You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@logging.apache.org by Ralph Goers <ra...@dslextreme.com> on 2018/02/18 07:44:48 UTC
Log4j Jenkins builds
While splitting out components into modules should have had the desired effect of reducing the dependencies in log4j-core, it didn’t help making the builds take less time. The Jenkins build for 3.x took about 43 minutes while the 2.x build took 38 minutes.
Ralph
Re: Log4j Jenkins builds
Posted by Matt Sicker <bo...@gmail.com>.
If Jenkins is configured to do "mvn clean" first, then the only time it
could save is in downloading dependencies. I usually set up my own Jenkins
jobs using "mvn verify" or "./gradlew build" or equivalent, though it tends
to work better with Gradle in the past (incremental compilation with Maven
might be better by now I hope).
On 18 February 2018 at 13:33, Ralph Goers <ra...@dslextreme.com>
wrote:
> It just occurred to me that maybe the build took longer than normal
> because this was the first time a standalone build for 3.x was run. I’ll
> review it again after it builds a few more times.
>
> Ralph
>
> > On Feb 18, 2018, at 12:20 PM, Gary Gregory <ga...@gmail.com>
> wrote:
> >
> > On Sun, Feb 18, 2018 at 12:44 AM, Ralph Goers <
> ralph.goers@dslextreme.com>
> > wrote:
> >
> >> While splitting out components into modules should have had the desired
> >> effect of reducing the dependencies in log4j-core, it didn’t help making
> >> the builds take less time. The Jenkins build for 3.x took about 43
> minutes
> >> while the 2.x build took 38 minutes.
> >>
> >
> > I am continuing on this path by implementing
> > https://issues.apache.org/jira/browse/LOG4J2-2237 "Move Jackson-based
> > layouts to their own modules."
> >
> > There is obviously some extra work the build needs to do for each
> > additional module, which explains the longer build.
> >
> > Gary
> >
> >
> >>
> >> Ralph
> >>
>
>
>
--
Matt Sicker <bo...@gmail.com>
Re: Log4j Jenkins builds
Posted by Ralph Goers <ra...@dslextreme.com>.
It just occurred to me that maybe the build took longer than normal because this was the first time a standalone build for 3.x was run. I’ll review it again after it builds a few more times.
Ralph
> On Feb 18, 2018, at 12:20 PM, Gary Gregory <ga...@gmail.com> wrote:
>
> On Sun, Feb 18, 2018 at 12:44 AM, Ralph Goers <ra...@dslextreme.com>
> wrote:
>
>> While splitting out components into modules should have had the desired
>> effect of reducing the dependencies in log4j-core, it didn’t help making
>> the builds take less time. The Jenkins build for 3.x took about 43 minutes
>> while the 2.x build took 38 minutes.
>>
>
> I am continuing on this path by implementing
> https://issues.apache.org/jira/browse/LOG4J2-2237 "Move Jackson-based
> layouts to their own modules."
>
> There is obviously some extra work the build needs to do for each
> additional module, which explains the longer build.
>
> Gary
>
>
>>
>> Ralph
>>
Re: Log4j Jenkins builds
Posted by Gary Gregory <ga...@gmail.com>.
On Sun, Feb 18, 2018 at 12:44 AM, Ralph Goers <ra...@dslextreme.com>
wrote:
> While splitting out components into modules should have had the desired
> effect of reducing the dependencies in log4j-core, it didn’t help making
> the builds take less time. The Jenkins build for 3.x took about 43 minutes
> while the 2.x build took 38 minutes.
>
I am continuing on this path by implementing
https://issues.apache.org/jira/browse/LOG4J2-2237 "Move Jackson-based
layouts to their own modules."
There is obviously some extra work the build needs to do for each
additional module, which explains the longer build.
Gary
>
> Ralph
>