The ploperations Forge user (and associated puppetlabs-operations GitHub
account) have some very useful modules that are kept up-to-date on
GitHub, but haven't been seeing releases to Forge (in more than a year,
actually).
Is this a matter of lack of time or interest by the PL ops team on
pus
I made the really minor tweaks required to make RPMs that build on and
are compatible with Amazon's RedHat-derived Linux distribution (for
EC2). Understandably PL might not have the customer demand for azn-linux
builds, but given how little the patches touch, I hope they can get
integrated so t
Is there by chance a torrent available of these files? For whatever
reason, downloads.puppetlabs.com is being especially slow with the ovf
file for me.
Jeff
On 04/17/2014 07:45 PM, Ben Ford wrote:
Hi,
We just rolled out the new and improved Learning VM — now with fun quests to
learn Puppet
While implementing restart is good, if for some reason you were unable
to do that you need only set *hasrestart => false* for the service
resource, per
http://docs.puppetlabs.com/references/latest/type.html#service . That
will cause Puppet to issue a service stop then service start, instead of
I don't have a Maverick box to test on, but the PRs in question are
applied to the 3.4.0 and onward tags, so I would bet that current OS X
packages would do the right thing and work with the Ruby 2.0 on Maverick.
Jeff
On 01/14/2014 01:15 PM, Khoury wrote:
Does anyone know what the status is on
On 01/09/2014 10:12 AM, Pablo Fernandez wrote:
I understand your point. I guess the SSL layer will render the request
as illegitimate, but even if it doesn't, it may be playing with fire :)
No, actually it doesn't verify certname against fqdn or any such, so
technically you could bake in a s
Depending on your comfort level you could use PowerShell to do it, see
http://powershell.com/cs/blogs/tobias/archive/2009/02/02/xml-part-2-write-add-and-change-xml-data.aspx
You could combine that XML read/manipulate/write technique with the
Powershell exec provider from
https://forge.puppetla
https://tickets.puppetlabs.com/browse/MCO-144
It's fixed in package agent 4.2.1 (not yet pushed to yum repo, alas)
Jeff
On 01/07/2014 01:18 PM, Nikolay Georgieff wrote:
Same issue here.
[root@puppet ucla_gateway]# mco package puppet update -F
foreman_env=testing -v
Discovering hosts using t
+1
I would not be at odds, it seems like that is a sensible default
behavior, while still preserving the ability to shoot one's self in the
foot if needs be.
Jeff
On 11/20/2013 01:40 PM, Rob Reynolds wrote:
With the changes for http://projects.puppetlabs.com/issues/18931
(Don't set mode on
ore descriptive name to use, that's just what I've
seen most often. If you like ::general you should roll with it,
Jeff
Thank you!
On Thursday, January 2, 2014 6:47:02 AM UTC-8, Jeff Bachtel wrote:
On 01/01/2014 08:38 PM, mjus...@gmail.com wrote:
> Hi all,
>
On 01/01/2014 08:38 PM, mjuszc...@gmail.com wrote:
Hi all,
I have a role/profile setup that's working quite well. However, I'm
finding that there are *super* simple things that don't really require
the setup of their own module, such as installing "nano" or setting up
a yumrepo {}. Do I re
On 12/18/2013 06:00 AM, asim patel wrote:
Hi All,
even i am facing the same prob, when i am trying to start the puppet
agent getting an error illegal instruction (core dump)
please let me know the resolution
On Tuesday, 28 May 2013 13:04:14 UTC+5:30, Aqeel Nazeer wrote:
Hi,
I just u
13 20:28, "Jeff Bachtel" <mailto:jbach...@bericotechnologies.com>> wrote:
(adding puppet-users@ back in case others have better information
on handling filebuckets)
On 12/20/2013 12:53 AM, Harshita Sinha wrote:
Dear Jeff,
Thanks for your guidance.
Bas
(adding puppet-users@ back in case others have better information on
handling filebuckets)
On 12/20/2013 12:53 AM, Harshita Sinha wrote:
Dear Jeff,
Thanks for your guidance.
Basically I want to achieve sending files from Windows to unix.
I missed "push the file from windows to unix" in your
Your default node definition does not define any resources. filebucket
does not itself define a file, it defines a location where files are
backed up to.
If you're just beginning with Puppet, then the method for using a
filebucket to distribute file content is not something you really want
to
[main] pluginsync = true
In /etc/puppet/puppet.conf on your agents is really the way to go (from
http://docs.puppetlabs.com/guides/plugins_in_modules.html#enabling-pluginsync).
Otherwise you're looking at something like MCollective to remotely force
a plugin sync when you make changes.
But
http://askubuntu.com/questions/150267/how-to-add-a-local-repository-in-kickseed-based-install
Has an example that doesn't seem to use late_command.
https://help.ubuntu.com/10.04/installation-guide/example-preseed.txt has
it integrated into the preseed file (grep for local0 in the file),
Jeff
stem => %w{Debian SLES OpenSUSE CentOS}
outside the setcode block?
Thanks,
Felix
On 12/18/2013 04:34 PM, Jeff Bachtel wrote:
Facts are autoloaded from all modules and distributed to all agents,
because that step occurs before the DSL is parsed for manifests (as it
should be, because the D
Facts are autoloaded from all modules and distributed to all agents,
because that step occurs before the DSL is parsed for manifests (as it
should be, because the DSL can be (is) impacted by facts).
It is up to the custom fact to case itself out of execution for certain
operating systems. For
The command should be in single quotes, try
exec { "ls":
command => 'test $(grep -c aa /tmp/aao) -gt 0',
returns => 1,
path => "/usr/bin:/usr/sbin:/bin",
}
Note also that you were missing a comma after the command property, that
must be present.
Jeff
On 12/18/2013 04:40 AM, shlo.af...@gmai
Do you have use_cached_catalog set on puppet agents or puppetmaster? Are
you using storeconfigs / puppetdb?
Offhand, I'd guess that when you modify mtime of site.pp by saving, that
forces puppet to invalidate the cached catalog and temporarily resolve
the problem.
If necessary, are you able
First, to clarify - do you have pkgng installed on your FreeBSD box?
That is, should you be using /usr/sbin/pkg instead of
/usr/local/bin/pkg_info if you are doing local package management?
If so, the current freebsd provider is not likely to work well for you,
as it's pretty tied into the old
Have you tried a yaml to csv converter like
https://github.com/tokland/yaml2csv ? Note that it seems to require the
csv module syntax from ruby 1.9.3.
Jeff
On 12/16/2013 03:29 PM, Stuart Cracraft wrote:
Is the back-end piece of Puppet Dashboard which implements the
CSV-capability
of the YAML
There is a problem with variable hooks in puppet.conf that is fixed in
the latest 3.4 release candidates (or so sayeth the bugreps).
Can you try explicitly specifying your directories instead of using
$confdir in that puppet.conf, at least to test?
Jeff
On 12/17/2013 09:18 AM, Chantal Rosmul
yum-config-manager --enable rhel-6-server-optional-rpms
Has worked on some RHEL machines, for me. It depends on how your system
is subscribed whether that will work or whether you'll need to use
another command.
Jeff
On 12/13/2013 04:50 PM, Stuart Cracraft wrote:
Great!
What is the .repo f
Felix, could that above behavior be linked to
http://projects.puppetlabs.com/issues/23349 ? That is, could there be a
hook/ordering issue?
Sprang to mind, anyway. Probably cargo cult.
Jeff
On Thu, Dec 12, 2013 at 12:36 PM, John Pyeatt wrote:
> OK,
> Here is something I found with this problem
I install puppet agent on provision (kickstart, VM template, whatever)
and do a puppet agent -t -w 1 (we don't autosign) at the end of the
provision script. MCollective in our environment is handled as part of
the puppet manifest, so after the agent run the host should be
registered in mco.
J
Are you seeing the not-working rules generate iptables rules (verify
with iptables-save after applying manifest)?
Also, it's been a while since I dug into the ruby that generates the
rules, you should probably use the before operator -> between rules to
enforce ordering, like
firewall { '0
In the case of the module you're using (
https://github.com/luxflux/puppet-openvpn it appears) and you seem to
have already understood this, the client resource has a hard dependency
on files that should only exist on the server (due to client cert
generation process).
All is really not lost,
On 12/11/2013 04:11 PM, F. Y. wrote:
Hi folks,
I'm running into an issue where a home grown fuse module is mounting a
file structure, and it works the first time. But on the second run,
Puppet complains:
Warning: /Stage[main]/Test/File[/home/myhome/dir1/dir2/dir3/dir4]:
Could not stat; per
ependencies:
facter : Depends: dmidecode but it is not installable
E: Broken packages
I added '--ignore-missing' but the result was the same.
On Wednesday, December 11, 2013 7:47:02 AM UTC-8, Jeff Bachtel wrote:
demidecode (which is a hard dependency for facter) is not
avail
demidecode (which is a hard dependency for facter) is not available for
armel in Debian's repository. I have confirmed that facter will not fail
outright when dmidecode is missing.
The following facter variables will not be created, however:
bios_release_date
bios_vendor
bios_version
boardmanuf
Try
case $hostname {
'puppetagent': {include sendmail}
}
BUT, that's really the wrong pattern to be following, you should instead
create a node definition for "puppetagent" that includes the sendmail class.
node puppetagent.yourdomain {
include sendmail
}
Jeff
On 12/11/2013 10:00 AM, sh
What messages are you seeing in /var/log/puppetdb/puppetdb-daemon.log
and puppetdb.log on 129.128.100.11? Did you previously have puppetdb
running on your puppet master?
Because you got the 403, puppetmaster is at least able to connect to and
receive a response from 129.128.100.11:8081 via HTT
Note that even with local file storage, the java container that puppetdb
runs in can take a longer-than-expected time to spawn listener threads.
No issue if you're running your development puppetdb somewhat persistent
(outside of your Vagrant VM), but if you put puppetdb on the Vagrant
machine
When git feeds the file to puppet parser on your git repo server, it is
doing so via a temporary file (evidenced by remote: Error: Could not parse
for environment production: No file(s) found for import of 'fix.pp' at
/tmp/tmp.hujjYZKPpC/manifests/nodes.pp:9 ). Because it is a temporary file
it doe
What is the output of "puppet agent -t -d" ? The -d debug flag should give
you much more detailed information about what child process, exactly, is
blocking.
Jeff
On Sat, Dec 7, 2013 at 2:03 PM, Arvind P R wrote:
> My puppet master server is working fine and i now installed razor module
> on t
If this is accurate, I can only warn that I've run into horribly annoying,
manifest-breaking issues trying to run 2.6/2.7 clients with 3.3 servers. It
may not be official advice, but if you can make puppet client versions
track your puppet server closer, you will have much less headache.
Jeff
On
When creating a provider that uses a command not in PATH, what is the
best-practice pattern for case'ing out different potential locations? As an
example, the puppetlabs rabbitmq pupmod has a rabbitmqplugins provider that
falls down on CentOS using the rabbitmq upstream package due to
rabbitmq-plug
I don't have Nagios XI (still using Icinga) but could you copy out the code
from */usr/local/nagiosxi/scripts/reconfigure_nagios.sh* needed to import
files (but not the removal code) and put it in an exec resource (directly
or via a new customer shell script) that subscribes to your target
director
Three thoughts. The first would be to have node definitions for monitor on
both dev and test environments, doing a minimal amount of work to generate
nagios host definitions in a disjoint directory that you include in your
nagios config.
So:
/var/lib/{puppet,puppet-dev,puppet-test}
/etc/{puppet,pu
The gem package provider pretty extensively reads from commands[] and does
its own execute calls instead of using the command functions provided. This
makes it impossible to create a child provider that points to a different
gem command.
Would this be considered a defect worthy of a bugrep, or is
Is there a way that I'm not seeing to make scheduled_task create a task
that runs on Windows startup?
Unrelated (but too lazy to send 2 emails) does anyone have an example of
configuring the Windows time service from Puppet?
This is all on 3.3.1,
Jeff
--
You received this message because you a
43 matches
Mail list logo
