This looks better then my workaround of having long bash command to log
"something" every 10 seconds. That worked fine as well, but didn't
really look pretty :-)
BTV. instead of looking at another alternative for Travis, we can maybe
spend the time to tune our testsuite? We have already JIRA for it
.
The advantage is, that it will help with the speed of build on our
laptops too, not just with the speed of CI. I've personally ended to
always run the build with "-DskipTests=true" on my laptop as it takes
forever to build with tests.
Marek
On 17/08/16 08:46, Stian Thorgersen wrote:
I've tweaked the stroppy toddler a bit
(
https://github.com/keycloak/keycloak/pull/3147). Ran that PR 5 times
and it passed every time, so hopefully it'll make it a bit more
stable. Still takes more than 1 hour to complete though.
On 16 August 2016 at 16:06, Bruno Oliveira <bruno(a)abstractj.org
<mailto:bruno@abstractj.org>> wrote:
There are some alternatives that we could try:
*
https://www.cloudbees.com/partners/platform/red-hat
<
https://www.cloudbees.com/partners/platform/red-hat>
*
https://developers.openshift.com/managing-your-applications/continuous-in...
<
https://developers.openshift.com/managing-your-applications/continuous-in...
* Make use of a beefy machine :)
* Or host Jenkins in some paid cloud infrastructure
If we don't have any volunteers, count me in.
On 2016-08-15, Stian Thorgersen wrote:
> My opinion is that i HATE Travis! It's slow and has a tendency
to act like
> a stroppy toddler.
>
> Removing -q doesn't work because we then end up generating more
log than
> Travis allows which kills the build as well.
>
> We used to have the caching enabled, but it used to result in a
lot of
> issues. Can't quite remember the details though. Maybe we can
try it again.
>
> There's also travis_wait option that could work (
>
https://docs.travis-ci.com/user/common-build-problems/#Build-times-out-be...
<
https://docs.travis-ci.com/user/common-build-problems/#Build-times-out-be...
> ).
>
> My preference would be to migrate away from Travis and use
something more
> beefy that can run tests quicker.
>
> I'll need to figure out someone to look into it though. Any
volunteers?
>
>
>
>
> On 28 July 2016 at 17:11, Pedro Igor Silva <psilva(a)redhat.com
<mailto:psilva@redhat.com>> wrote:
>
> > Let's see what Stian thinks about it :) For me, it is also an
option.
> >
> > ----- Original Message -----
> > From: "Bruno Oliveira" <bruno(a)abstractj.org
<mailto:bruno@abstractj.org>>
> > To: "Pedro Igor Silva" <psilva(a)redhat.com
<mailto:psilva@redhat.com>>
> > Cc: "Bill Burke" <bburke(a)redhat.com
<mailto:bburke@redhat.com>>, "keycloak-dev" <
> > keycloak-dev(a)lists.jboss.org
<mailto:keycloak-dev@lists.jboss.org>>
> > Sent: Wednesday, July 27, 2016 10:45:35 PM
> > Subject: Re: [keycloak-dev] travis seems to be down
> >
> > Another alternative, but you might not enjoy it, is :)
> >
> > Before start building Keycloak, pre-download all the '.m2' dir
with
> > the dependencies required. Second alternative, is to try some
caching[1],
> > but it may
> > or may not work.
> >
> > [1] -
https://docs.travis-ci.com/user/caching/
<
https://docs.travis-ci.com/user/caching/>
> >
> > On 2016-07-27, Pedro Igor Silva wrote:
> > > As Bruno suggested, the problem is probably related with the
build
> > taking too long. If you look at the first command being executed:
> > >
> > > mvn install -Pdistribution -DskipTests=true -B -V -q
> > >
> > > The -q option tells maven to run in quite mode and only show
errors.
> > Even in my laptop, if I run the command above it takes a bit
more time to
> > show any output (not too long, but more than usual). That may
explain that
> > message from Travis.
> > >
> > > Maybe we should remove -q and just log everything. That
could make the
> > builds more stable.
> > >
> > > I've forced a new build to your
https://github.com/keycloak/
> > keycloak/pull/3079 (without changing anything to .travis.yml).
Now it is
> > running.
> > >
> > > ----- Original Message -----
> > > From: "Bill Burke" <bburke(a)redhat.com
<mailto:bburke@redhat.com>>
> > > To: "Pedro Igor Silva" <psilva(a)redhat.com
<mailto:psilva@redhat.com>>, "Bruno Oliveira" <
> > bruno(a)abstractj.org <mailto:bruno@abstractj.org>>
> > > Cc: "keycloak-dev" <keycloak-dev(a)lists.jboss.org
<mailto:keycloak-dev@lists.jboss.org>>
> > > Sent: Wednesday, July 27, 2016 8:53:16 PM
> > > Subject: Re: [keycloak-dev] travis seems to be down
> > >
> > > Wrong... :( I still get this problem.
> > >
> > >
> > > On 7/27/16 7:02 PM, Pedro Igor Silva wrote:
> > > > I think Travis decided to not mess with Bill :)
> > > >
> > > > ----- Original Message -----
> > > > From: "Bruno Oliveira" <bruno(a)abstractj.org
<mailto:bruno@abstractj.org>>
> > > > To: "Pedro Igor Silva" <psilva(a)redhat.com
<mailto:psilva@redhat.com>>
> > > > Cc: "Bill Burke" <bburke(a)redhat.com
<mailto:bburke@redhat.com>>, "keycloak-dev" <
> > keycloak-dev(a)lists.jboss.org
<mailto:keycloak-dev@lists.jboss.org>>
> > > > Sent: Wednesday, July 27, 2016 6:37:44 PM
> > > > Subject: Re: [keycloak-dev] travis seems to be down
> > > >
> > > > Seems like the balance of the Force was established again.
> > > >
> > > > On 2016-07-27, Pedro Igor Silva wrote:
> > > >> Got that error, but now the build was successful.
> > > >>
> > > >> ----- Original Message -----
> > > >> From: "Bruno Oliveira" <bruno(a)abstractj.org
<mailto:bruno@abstractj.org>>
> > > >> To: "Bill Burke" <bburke(a)redhat.com
<mailto:bburke@redhat.com>>
> > > >> Cc: "keycloak-dev"
<keycloak-dev(a)lists.jboss.org
<mailto:keycloak-dev@lists.jboss.org>>
> > > >> Sent: Wednesday, July 27, 2016 6:01:47 PM
> > > >> Subject: Re: [keycloak-dev] travis seems to be down
> > > >>
> > > >> Seems like they are fully operational[1]. I believe that
this is the
> > > >> reason:
> > > >>
> > > >> "No output has been received in the last 10 minutes,
this
potentially
> > indicates a stalled build or something wrong with the build
itself."
> > > >>
> > > >> Based on my previous experiences, it happens when the
build takes too
> > > >> long. Not really sure if that's the root cause, but I
can
help
> > > >> investigating it.
> > > >>
> > > >> [1] -
https://www.traviscistatus.com/
<
https://www.traviscistatus.com/>
> > > >>
> > > >> On 2016-07-27, Bill Burke wrote:
> > > >>> A lot of builds failing in initial setup
> > > >>>
> > > >>>
> > > >>> Bill
> > > >>>
> > > >>> _______________________________________________
> > > >>> keycloak-dev mailing list
> > > >>> keycloak-dev(a)lists.jboss.org
<mailto:keycloak-dev@lists.jboss.org>
> > > >>>
https://lists.jboss.org/mailman/listinfo/keycloak-dev
<
https://lists.jboss.org/mailman/listinfo/keycloak-dev>
> > > >> --
> > > >>
> > > >> abstractj
> > > >> PGP: 0x84DC9914
> > > >> _______________________________________________
> > > >> keycloak-dev mailing list
> > > >> keycloak-dev(a)lists.jboss.org
<mailto:keycloak-dev@lists.jboss.org>
> > > >>
https://lists.jboss.org/mailman/listinfo/keycloak-dev
<
https://lists.jboss.org/mailman/listinfo/keycloak-dev>
> > > > --
> > > >
> > > > abstractj
> > > > PGP: 0x84DC9914
> > >
> >
> > --
> >
> > abstractj
> > PGP: 0x84DC9914
> > _______________________________________________
> > keycloak-dev mailing list
> > keycloak-dev(a)lists.jboss.org <mailto:keycloak-dev@lists.jboss.org>
> >
https://lists.jboss.org/mailman/listinfo/keycloak-dev
<
https://lists.jboss.org/mailman/listinfo/keycloak-dev>
> >
--
abstractj
PGP: 0x84DC9914
_______________________________________________
keycloak-dev mailing list
keycloak-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/keycloak-dev