
Author: Tamas Cservenak
Date: Mon May 26 19:07:59 2025 +0200
[MNG-8396] Backpor
Sergey, I added another commit, you missed.
T
On Mon, May 26, 2025 at 7:33 PM Sergey Chernov wrote:
>
> I built a commit from your fork and tried the build with filtering. It's
> still hanging (the problem is not addressed).
> ```
>
> commit 5b37f4eacfa25e81545650f292c8ea0d00035512 (HEAD,
> csta
Seems to be ok now 👍
On Mon, May 26, 2025 at 7:37 PM Tamás Cservenák wrote:
> Sergey, I added another commit, you missed.
>
> T
>
> On Mon, May 26, 2025 at 7:33 PM Sergey Chernov
> wrote:
> >
> > I built a commit from your fork and tried the build with filtering. It's
> > still hanging (the pr
dev distribution updated:
https://dist.apache.org/repos/dist/dev/maven/maven-3/3.9.10-SNAPSHOT/
svnid: 77141
Apache Maven 3.9.10-SNAPSHOT (1da6ce51230b0b3247c983ec25bdeae21d52395f)
On Mon, 26 May 2025 at 19:48, Sergey Chernov wrote:
> Seems to be ok now 👍
>
>
> On Mon, May 26, 2025 at 7:37 PM
Just tried the 4.0.0-rc-3 comparing with 3.9.9, it's 4x times slower (!) on
a project of ~900 modules (700 of them are jar).
Maven 3 builds it in 2m57s, while Maven 4 in 11m48s (reproducible, with all
m2 caches). No build cache solutions enabled to compare pure build time.
Project is nothing specia
We cannot do that. We'd have to cut a new release and vote on it.
If the goal is to evaluate the release, we need to start voting a
4.0.0, open the vote, if any issue is considered blocking, cancel the
vote, fix the issue and cut another 4.0.0 release. If the issue is not
blocking, we can do a 4.0.
Some plugins are known to cause important slow downs when executing in Maven 4.
This is the case for example of
https://github.com/tisonkun/os-detector which can be easily replaced
with https://github.com/maveniverse/nisse.
But I suppose there are other plugins that could cause the same behavior.
There is a problem using apache-maven-3.9.10-20250523.195106-41-bin.zip (
3.9.10-SNAPSHOT) version locally. When I run it with the existing mvnw wrapper
(which works quite well with apache-maven-3.9.9-bin.zip), there is a
failure mvnw.url: No such file or directory. The full log and the
configurati
I think we should not change the archive folder structure in a patch
release, and rollback to the previous name before GA.
On Mon, May 26, 2025 at 9:09 PM James Gao wrote:
> @Sergey , the top folder name inside the zip file is different:
> 3.9.9: folder name is "apache-maven-3.9.9/"
> 3.9.10: fo
This is "CI build", which does NOT contain the version.
Reason for it is this:
https://github.com/apache/maven/blob/maven-3.9.x/Jenkinsfile#L56
The `-PversionlessMavenDist`. By the way, master does not use this
anymore as it just creates a mixup like this one...
Thanks
T
On Mon, May 26, 2025 at 3
Totally agree! I'm not trying to advocate "no more RC's", I'm trying to
advocate "last RC should be as close as possible to the final product".
And given what has happened between the last RC and today, I don't think
we can cut 4.0.0 *today*. The diff is quite large IMO.
Thanks,
Maarten
On
Hi,
For easy testing I have published latest version at:
https://dist.apache.org/repos/dist/dev/maven/maven-3/3.9.10-SNAPSHOT/
The same distribution will be done during releases.
On Mon, 26 May 2025 at 15:45, Tamás Cservenák wrote:
> This is "CI build", which does NOT contain the version.
>
That definition makes a lot of sense to me. If we would adhere to it, we
should discipline ourselves after "a" next RC and *not* accept any code
changes *except* for fixing critical uses.
I think there have been too many changes since last RC to cut a 4.0.0 today.
Maarten
On 26/05/2025 14:14
On Mon, May 26, 2025 at 2:25 PM Maarten Mulders wrote:
>
> That definition makes a lot of sense to me. If we would adhere to it, we
> should discipline ourselves after "a" next RC and *not* accept any code
> changes *except* for fixing critical uses.
Be careful not to let the tail wag the dog. if
I believe having a "Release Candidate" is a good idea
As long as it is indeed a candidate: a version that is exactly the same as
a release, except for discovered critical issues and version numbers.
I wonder if promoting a candidate to a GA without any code changes requires
modification of existin
@Sergey , the top folder name inside the zip file is different:
3.9.9: folder name is "apache-maven-3.9.9/"
3.9.10: folder name is "apache-maven/"
so the script breaks since the folder pattern is changed.
On Mon, May 26, 2025 at 7:54 PM Sergey Chernov
wrote:
> There is a problem using apache-mav
21 matches
Mail list logo
