Could you cheat and create /opt2/puppetlabs and a symlink
from /opt/puppetlabs -> /opt2/puppetlabs before installation?
Otherwise, you can likely unpack the rpm, and futz with it,
but I could imagine that /opt/puppetlabs is buried in a
binary or two and thus awkward to change.
Hope that helps - c
Ah, you need the whole thing to be relocatable!
Yeah, that's much more difficult, even with a SRPM to be honest depending
on how the components are built.
I think that it's configurable in Vanagon so you can probably build the
container image, hop into it and dig around to see where you might hav
Thank you! Unfortunately that won't work for me. The issue I have is
due to /opt is some NFS read-only mounts. I am wondering about how to
get a new RPM created for puppet-agent6, so that it could be installed
on some folder like /opt2 instead. Right now I could not install this
rpm. The other work
This community hosted project contains dockerfiles that outline the steps
required for building from source
https://github.com/simp/releng-misc/tree/master/puppet_agent_build
On Mon, Apr 20, 2020 at 2:22 PM Morgan Rhodes wrote:
> Hi,
>
> We no longer provide source RPMs for puppet6 packages due
Hi Team,
Though i have executed the puppet task successfully. Unable to call the
plan with my ruby task from console getting the below task:
Error: Task exited 127:
/usr/bin/env: ‘powershell’: No such file or directory.
>From bolt command process has been started but no progress it has stuck at
Bram and Konrad,
I'm personally interested in your perspective on multi-node testing.
In my experience with our stack of around 100 modules, I've found that
single node testing is marginally better than spec testing in that it tells
you if a service actually starts. However, my project has found
