On Tue, Jul 21, 2015 at 1:27 AM, Fabrice Bacchella
fabrice.bacche...@orange.fr wrote:
For this (and a few other reasons), the breaking change of removing the
-p switch in Facter 3.0.0 is being reverted in 3.0.2, which will be
released very soon.
When do you expect it to be released ?
There have been a number of regression fixes including fixes to command
execution, inclusion of the -p option, porting the xendomains fact, OS fact
fixes, and the fqdn fact reporting incorrectly on some systems.
If there's an issue you're running into that you don't see being addressed in
On 07/21/2015 04:14 PM, jcbollinger wrote:
2)
I read in the changelog that the older clients should work with
the backward compatibility, but is it worth even trying? Or should
I just push out the updated puppet client wherever I can before
adding the server into my new
Systemd services do not have access the normal environment variables that are
present in yor shell.
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
On Tue, Jul 21, 2015 at 9:50 AM, Fabrice Bacchella
fabrice.bacche...@orange.fr wrote:
On a linux, if I enumerate interfaces using standards tools I get :
~# ip link list
1: lo: LOOPBACK,UP,LOWER_UP mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd
I agree that Facter could do a better job merging the secondary interface
into the primary one here. It shares the networking fact code with a few
other platforms (mainly OSX and the BSDs) and it currently doesn't do any
specific logic for Linux to merge bonded interfaces together.
I
On a linux, if I enumerate interfaces using standards tools I get :
~# ip link list
1: lo: LOOPBACK,UP,LOWER_UP mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP mtu 1500 qdisc mq master bond0
state UP
I am probably confused when it comes RHEL7. But systemd comes into play
here only when I restart my puppet master with # systemctl restart
pe-httpd right? I thought I had confirmed that fact pod_prefix is getting
set on the node when I execute:
[root@vii-osc4-mgmt-001 ~]# facter pod_prefix
I thought I was doing those things. I set up a FACTER_ env var on one of
my nodes to test like so ...
[root@vii-osc4-mgmt-001 ~]# cat /etc/profile.d/POD_prefix.sh export
FACTER_pod_prefix=vii-osc4
I rebooted the node, vii-osc4-mgmt-001, and checked that env. var is being
set (maybe it's not)
Le 21 juil. 2015 à 20:33, Peter Huene peter.hu...@puppetlabs.com a écrit :
Thus there would be no eth0:1 in the interfaces list; it would just show up
as the first element in the secondary array. e.g. eth0:1 becomes
eth0.secondary.0, eth0:2 becomes eth0.secondary.1, etc.
Thoughts on
Gotcha! Thanks
On Tue, Jul 21, 2015 at 11:33 AM, Thomas Müller tho...@chaschperli.ch
wrote:
Are you saying that my test is no good because it is using the environment
from the shell?
yes systemd does not send provide /etc/profile.d environment to its
services.
just don't use FACTER_
If you have access to Facter 2.x on this same system, I'd be curious to see
if it would output the secondary interface / address information. If not, I
think we could consider Facter's 3.x behavior to be a bug and fix it by
moving the secondary interfaces to where they belong.
On
Right. But with a caveat: the random number is not really random. In
fact the random generator is seeded with the node fqdn. Since this is a
constant, the generated random number won't change from run to run for a
given node.
Or if you want to manage cron resource directly:
cron {
On Tue, Jul 21, 2015 at 11:47 AM, Fabrice Bacchella
fabrice.bacche...@orange.fr wrote:
Le 21 juil. 2015 à 20:33, Peter Huene peter.hu...@puppetlabs.com a
écrit :
Thus there would be no eth0:1 in the interfaces list; it would just
show up as the first element in the secondary array.
ah you mentioned it already ... fact.d
On Tue, Jul 21, 2015 at 11:03 AM, Russell Cecala red.cricket.b...@gmail.com
wrote:
I am probably confused when it comes RHEL7. But systemd comes into play
here only when I restart my puppet master with # systemctl restart
pe-httpd right? I thought I
On Tue, Jul 21, 2015 at 11:06 AM, Fabrice Bacchella
fabrice.bacche...@orange.fr wrote:
I agree that Facter could do a better job merging the secondary
interface into the primary one here. It shares the networking fact code
with a few other platforms (mainly OSX and the BSDs) and it currently
Are you saying that my test is no good because it is using the environment
from the shell?
yes systemd does not send provide /etc/profile.d environment to its
services.
just don't use FACTER_ env variables. Use facter-dot-d or a custom fact in
a puppet module.
- Thomas
--
You
I am using the VLAN resource type to manage VLANs on a Cisco 4500.
On every run puppet reports that it created the VLAN (even though it
exists).
Grabbing data from hiera:
network::layer3::vlans:
site:
name: 703
description: 'Site'
management:
name: 712
description:
How to get all puppet agent/master logs to rsyslog? intend to ship all
puppet agent/master messages to central loghost and parse with logstash and
ship to elasticsearch. Currently puppet logs some messages to
/var/log/messages via rsyslog while puppet daemons write directly to
For the puppet agent, you can always try tdis rsyslog syntax:
if $programname == 'puppet-agent' then /var/log/puppet.log
if $programname == 'puppet-agent' then ~
Same for the master process.
On Tue, Jul 21, 2015 at 11:29 PM, eric odell cirell...@gmail.com wrote:
How to get all puppet
On Tuesday, July 21, 2015 at 9:14:04 AM UTC-5, jcbollinger wrote:
Questions!
1a)
I am debating as to whether I need to build multiple puppet masters or
not. Is the following documentation still valid for the 4.x series of
puppet?
Doh! The /etc/facter/facts.d/pod_prefix.txt file goes on the nodes NOT the
puppet master. And it works!
On Tue, Jul 21, 2015 at 1:10 PM, Russell Cecala red.cricket.b...@gmail.com
wrote:
OK I tried looking into the facter-dot-d way of setting a custom fact, but
it does not seem suitable for
On Tue, Jul 21, 2015 at 12:26 PM, Fabrice Bacchella
fabrice.bacche...@orange.fr wrote:
If you have access to Facter 2.x on this same system, I'd be curious to
see if it would output the secondary interface / address information. If
not, I think we could consider Facter's 3.x behavior to be
OK I tried looking into the facter-dot-d way of setting a custom fact, but
it does not seem suitable for what I am trying to do.
I want some nodes to report the fact pod_prefix as 'vii-osc4' and I will
want other nodes to report pod_prefix as 'ix-xyz' so that my hiera.yaml
hierachy of ...
On Tuesday, July 21, 2015 at 9:44:45 AM UTC-5, Martijn wrote:
Op dinsdag 21 juli 2015 01:28:09 UTC+2 schreef Stack Kororā:
If I am willing to take the time to migrate to puppet-forge modules
anyway (plus hopefully fix any bad code I can't replace), wouldn't this be
a good time to switch
On Tuesday, July 21, 2015 at 11:33:29 AM UTC-5, Felix.Frank wrote:
On 07/21/2015 04:14 PM, jcbollinger wrote:
2)
I read in the changelog that the older clients should work with the
backward compatibility, but is it worth even trying? Or should I just push
out the updated puppet
One last update on my progress.
I have 4 installed on Scientific Linux with the only client being itself at
the moment. The current plan is to either find replacement modules or move
over as many modules as I can. If I can get the bulk of the modules and a
few important ones, then I will be
Hi all,
Is there currently a preferred package provider for Mac? I'm trying to
determine whether pkgdmg or gem is the better option for a set of packages
I'm installing.
Thanks,
Mike Turner
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To
you need to adjust the hiera hierarchy. To adjust it you need a top scope
variable (e.g. pod_prefix)
(https://docs.puppetlabs.com/puppet/latest/reference/lang_scope.html#top-scope
) to be used in hiera.yaml configuration file.
you could also make a custom fact
For this (and a few other reasons), the breaking change of removing the -p
switch in Facter 3.0.0 is being reverted in 3.0.2, which will be released
very soon.
When do you expect it to be released ?
I've found many bugs with facter 3 (and I've been using it only for a few
days),
Hi All,
I am getting following error while running puppet agent
puppet agent -tv
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Loading facts
Unable to add resolve nil for fact ks_memsql_nodeid: undefined method
`num_rows' for Error: 2003:String
Unable to add resolve nil for fact
On Monday, July 20, 2015 at 9:45:12 PM UTC-5, Fabien Delpierre wrote:
Peter, Felix,
Thank you for your answers!
I guess my improperly worded question was about figuring out whether
Puppet has a concept similar to what Chef calls a wrapper cookbook.
I know nothing more about wrapper
I want to both install puppet 4 and some mcollective plugins on a CentOS 6
As the the mcollective plugins are not in the PC1 repository, I need the
products repository. But then :
yum install
https://yum.puppetlabs.com/el/6/products/x86_64/puppetlabs-release-6-11.noarch.rpm
Figured it out! The issue was not with my Puppet code, but simply with my
Vagrantfile.
I had the following:
config.vm.provision :puppet do |puppet|
puppet.manifests_path = manifests
puppet.module_path = modules
puppet.working_directory = /vagrant
puppet.facter = {
}
end
I edited the
On Tuesday, July 21, 2015 at 12:25:25 AM UTC-5, Thomas Müller wrote:
Saz/ssh seems to use the standard sshkey type to export the host keys.
Code: https://github.com/saz/puppet-ssh/blob/master/manifests/hostkeys.pp
Seems you can export the non-puppet managed keys on a puppet managed node
On Monday, July 20, 2015 at 6:28:09 PM UTC-5, Stack Kororā wrote:
Greetings,
First, a short intro. :-)
Last year I invested quite a bit of time building out a puppet
infrastructure. I have +150 Scientific Linux 6 and RHEL 6 servers running
puppet 3.7.1. It has been working really well
On Tuesday, July 21, 2015 at 9:46:56 AM UTC-4, jcbollinger wrote:
I know nothing more about wrapper cookbooks than I can deduce from their
name and your code, but they sound like they may have some similarity with
the Puppet pattern called Roles Profiles, conceived by Craig Dunn. You
Op dinsdag 21 juli 2015 01:28:09 UTC+2 schreef Stack Kororā:
If I am willing to take the time to migrate to puppet-forge modules anyway
(plus hopefully fix any bad code I can't replace), wouldn't this be a good
time to switch puppet versions too?
I would definitely think so, but one thing
38 matches
Mail list logo