@elukey reported via IRC:
hey folks, I am bootstrapping a Cassandra cluster and getting Error: /Stage[main]/Cassandra::Twcs/Scap::Target[cassandra/twcs]/Package[cassandra/twcs]: Provider scap3 is not functional on this host
I suspect that this is because /usr/bin/scap was not functional on the host at the point this executed, possibly because puppet had not recently run on deploy1002.
modules/cassandra/manifests/twcs.pp
class cassandra::twcs(
) {
require ::cassandra
scap::target { 'cassandra/twcs':
deploy_user => 'deploy-service',
manage_user => true,
}
}modules/scap/manifests/target.pp
...
package { $package_name:
ensure => $ensure,
install_options => [{
owner => $deploy_user
}],
provider => 'scap3',
require => $require_package,
}
...modules/scap/spec/types/package/scap3_spec.rb
...
provider_class = Puppet::Type.type(:package).provider(:scap3)
describe provider_class do
before do
@resource = Puppet::Type.type(:package).new(name: 'foo/deploy')
@provider = provider_class.new(@resource)
@resource.provider = @provider
@resource[:install_options] = [{ 'owner' => 'mwdeploy' }]
# Stub all filesystem operations
allow(FileUtils).to receive(:chown_R)
allow(FileUtils).to receive(:makedirs)
allow(FileUtils).to receive(:rm_rf)
# Stub our mwdeploy user
allow(Etc).to receive(:getpwnam).with('mwdeploy').and_return(OpenStruct.new(uid: 666))
# Stub the existance of our deploy-local command
allow(@provider.class).to receive(:command)
.with(:scap)
.and_return('/usr/bin/scap')
end
...