I extended my travis experiments and also activated support for
pax-wicket [1]. Still pretty surprised how straight forward this
works...

Would be cool if someone could try cloudbee for one or two other
projects that we get a feeling how it works out there... I might have
the time to give it a shot at the weekend otherwise.

Kind regards,
Andreas

[1] http://travis-ci.org/#!/ops4j/org.ops4j.pax.wicket

On Thu, Mar 15, 2012 at 16:52, Andreas Pieber <[email protected]> wrote:
> definitely, that way you avoid the necessarity to bisect in case such
> a collection of commits fail.
>
> Kind regards,
> Andreas
>
> On Thu, Mar 15, 2012 at 16:30, Jimmi Dyson <[email protected]> wrote:
>> Would you really want to run a build for every single commit?
>>
>> Cheers,
>> Jimmi
>>
>>
>> On 15 March 2012 11:07, Andreas Pieber <[email protected]> wrote:
>>>
>>> On Thu, Mar 15, 2012 at 10:51, Jimmi Dyson <[email protected]>
>>> wrote:
>>> > If you wanted to remain on Jenkins but move away from self-hosting,
>>> > Cloudbees (http://www.cloudbees.com/) is a pretty good option, depending
>>> > on
>>> > your utilisation. They offer free cloud hosting of Jenkins to open
>>> > source
>>> > projects (http://www.cloudbees.com/foss/index.cb), although obviously
>>> > with
>>> > resource constraints. I have set this up with some projects I have in
>>> > Github
>>> > & all works nicely.
>>> >
>>> > Just an alternative, although I must say http://travis-ci.org looks
>>> > pretty
>>> > interesting.
>>>
>>> I like the simplicity and the the fact that every build is run in a
>>> completely new VM; this helps to discover problems which would have
>>> gone unnoticed otherwise... (like not public available snapshots, old
>>> versions, ...
>>>
>>> >
>>> > Not quite sure what you mean by:
>>> >
>>> >
>>> > "compared to hudson where you need to setup a build seperately and the
>>> > single commits are gathered together till no service is available and
>>> > are
>>> > build afterwards all together"
>>>
>>> typically you push; then jenkins/hudson starts a build; now you push
>>> and push and push while hudson is still building; once hudson is
>>> finished he'll get your changes and build again. The disadvantage is
>>> that he'll test all changes you've pushed in the meantime at once;
>>> which means you don't get if you break the build in between. travis
>>> starts a new build for each and every commit :-)
>>>
>>> >
>>> > The Git plugin for Jenkins can build all branches & can also perform
>>> > pre-tested commits from integration branches which we've started using
>>> > at
>>> > work. You can also send post-commit hooks to trigger builds
>>> > automatically.
>>> > Maybe I've misunderstood what you mean?
>>> >
>>> > Thanks,
>>> > Jimmi
>>> >
>>> > On 14 March 2012 19:49, Andreas Pieber <[email protected]> wrote:
>>> >>
>>> >> Hey guys,
>>> >>
>>> >> Because of the current problems with our CI&T system we did some time
>>> >> searching what we could find of hosted CI systems. After various ppl
>>> >> thinking travis-ci.org for a good choice Achim and I invested some
>>> >> time today to evaluate that service and setup pax-web today [1]. And
>>> >> TBH we're pretty impressed. Anyone with owner permissions to ops4j can
>>> >> simply setup a travis hook to an ops4j project which will
>>> >> automatically setup a build for the project. Once this is done the
>>> >> entire configuration is done via a ".travis.yml" file in the root of
>>> >> the repository. [2] is the pax-web one for example. As you see this
>>> >> one is pretty simple to create an maintain. Travis automatically
>>> >> builds all branches and every commit (compared to hudson where you
>>> >> need to setup a build seperately and the single commits are gathered
>>> >> together till no service is available and are build afterwards all
>>> >> together). This will mean you'll get the EXACT commit where you break
>>> >> the build; even if pushing many changes in a short time frame. Another
>>> >> advantage is that each build is started in a completely fresh VM which
>>> >> means that also dependency problems between projects are identified.
>>> >>
>>> >> Nevertheless, there is also one big disadvantage. There is no
>>> >> possibility (as far as I see) to get a key file into the system to
>>> >> deploy snapshots to sonatype for each build. Since there is no
>>> >> surviving .m2 repo (and no snapshots) we'll need to either deploy
>>> >> snapshots manually or extend the install script to checkout the right
>>> >> branch of the related projects and build them separately (without
>>> >> tests) before we start to build/test the main repo.
>>> >>
>>> >> Nevertheless, after seeing all the advantages of travis-ci I would
>>> >> almost accept this single drawback. At least this will allow us to
>>> >> drop another self hosted service from the list... One vulnerability
>>> >> less!
>>> >>
>>> >> WDYT?
>>> >>
>>> >> Kind regards,
>>> >> Andreas
>>> >>
>>> >> [1] http://travis-ci.org/#!/ops4j/org.ops4j.pax.web
>>> >> [2] https://github.com/ops4j/org.ops4j.pax.web/blob/master/.travis.yml
>>> >>
>>> >> _______________________________________________
>>> >> general mailing list
>>> >> [email protected]
>>> >> http://lists.ops4j.org/mailman/listinfo/general
>>> >
>>> >
>>> > ________________________________
>>> >
>>> > NOTICE: This message contains privileged and confidential information
>>> > intended only for the addressee. If you have received this message in
>>> > error
>>> > you must not disseminate, copy or take action on it; please notify
>>> > [email protected] Opinions expressed in this message
>>> > are
>>> > those of the sender and do not necessarily represent those of
>>> > Specsavers.
>>> > Although this e-mail and any attachments are believed to be virus free,
>>> > e-mail communications are not 100% secure and Specsavers makes no
>>> > warranty
>>> > that this message is secure or virus free. All references to Specsavers
>>> > means Specsavers Optical Superstores Limited, a company limited by
>>> > shares
>>> > and registered in England under number 1721624 of Forum 6, Parkway,
>>> > Solent
>>> > Business Park, Whiteley, Fareham, Hampshire, PO15 7PA. Nothing in this
>>> > transmission shall or shall be deemed to constitute an offer or
>>> > acceptance
>>> > of an offer or otherwise have the effect of forming a contract by
>>> > electronic
>>> > communication. Your name and address may be stored to facilitate
>>> > communication.
>>> >
>>> > ________________________________
>>> > .
>>> > _______________________________________________
>>> > general mailing list
>>> > [email protected]
>>> > http://lists.ops4j.org/mailman/listinfo/general
>>> >
>>>
>>> _______________________________________________
>>> general mailing list
>>> [email protected]
>>> http://lists.ops4j.org/mailman/listinfo/general
>>
>>
>> ________________________________
>>
>> NOTICE: This message contains privileged and confidential information
>> intended only for the addressee. If you have received this message in error
>> you must not disseminate, copy or take action on it; please notify
>> [email protected] Opinions expressed in this message are
>> those of the sender and do not necessarily represent those of Specsavers.
>> Although this e-mail and any attachments are believed to be virus free,
>> e-mail communications are not 100% secure and Specsavers makes no warranty
>> that this message is secure or virus free. All references to Specsavers
>> means Specsavers Optical Superstores Limited, a company limited by shares
>> and registered in England under number 1721624 of Forum 6, Parkway, Solent
>> Business Park, Whiteley, Fareham, Hampshire, PO15 7PA. Nothing in this
>> transmission shall or shall be deemed to constitute an offer or acceptance
>> of an offer or otherwise have the effect of forming a contract by electronic
>> communication. Your name and address may be stored to facilitate
>> communication.
>>
>> ________________________________
>> .
>>
>> _______________________________________________
>> general mailing list
>> [email protected]
>> http://lists.ops4j.org/mailman/listinfo/general
>>

_______________________________________________
general mailing list
[email protected]
http://lists.ops4j.org/mailman/listinfo/general

Reply via email to