hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Busbey <bus...@cloudera.com>
Subject Re: Testing and CI -- Apache Jenkins Builds (WAS -> Re: Testing)
Date Sun, 14 Jun 2015 01:28:32 GMT
sigh. that should have been "is now a matrix build".

On Sat, Jun 13, 2015 at 7:31 PM, Sean Busbey <busbey@cloudera.com> wrote:

> The HBase-1.2 build[1] is not a matrix build that runs JDK7 and JDK8 in
> parallel.
>
> I saved the old job for now and left it disabled[2].
>
>
> [1]: https://builds.apache.org/view/H-L/view/HBase/job/HBase-1.2/
> [2]:
> https://builds.apache.org/view/H-L/view/HBase/job/HBase-1.2%20non-matrix/
>
> On Fri, Apr 10, 2015 at 3:42 PM, Sean Busbey <busbey@cloudera.com> wrote:
>
>> Sorry for the noise. I also updated the checkstyle step on HBase-TRUNK
>>
>> from
>>     mvn checkstyle:checkstyle
>> to
>>     mvn -DskipTests package checkstyle:checkstyle
>>
>> to deal with the same issue. Looks all clear now.
>>
>> On Fri, Apr 10, 2015 at 3:03 PM, Sean Busbey <busbey@cloudera.com> wrote:
>>
>>> Updated the following builds:
>>>
>>> * HBase-TRUNK
>>>
>>> moved from
>>>
>>>     mvn clean compile findbugs:findbugs
>>>
>>> to
>>>
>>>    mvn clean -DskipTests package findbugs:findbugs
>>>
>>> To work around the known issue where we can't do a bootstrap compile
>>> without previous install or remote SNAPSHOT artifacts. (recently triggered
>>> by the addition of the procedure module on master)
>>>
>>> On Mon, Mar 30, 2015 at 8:49 PM, 张铎 <palomino219@gmail.com> wrote:
>>>
>>>> Make findbugs and checkstyle plugins always run.
>>>> The default behavior only publishes result for stable builds, but
>>>> master is
>>>> not always stable and sometimes we introduce new warnings in unstable
>>>> builds(see builds 6310-6314).
>>>> And findbugs and checkstyle will not fail unless we abort the building
>>>> process I think, so it is safe to turn it on every time.
>>>>
>>>> Thanks.
>>>>
>>>> 2015-03-22 22:44 GMT+08:00 Ted Yu <yuzhihong@gmail.com>:
>>>>
>>>> > +1 on letting HBase-TRUNK jenkins show coverage report.
>>>> >
>>>> > Cheers
>>>> >
>>>> > On Sun, Mar 22, 2015 at 5:51 AM, 张铎 <palomino219@gmail.com>
wrote:
>>>> >
>>>> > > Going to change the config of HBase-TRUNK jenkins to show findbugs,
>>>> > > checkstyle and jacoco coverage report.
>>>> > > The config has been tested on
>>>> > > https://builds.apache.org/job/HBase-TRUNK-jacoco for nearly 30
>>>> times.
>>>> > Not
>>>> > > much different from HBase-TRUNK unless it runs ~30% slower(the
>>>> overhead
>>>> > of
>>>> > > collecting information for code coverage).
>>>> > > Thanks.
>>>> > >
>>>> > > 2015-03-12 5:08 GMT+08:00 Andrew Purtell <apurtell@apache.org>:
>>>> > >
>>>> > > > +1, thanks a lot for improving our build hygiene.
>>>> > > >
>>>> > > > On Wed, Mar 11, 2015 at 2:01 PM, Enis Söztutar <
>>>> enis.soz@gmail.com>
>>>> > > wrote:
>>>> > > >
>>>> > > > > Thanks Sean. This is great.
>>>> > > > >
>>>> > > > > Enis
>>>> > > > >
>>>> > > > > On Wed, Mar 11, 2015 at 1:54 PM, Sean Busbey <
>>>> busbey@cloudera.com>
>>>> > > > wrote:
>>>> > > > >
>>>> > > > > > FYI I just finished chasing down the breakage for
mvn site on
>>>> all
>>>> > > patch
>>>> > > > > > builds.
>>>> > > > > >
>>>> > > > > > HBASE-13191 consolidates the few places in the test-patch
code
>>>> > where
>>>> > > we
>>>> > > > > > hard coded MAVEN_OPTS.
>>>> > > > > >
>>>> > > > > > If you look at the PreCommit job now, we use the
"set
>>>> environment
>>>> > > > > > variables" option to set MAVEN_OPTS and then everything
else
>>>> > respects
>>>> > > > > that
>>>> > > > > > setting.
>>>> > > > > >
>>>> > > > > > I set the initial value to be a combination of the
memory
>>>> > limitations
>>>> > > > > we've
>>>> > > > > > been actually running with (the ~6G was getting
ignored) and
>>>> the
>>>> > > > permgen
>>>> > > > > > needed for site.
>>>> > > > > >
>>>> > > > > > MAVEN_OPTS=-Xmx3100M -XX:-UsePerfData -XX:MaxPermSize=256m
>>>> > > > > >
>>>> > > > > > On Mon, Feb 23, 2015 at 11:46 PM, Stack <stack@duboce.net>
>>>> wrote:
>>>> > > > > >
>>>> > > > > > > I just made TRUNK and branch-1 builds use same
jvm as
>>>> patch-build
>>>> > > > > > > (hadoopqa) -- i.e. jdku51 -- and I set the
MAVEN_OPTS to be
>>>> the
>>>> > > same
>>>> > > > as
>>>> > > > > > > those of trunk build too, setting MAVEN_OPTS="-Xmx6100m"...
>>>> it
>>>> > had
>>>> > > > been
>>>> > > > > > > 3000.
>>>> > > > > > >
>>>> > > > > > > Yours,
>>>> > > > > > > St.Ack
>>>> > > > > > >
>>>> > > > > > > On Wed, Dec 31, 2014 at 4:22 PM, Stack <stack@duboce.net>
>>>> wrote:
>>>> > > > > > >
>>>> > > > > > > > I upped hadoopqa retention to keep last
100 builds and or
>>>> last
>>>> > 7
>>>> > > > > days,
>>>> > > > > > > > whichever comes first.
>>>> > > > > > > > St.Ack
>>>> > > > > > > >
>>>> > > > > > > > On Tue, Nov 4, 2014 at 9:38 AM, Stack
<stack@duboce.net>
>>>> > wrote:
>>>> > > > > > > >
>>>> > > > > > > >> Branch-1 and master have stabilized
and now run mostly
>>>> blue
>>>> > > (give
>>>> > > > or
>>>> > > > > > > take
>>>> > > > > > > >> the odd failure) [1][2]. Having a
mostly blue branch-1
>>>> has
>>>> > > helped
>>>> > > > us
>>>> > > > > > > >> identify at least one destabilizing
commit in the last
>>>> few
>>>> > days,
>>>> > > > > maybe
>>>> > > > > > > two;
>>>> > > > > > > >> this is as it should be (smile).
>>>> > > > > > > >>
>>>> > > > > > > >> Lets keep our builds blue. If you
commit a patch, make
>>>> sure
>>>> > > > > subsequent
>>>> > > > > > > >> builds stay blue. You can subscribe
to
>>>> > builds@hbase.apache.org
>>>> > > to
>>>> > > > > get
>>>> > > > > > > >> notice of failures if not already
subscribed.
>>>> > > > > > > >>
>>>> > > > > > > >> Thanks,
>>>> > > > > > > >> St.Ack
>>>> > > > > > > >>
>>>> > > > > > > >> 1.
>>>> > https://builds.apache.org/view/H-L/view/HBase/job/HBase-1.0/
>>>> > > > > > > >> 2.
>>>> > > https://builds.apache.org/view/H-L/view/HBase/job/HBase-TRUNK/
>>>> > > > > > > >>
>>>> > > > > > > >>
>>>> > > > > > > >> On Mon, Oct 13, 2014 at 4:41 PM, Stack
<stack@duboce.net
>>>> >
>>>> > > wrote:
>>>> > > > > > > >>
>>>> > > > > > > >>> A few notes on testing.
>>>> > > > > > > >>>
>>>> > > > > > > >>> Too long to read, infra is more
capable now and after
>>>> some
>>>> > > work,
>>>> > > > we
>>>> > > > > > are
>>>> > > > > > > >>> seeing branch-1 and trunk mostly
running blue. Lets try
>>>> and
>>>> > > keep
>>>> > > > it
>>>> > > > > > > this
>>>> > > > > > > >>> way going forward.
>>>> > > > > > > >>>
>>>> > > > > > > >>> Apache Infra has new, more capable
hardware.
>>>> > > > > > > >>>
>>>> > > > > > > >>> A recent spurt of test fixing
combined with more capable
>>>> > > hardware
>>>> > > > > > seems
>>>> > > > > > > >>> to have gotten us to a new place;
tests are mostly
>>>> passing
>>>> > now
>>>> > > on
>>>> > > > > > > branch-1
>>>> > > > > > > >>> and master.  Lets try and keep
it this way and start to
>>>> trust
>>>> > > our
>>>> > > > > > test
>>>> > > > > > > runs
>>>> > > > > > > >>> again.  Just a few flakies remain.
 Lets try and nail
>>>> them.
>>>> > > > > > > >>>
>>>> > > > > > > >>> Our tests now run in parallel
with other test suites
>>>> where
>>>> > > > previous
>>>> > > > > > we
>>>> > > > > > > >>> ran alone. You can see this sometimes
when our zombie
>>>> > detector
>>>> > > > > > reports
>>>> > > > > > > >>> tests from another project altogether
as lingerers (To
>>>> be
>>>> > > fixed).
>>>> > > > > > > Some of
>>>> > > > > > > >>> our tests are failing because
a concurrent hbase run is
>>>> > undoing
>>>> > > > > > > classes and
>>>> > > > > > > >>> data from under it. Also, lets
fix.
>>>> > > > > > > >>>
>>>> > > > > > > >>> Our tests are brittle. It takes
75minutes for them to
>>>> > complete.
>>>> > > > > Many
>>>> > > > > > > >>> are heavy-duty integration tests
starting up multiple
>>>> > clusters
>>>> > > > and
>>>> > > > > > > >>> mapreduce all in the one JVM.
It is a miracle they pass
>>>> at
>>>> > all.
>>>> > > > > > > Usually
>>>> > > > > > > >>> integration tests have been cast
as unit tests because
>>>> there
>>>> > > was
>>>> > > > no
>>>> > > > > > > where
>>>> > > > > > > >>> else for them to get an airing.
 We have the hbase-it
>>>> suite
>>>> > now
>>>> > > > > which
>>>> > > > > > > would
>>>> > > > > > > >>> be a more apt place but until
these are run on a regular
>>>> > basis
>>>> > > in
>>>> > > > > > > public
>>>> > > > > > > >>> for all to see, the fat integration
tests disguised as
>>>> unit
>>>> > > tests
>>>> > > > > > will
>>>> > > > > > > >>> remain.  A review of our current
unit tests weeding the
>>>> old
>>>> > > cruft
>>>> > > > > and
>>>> > > > > > > the
>>>> > > > > > > >>> no longer relevant or duplicates
would be a nice
>>>> undertaking
>>>> > if
>>>> > > > > > > someone is
>>>> > > > > > > >>> looking to contribute.
>>>> > > > > > > >>>
>>>> > > > > > > >>> Alex Newman has been working on
making our tests work
>>>> up on
>>>> > > > travis
>>>> > > > > > and
>>>> > > > > > > >>> circle-ci.  That'll be sweet when
it goes end-to-end.
>>>> He
>>>> > also
>>>> > > > > added
>>>> > > > > > in
>>>> > > > > > > >>> some "type" categorizations --
client, filter,
>>>> mapreduce --
>>>> > > > > alongside
>>>> > > > > > > our
>>>> > > > > > > >>> old "sizing" categorizations of
small/medium/large.  His
>>>> > > thinking
>>>> > > > > is
>>>> > > > > > > that
>>>> > > > > > > >>> we can run these categorizations
in parallel so we
>>>> could run
>>>> > > the
>>>> > > > > > total
>>>> > > > > > > >>> suite in about the time of the
longest test, say
>>>> > 20-30minutes?
>>>> > > > We
>>>> > > > > > > could
>>>> > > > > > > >>> even change Apache to run them
this way.
>>>> > > > > > > >>>
>>>> > > > > > > >>> FYI,
>>>> > > > > > > >>> St.Ack
>>>> > > > > > > >>>
>>>> > > > > > > >>>
>>>> > > > > > > >>>
>>>> > > > > > > >>>
>>>> > > > > > > >>>
>>>> > > > > > > >>>
>>>> > > > > > > >>>
>>>> > > > > > > >>
>>>> > > > > > > >
>>>> > > > > > >
>>>> > > > > >
>>>> > > > > >
>>>> > > > > >
>>>> > > > > > --
>>>> > > > > > Sean
>>>> > > > > >
>>>> > > > >
>>>> > > >
>>>> > > >
>>>> > > >
>>>> > > > --
>>>> > > > Best regards,
>>>> > > >
>>>> > > >    - Andy
>>>> > > >
>>>> > > > Problems worthy of attack prove their worth by hitting back.
-
>>>> Piet
>>>> > Hein
>>>> > > > (via Tom White)
>>>> > > >
>>>> > >
>>>> >
>>>>
>>>
>>>
>>>
>>> --
>>> Sean
>>>
>>
>>
>>
>> --
>> Sean
>>
>
>
>
> --
> Sean
>



-- 
Sean

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message