[Building Sakai] Jenkins deployment issue

John Bush jbush at anisakai.com
Fri Apr 25 10:51:27 PDT 2014


We use jenkins to create tag, branches, builds, and release tar balls
of Sakai which we upload into s3.  We don't use jenkins to deploy.

We use puppet for configuration management which lays down a deploy
script onto our app nodes, the deploy script grabs the release from s3
and then cleanup out tomcat, redeploys and recycles it.  Puppet
manages everything on the box, we start with a clean base image and
puppet lays down everything else, tomcat, mysql, apache, keys for
various people, cron jobs, scripts, network conf, monitoring, and
config for all of it.  It's 100% automated, across 1 or 10 nodes.
Puppet determines what type of box it is (db, app, utility, load
balancer, etc) by the name of the machine. So we can do single
instance boxes that run the full stack or split everything out into
separate boxes.

We've used capistrano in the past for co-ordinating deployments on
multiple nodes at the same time, but are longer term plan is to use
rundeck for that.  There are good rundeck/jenkins plugins.  IMHO,
jenkins is a great as a continuous integration build server, but there
are far better tools out there for deployment and configuration
management.  These things are really separate concerns, but if you
don't have the scale that requires using multiple tools, or the
resources to execute on it, you can get pretty far with just jenkins.

On Fri, Apr 25, 2014 at 9:36 AM, Qu, Yuanhua <yq12 at txstate.edu> wrote:
> We, Txstate University, have fully build/deploy our local sakai "TRACS"
> instance on dev environment and plan to do the similar thing on staging
> environment.
>
> Basically, it builds up all the binaries including tomcat servers and
> other setups to make a tar ball and then deploy to our dev server and
> untar it and start the server.
>
> We haven't use Jenkins for production deployment though.
>
> -Qu
>
> On 4/25/14 9:49 AM, "Liu, Peter" <peter.liu at yale.edu> wrote:
>
>>Hi Mathew,
>>
>>Thanks a lot for your comment. At Yale, we try to set up a Jenkin's
>>Build/deployment with artiFactory in a fully automatically way.
>>Currently, we have encountered quiet few obstacles since the Size of
>>Sakai are so big.
>>
>>Now, I am just wondering which institution has set up such a fully
>>Jenkin's build/deploy environment.
>>
>>Thanks,
>>Peter
>>
>>-----Original Message-----
>>From: Matthew Buckett [mailto:matthew.buckett at oucs.ox.ac.uk]
>>Sent: Friday, April 25, 2014 4:55 AM
>>To: Liu, Peter
>>Cc: sakai-dev; Neal Caidin
>>Subject: Re: Jenkins deployment issue
>>
>>On 24 April 2014 18:42, Liu, Peter <peter.liu at yale.edu> wrote:
>>>
>>> Should we just to pre-build the entire sakai-binary tar file and then
>>> un-tar it instead to run the sakai:deploy (take too long)?
>>
>>Locally we take the Sakai .zip and put it into a debian package. Then
>>when we come todo the upgrade we just upgrade the version of the debian
>>package. You can have the debian package upgrade scripts also do the
>>service restarting as well although I don't believe we use this.
>>
>>> Any suggestion (or how your institution does the deployment via
>>> Jenkins) will be highly appreciated.
>>
>>We do use jenkins todo continuous  integration, but we don't have it make
>>our final builds or package them.
>>
>>--
>>  Matthew Buckett, VLE Developer, LTG, Oxford University Computing
>>Services
>>_______________________________________________
>>sakai-dev mailing list
>>sakai-dev at collab.sakaiproject.org
>>http://collab.sakaiproject.org/mailman/listinfo/sakai-dev
>>
>>TO UNSUBSCRIBE: send email to
>>sakai-dev-unsubscribe at collab.sakaiproject.org with a subject of
>>"unsubscribe"
>
> _______________________________________________
> sakai-dev mailing list
> sakai-dev at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/sakai-dev
>
> TO UNSUBSCRIBE: send email to sakai-dev-unsubscribe at collab.sakaiproject.org with a subject of "unsubscribe"



-- 
John Bush
602-490-0470

** This message is neither private nor confidential in fact the US
government is storing it in a warehouse located in Utah for future
data mining use cases should they arise. **


More information about the sakai-dev mailing list