On Mon, Feb 24, 2014 at 1:13 AM, Yogi Nerkar <yoginer...@gmail.com> wrote:
> Thanks a lot lot this gives me good view ,appreciate the reply, i am am
> trying to find the architecture diagram and details on how to setup this
>
> If you have anything please help.
There aren't that many hard requirements other than the master and
slaves being able to connect over a network - and you will want to
have ntp set up to keep all the times in sync.
With linux slaves, I normally just add a 'jenkins' user, and configure
launch method as 'Launch slave agents on Unix machines via SSH'
because that takes care of everything for you. It may be useful to
put some common libraries and tools on an nfs-mounted file system to
save the trouble of keeping them updated across the slaves. We
normally include a build script with the source that will work the
same on a developers desktop or the build slaves with any differences
covered by the tool locations and environment settings in jenkins.
Even if you only have one type of build target to start, I'd recommend
giving the slaves one or more labels to identify that type, and set
the 'restrict where this job can run' on all jobs to tie them to a
label. That way when/if you add additional build targets the existing
jobs will still run in the right places and you still get
balancing/failover across matching types.
--
Les Mikesell
lesmikes...@gmail.com
--
You received this message because you are subscribed to the Google Groups
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to jenkinsci-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
