Closed ghoneycutt closed 7 years ago
Fixes #77
Need to address #50 before merging. This is to remove inheritance.
Fixes #32
Fixes #42
Let's try to get @nalyanyam onboard, he is our Solaris'n'Puppet specialist. @nalyanyam hope you have some time to help :)
Installation on which Solaris release? 10 or 11? Solaris 10 should already be having the packages installed as part of the base OS. Probably what's causing the puppet not to detect the installed packages is the package provider. By default Solaris 10 uses the legacy provider "sun" while Solaris 11 uses the provider "pkg" and for Solaris 11 it's not necessary to provide adminfile as long as the repositories are properly configured.
Hi @nalyanyam
The issue is not with nfs, but with the testing framework. I'm attempting to install puppet enterprise and that package is prompting for user input. The goal is that you can do vagrant up
with the solaris10-client and solaris11-client systems without any user input, like the other platforms.
Check out theses two lines from the vagrant/provision_solaris.sh
script.
@ghoneycutt The correct command syntax is something like below: pkgadd -n -a /var/sadm/install/admin/puppet -d /vagrant/.puppetagents/${pkg} all
This assumes that ${pkg} is a datastream package. And the adminfile should also be ok. The original syntax assumed a device instead of a package. Please see below notes from the man page. Please test and let me know in case that fails
From man page:
By default, pkgadd looks in the /var/spool/pkg directory when searching for instances of a package to install or spool. Optionally, the source for the package instances to be installed or spooled can be specified using:
-d device
-d datastream pkgname,... | all
@Phil-Friderici Having a hard time getting the installation to happen without intervention on Solaris, even though I'm providing an adminfile. Any ideas?