This smells like you have a second copy of facter or some other facts
somewhere in your RUBYLIB, as the latest version no longer uses
Facter::IPAddress. Are you sure you haven't got an RPM or local copy
installed somewhere else?
Try running facter --trace as well so we can see the output. The path
What are you actually trying to do with the YAML file today Marek
whereby the links are causing such problems? This is a semi-loaded
question ... call me curious :-).
On Wed, Feb 15, 2012 at 9:55 PM, Marek Dohojda wrote:
> :: sigh :::
>
> Back to the ol' drawing board. LOL.
>
> Well I guess that
This ordering behaviour is as you state, and the numbers in the
namevar are ultimately for how they get ordered in the file ruleset as
you state - but not what order they are _inserted_. Ideally it would
be great to have insertion order and order in the firewall list to be
the same - but this doesn
> You said:
>>
>> the numbers in the namevar are ultimately for how they get
>> ordered in the file ruleset as you state - but not what order
>> they are _inserted_.
>
> Which makes me still think that the order various modules kick can affect
> the firewall rules. Thus, a stage after main is still
> Since the whole fwpre class is run before everything else, is it necessary
> to define each resource with dependencies with firewall {"002 testing":
> ...}->firewall {... as in your gist?
No its not.
> Anyway, works great for us now. Thanks much!
Good to hear - I'll get the documentation fixed
Hi Amos - its been a long long long time mate :-).
> example for (1): Our vagrant (http://vagrantup.com/) dev base boxes
> still come with Puppet 2.6.3 while the manifest depends on Puppet 2.7
> features. I can manually upgrade puppet manually (and that's what I do
> on dev), but when the time to
>> 3. More forethought and discussion on the dev list prior to making a
>> pull request/patch.
>
> That'd be really great. And I noticed some attempts lately in this
> direction, which is really good.
I've been moving more discussions onto puppet-dev in the last few
weeks, as I've been delving mo
(I don't have a direct tool - but its an interesting conceptual problem)
You would ordinarily want an intermediate system for this, either
something custom - or a CI system like Jenkins (and perhaps Travis CI
does this as well) where you would put your 'publish' logic.
The problem of course, bein
A pm2rpm tool perhaps Todd? :-).
On Mon, Apr 16, 2012 at 7:36 PM, Todd Zullinger wrote:
> Michael Stahnke wrote:
>>
>> For the next major Puppet version, code-named Telly, we have some changes
>> coming. This is the first in a series of emails around these changes and
>> may require some input f
I'm going to review this now. Its destined for master, so someone from
the release team can probably comment on the next major release
schedule for stdlib.
On Tue, Apr 17, 2012 at 7:35 PM, Geoff Davis wrote:
> That's what I'm looking for. I'll fold in that branch into my testing until
> it goes m
Don't stress, I'm sure its topical :-).
On May 8, 2012 5:49 AM, "Brian Gupta" wrote:
> My apologies, this was supposed to go to the puppet-nyc mailing list. :(
> -Brian
>
> On Tue, May 8, 2012 at 12:42 AM, Brian Gupta
> wrote:
> > Ohad did a great job covering Foreman. If you couldn't make it go
Hi all,
Just FYI. I've just renamed some of the modules that are stored as
github repos in our Github organisation
(http://github.com/puppetlabs/) to use the puppetlabs-
convention so we can get a bit of consistency. The following changes
were made:
puppet-lvm -> puppetlabs-lvm
puppet-sudo -> pup
Perhaps look at the Puppet Dashboard or Foreman schemas as a starting
point? These are both ENC's that are already working.
ken.
On Wed, May 30, 2012 at 9:13 PM, erkan yanar wrote:
>
> Moin,
>
> I am thinking of using a RDBM as a best practise.
> I am missing some info/examples how to best organ
Why don't you try using PuppetDB for stored configs instead? Its
asynchronous, uses activemq behind the scenes and supports postgres.
https://github.com/puppetlabs/puppetdb
On Thu, May 31, 2012 at 10:32 AM, Svein wrote:
> How can I set up both Storeconfig and mcollective using activemq for both?
> Turns out yes, it's the leap second, but boy was the fix I found
> easier than that:
>
> http://artipc10.vub.ac.be/wordpress/2012/07/01/leap-second-causing-ksoftirqd-and-java-to-use-lots-of-cpu-time/
>
> $ sudo date -s "`date`"
>
> Cleared it rigt up.
Huh. What a weird fix :-).
ken.
--
You re
essage' as '22 100 2323'
notice: 22 100 2323
notice: /Stage[main]/Firewall_override/Notify[msg0]/message: defined
'message' as '22 100 2323'
notice: Finished catalog run in 0.01 seconds
kbarber:tmp ken$
ken.
On Fri, Jun 17, 2011 at 2:56 AM, Ashley Gould wrote:
&
You should usually run the exec with:
exec { "asdf":
...
logoutput => on_failure,
...
}
This prints to screen any error. Anyway the clue for your case is in
the error message:
err: /Stage[main]//Node[lucas.com.br]/Wordpress::Install[ ]/Exec[tar
-vzxf /var/www/wordpress.tar.gz]/returns:
> My initial "layout" was motivated by a need to "harden" our Linux
> systems. I grouped the various hardening configs into:
>
> Kernel
> OS
> Network
> Shell
> Files
> Application
>
> I'm hoping I can create the same module structure within puppet.
Start here:
http://docs.puppetlabs.com/guides/m
Augeas is a resource - I don't see how it fits in as a module. You may
_use_ it in your modules if you like.
On Mon, Jun 20, 2011 at 12:09 PM, Bruce Bushby wrote:
> One last question:
>
> Would the list suggest implementing "augeas" where possible? and would
> "augeas" fit into the "module layout
Do you have a lot of boxes that perform the same function? You can use
the regex node syntax:
node /^db/ {
include db
}
Or device some other way of extracting function out of the machine
name. It might save you some time.
Otherwise if your machines are fairly unique using ENC to configure
that
The main reason I separate title from namevar is so when I have
references to the exec they stay consistent - but the command & args
can be updated independently. This matters more when I have references
that cross file boundaries.
ken.
On Sat, Jun 18, 2011 at 7:35 PM, vagn scott wrote:
> On 06/
Did you try the logouput => on_failure in the exec?
Can you try that and run the content with the --debug flag so we can
all see the output? (try putting the output in pastie.org if its a lot
of information).
ken.
On Mon, Jun 20, 2011 at 1:28 PM, lucas.brig...@ymail.com
wrote:
> I could not do
> Getting back to my ultra simple setup, I'm finding that I can't run
> "puppetd --test":
> [root@msukpuppet02 puppet]# puppetd --test
> err: Could not retrieve catalog from remote server: SSL_connect
> returned=1 errno=0 state=SSLv3 read server certificate B: certificate
> verify failed
> warning:
> I tried using "try the logouput => ON_FAILURE in the exec."
> But my client is a puppet CentOS. (The puppet version is outdated).
> Not recognizing this parameter.
There is a newer version of Puppet in EPEL testing. For example, RHEL/CentOS 5:
http://download.fedora.redhat.com/pub/epel/testing/
It does seem like its not being included :-) ...
What does:
echo "# foo" >> /etc/ntp.conf
puppet apply -d -e 'include ntp'
Do?
ken.
On Mon, Jun 20, 2011 at 11:46 PM, Craig White wrote:
>
> On Jun 20, 2011, at 3:37 PM, Jacob Helwig wrote:
>
>> On Mon, 20 Jun 2011 15:11:09 -0700, Craig White wr
So it would appear the parameter 'name' acts as namevar for file and
exec at least ... its not defined in the type for these explicitly -
it seems implicit. This seems historical and I haven't seen it used
before (at least I've never used it myself). Does anyone know the
history behind this?
The f
Its just using the Ruby Etc library to do the lookup according to
puppet/util/posix.rb. Which won't match NIS accounts. So I'd raise a
feature request if you want this support.
You can work around this by using something like:
$username = "bob"
ssh_authorized_key { "keyfor-${username}":
key =>
Another mechanism. You can use mcollective and this resource to
trigger mco calls inside puppet:
https://github.com/ripienaar/puppet-mcollective
The resource acts like an exec ... and can be refreshed and notified.
For example:
file {"/tmp/foo1":
# content => "foo",
content => inline_template
I concur. Using mcollective to trigger Puppet is sweet.
ken.
On Fri, Jun 24, 2011 at 3:31 PM, Daniel Maher wrote:
> On 06/24/2011 04:10 PM, Nathan Clemons wrote:
>>
>> This is what we do at Livemocha; we're actually in the process of
>> switching over to using MCollective to trigger Puppet runs.
Thanks Stefan. A bug (originally a doc bug) and subsequent discussion
at Puppetlabs is documented here:
https://projects.puppetlabs.com/issues/8096
ken.
On Thu, Jun 23, 2011 at 6:11 PM, Stefan Schulte
wrote:
> On Wed, Jun 22, 2011 at 08:37:25PM +0100, Ken Barber wrote:
>> So it wou
How big are these manifests (ie. number of files, total lines for all
files that get imported)? For example - are you pulling in a large
file of all your nodes or something? I recall there has been problems
in the past with parsing large files - its fixed now though.
The obvious suggestion is to u
Hey Bjørge,
Looks like a bug ... can you create an issue here?
https://github.com/puppetlabs/puppetlabs-firewall/issues
Can you supply the output of the command iptables-save as well in the
ticket? It looks like its having trouble parsing that.
ken.
2011/7/4 Bjørge Solli :
> Hi,
>
> I am testi
> my manifest contains number of file (flat and templete ) and for
> pulling some jars on client.
I guess I was looking to find out the specifics of how big these
manifests and templates are in total on the nodes you are having
specific performance issues with. You say 'big manifests' but this is
inished catalog run
>> in 1.22 seconds
>>
>
>> -Luke
>>
>> On Jul 4, 4:13 pm, "sanjiv.singh" wrote:
>>
>>
>>
>>
>>
>>
>>
>> > On Jul 4, 7:36 pm, Ken Barber wrote:
>>
>> > > > my man
FYI - for performance I would upgrade Puppet (there were performance
improvements in newer revisions), and use something like
Apache/Passenger as a delivery mechanism.
One question - does this happen each time you run it on the same box?
If there is no new changes - is the performance still terrib
Whats wrong with using chained resources?
class x {
file { "a":
} ->
exec { "b":
} ->
exec { "c":
} ->
exec { "d":
}
}
ken.
On Tue, Jul 5, 2011 at 4:40 AM, vagn scott wrote:
> Sometimes things just have to happen in sequence.
> It is the simplest of relations, but puppet really
What does your service {} block look like? Are you using hasstatus =>
true? For example:
service { "puppet":
ensure => running,
enable => true,
hasstatus => true,
}
ken.
On Tue, Jul 5, 2011 at 9:53 AM, Chris Phillips wrote:
> Hi,
> I've a puppet module running which ensures the puppet ser
{ hasstatus => false }
service { "b": }
}
}
ken.
On Tue, Jul 5, 2011 at 10:26 AM, vagn scott wrote:
> On 07/05/2011 04:35 AM, Ken Barber wrote:
>>
>> Whats wrong with using chained resources?
>>
>
> It doesn't scale.
>
> Try expanding th
Hi Kyle,
The source paths need to be accessible by the same user that
puppetmasterd is running as ... which is usually puppet. Some of those
files are owned by root and 700 or 600 so it can't read them:
> drwx-- 2 root root 4.0K 2011-07-05 07:39 autostart
Look at the error:
> err: /Stage[m
Do you get any errors when you run puppetd/puppet agent on the server
side? (as apposed to running it on the client).
What version of puppet are you running btw?
ken.
On Tue, Jul 5, 2011 at 9:11 AM, alan bover wrote:
> Hi, I've been doing some testings with an iptables puppet recepy on my
> own
Try:
<% if includernw == 1 %> include "/etc/dhcp/dhcpd.rnw";<% end %>
Note the '==' :-).
ken.
On Wed, Jul 6, 2011 at 2:45 PM, Jonathan Gazeley
wrote:
> I have a puppet class called dhcp which sets up the daemon and installs a
> base dhcpd.conf.
>
> I have also have subclasses like dhcp::pool1,
If its a pure ruby provider, try running something like 'puppet agent
-t --pluginsync' on the server first, so it gets synced there. Then
try your client again.
The problem is your server isn't scanning your modulepath for types
... its expecting types to be in its own libdir first. Hence why an
p
ps someone can correct me on that.
ken.
On Thu, Jul 7, 2011 at 2:31 PM, JohnW wrote:
> Thanks Ken
>
> That did the trick.
>
> I was afraid I had to reinvent a wheel here, as I was a bit on a
> deadline
>
> John
>
>
> On Jul 7, 2:50 pm, Ken Barber wrote:
Let me understand this correctly - you want to use the Puppet RAL
layer to interrogate data such as free space on a mount?
At the moment you can't use RAL data elsewhere in Puppet content ...
but you can use the RAL layer in your own code and in things like
MCollective using the ralsh plugin ...
'defined' might be what you are after ... something like:
class foo {
}
class { "foo":
}
if defined(Class["foo"]) {
notice("foo is defined")
} else {
notice("foo is not defined")
}
When you comment out the class { "foo": } then the expression results
in a negative.
ken.
On Thu, Jul 7, 201
> notice: /Stage[main]/Nagios::Nrpe/Service[nagios-nrpe-server]/ensure: ensure
> changed 'stopped' to 'running'
That looks like its trying to start it, not restart it. Do you get
this message each time you do a puppet run (even when nothing has
changed)? If so - you might need a valid 'status' com
Use the pattern attribute, documented here:
http://docs.puppetlabs.com/references/stable/type.html#service
ken.
On Fri, Jul 8, 2011 at 1:18 AM, Joehillen wrote:
> Yes, it says it every time. The actual process is called "nrpe", so that
> makes sense that it can't find it.
> Unfortunately, it do
> The reason we are doign this is to make sure that we have sufficient free
> disk space on various mountpoints to actually run the application on the
> server.
Oh - and your monitoring doesn't do this already?
> So this is something I'd like to enforce (or "ensure") as part of the
> configuratio
You can add new entries using the 'host' type:
http://docs.puppetlabs.com/references/stable/type.html#host
For example:
host { "myhostname":
ip => "1.1.1.1",
}
By itself, it won't remove existing entries (but there is a way to do
this as well using purging).
If you haven't read it already, t
So ...
> servermd5 = YAML.load(apitruthtag("metadata")).ivars["checksum"] #
> When executed from a puppet run I tells me that ivars is undefined.
What does the output of apitruthtag("metadata") show you between each
run in facter, irb and puppet? Can you output each to a file and
analyze the diff
> Among the modules iv tried:
>
> https://github.com/puppetlabs/puppetlabs-firewall
What was the problem you had with this module?
ken.
--
"Join us for PuppetConf, September 22nd and 23rd in Portland, OR:
http://bit.ly/puppetconfsig";
--
You received this message because you are subscribed t
Hi Ronen,
Making the rules persistent is a matter of running iptables-save
afterwards. If you drop this in your top scope it should work:
exec { "persist-firewall":
command => $operatingsystem ? {
"debian" => "/sbin/iptables > /etc/iptables/rules.v4",
/(RedHat|CentOS)/ => "/sbin/iptable
, Jul 11, 2011 at 10:59 AM, Ronen Narkis wrote:
> Just did,
>
> Thank you!
> Ronen
>
> On Mon, Jul 11, 2011 at 1:50 AM, Ken Barber wrote:
>>
>> Hi Ronen,
>>
>> Making the rules persistent is a matter of running iptables-save
>> afterwards. I
Have you looked at tags?
http://docs.puppetlabs.com/references/stable/metaparameter.html#tag
http://docs.puppetlabs.com/references/stable/function.html#tag
It allows you to tag parts of your content, and selectively run it
with --tags on the command line.
Classes are already tagged implicitly w
IMHO I don't think I want nested facts to work :-). The evaluation
order alone is trouble some ... ie. the inner fact wouldn't get added
until the enclosed fact is evaluated ... doesn't sound like something
you would want in ordinary circumstances :-). Remember sometimes fact
information is gather
> Because about one in five of them is not actually *finished*
So currently the incomplete functions are:
date
is_float
is_integer
is_numberic
is_valid_domain_name
is_valid_ip_address
is_valid_mac_address
is_valid_netmask
rand
squeeze
If anyone wants to help complete these - I'm happy to take pa
Take a look at Jordan Sissel's FPM software if you want a cheap way of
making a deb.
Ken.
On Aug 1, 2011 7:10 PM, "Daniel Pittman" wrote:
> Yup. This is totally the right way to do that.
>
> If you don't want to, I would suggest that you compile once, tar up
> the content, and just untar it on th
Try:
notice($t['given'])
Or:
notice("${t['given']}")
Surrounding the var in ${} within quotes helps the
parser/interpolation find the edges of your var declaration.
ken.
On Thu, Aug 4, 2011 at 2:09 AM, Douglas Garstang
wrote:
> Well, this is weird...
> If I have this yaml file:
> ---
> custo
Hey Oliver :-).
> * using class introspection, generate class interface YAML that is
> read and used by our ENC instead of using a separate configuration
> specification, store this with the versioned Puppet classes.
I guess resource_type is your friend in this case. Info should be
available in
You need to add a 'subscribe' or 'notify' parameter to your server or
file resource (respectively). For example:
http://docs.puppetlabs.com/references/stable/metaparameter.html#notify
There is an example in the puppetlabs-ntp module:
https://github.com/puppetlabs/puppetlabs-ntp/blob/master/manif
If you want the sysv equivalent to:
service puppetmaster status
Its:
/etc/init.d/puppetmaster status
You used the agent's one which is not going to be the same service:
> [root@puppet puppet]# /etc/init.d/puppet status
ken.
On Wed, Aug 10, 2011 at 1:52 PM, Sans wrote:
> Dear all,
>
> Can an
Hi Scott,
I think its fair enough thing to expect it to do something like this ...
Incidentally, there is already a feature request for this:
http://projects.puppetlabs.com/issues/3789
Do you want to watch/vote for it ...? Or you can always provide a
patch ... we won't complain :-).
ken.
On W
Hi Steve,
Did you add pluginsync = true on the client? Is it in the [agent] or
[main] section?
What happens when you just do:
puppet agent -t --pluginsync
(obviously inserting any other switches you normally require)
ken.
On Wed, Aug 10, 2011 at 8:22 PM, Steve wrote:
> I've tried to sync cus
t; Thank you
>
> On Aug 10, 12:25 pm, Ken Barber wrote:
>> Hi Steve,
>>
>> Did you add pluginsync = true on the client? Is it in the [agent] or
>> [main] section?
>>
>> What happens when you just do:
>>
>> puppet agent -t --plug
If you look at this example:
>> I have a Ruby class in "nagios" module - it's located in nagios/
>> manifests/ssa_nagios_checks.rb
>> and looks like this
>>
>> hostclass :ssa_nagios_checks do
>> ...
>> end
>>
>> In nagios/manifests/init.pp I have
>>
>> class nagios::server {
>> ...
>> include s
e $type"
> end
>
> nagios_service( "ssa_#{x['doc']['_id']}",
> :target => "${nagios_confdir}/app_alerts/ssa/
> services/${name}.cfg",
> :host_name => x['doc']['hosts'],
&g
Is your config facts_terminus=yaml for a reason? Usually its 'facter'.
ken.
On Thu, Aug 11, 2011 at 10:59 PM, John T. Guthrie wrote:
> Hello all,
>
> I am running puppet 2.7.1 with mongrel and an apache proxy. I have been
> seeing an issue where it appears that the puppetmaster is completely
>
Hi Bjørge,
The clue is in looking at the code for the 'fqdn' fact:
host = Facter.value(:hostname)
domain = Facter.value(:domain)
if host and domain
[host, domain].join(".")
And drilling into the 'domain' fact:
if name = Facter::Util::Resolution.exe
> The first two could be solved as one using 'hostname --fqdn', meybe even
> the third one.
Perhaps the methodology chosen was more compatible with multiple UNIX
operating systems. For example hostname --fqdn doesn't seem to work on
Mac OS X. A lot of facter code may appear quirky on the surface f
The user resource won't clean up home directories directories for you.
In fact there was a feature request to rename the attribute for this
very reason:
http://projects.puppetlabs.com/issues/7002
It was originally pegged in the original managehome feature request
many years back but never made it
'notice' is a function. It runs on the server. Try using the 'notify'
resource instead.
ken.
On Sat, Sep 10, 2011 at 12:24 AM, Douglas Garstang
wrote:
> Maybe it's because it's Friday, but I'm suddenly not sure if I am
> going crazy or not...
>
> I have a server running 2.7.3, and clients runnin
>> 'notice' is a function. It runs on the server. Try using the 'notify'
>> resource instead.
>
> but surely those two variables would be based on the facts and not some
> master state?
>
> notify would run on the node yes, but the resolution of variable to string
> would still happen on the master
What does:
puppet -e 'notice("$puppetversion")'
Show on the _client_? I know its probably obvious but you never know.
ken.
On Sat, Sep 10, 2011 at 12:39 AM, Douglas Garstang
wrote:
> On Fri, Sep 9, 2011 at 4:36 PM, Ken Barber wrote:
>>>> 'notice&#x
>> puppet -e 'notice("$puppetversion")'
>>
>> Show on the _client_? I know its probably obvious but you never know.
>>
>
> Actually, it's not obvious. I didn't know you could do that. Never
> seen it documented anywhere...
puppet help apply
(I really should use puppet apply these days as that way
> I assume you mean't to use notify(), not notify{} ?
No - he meant what he said. Its a resource for client notifications.
Give it a try can't hurt.
Can you send your puppet.conf from both boxes (via pastie or
somesuch)? Maybe there is something there. I'm tempted to suggest
running the pupp
Out of morbid curiosity - how are you running the puppet client on
that client box? You aren't running an agent are you and using kick or
some such?
On Sat, Sep 10, 2011 at 12:56 AM, Ken Barber wrote:
>> I assume you mean't to use notify(), not notify{} ?
>
> No - he m
>>> I'm launching puppet with 'service puppet restart'.
Do you get any different results if you run it like:
puppet agent -t
Instead of using that service?
Are you absolutely certain there isn't a stray ruby process running
your old 0.25 puppet agent?
> Well, I can't seem to work out what's go
> On Client:
> service puppet stop
> yum clean all
> rpm --erase puppet
> rpm --erase facter
> rm -fR /var/lib/puppet
> yum upgrade puppet
>
> On server:
> puppetca --clean hproxy11.h.foo.com
>
> On Client:
> service puppet start
Where are the RPM's from that you are using for both 0.25.5 and 2.7.
>> Where are the RPM's from that you are using for both 0.25.5 and 2.7.3?
>> I wonder if this can be replicated elsewhere ... what OS distribution
>> and version are you using?
If you can let me know this info I can probably try to replicate.
One thing was troubling me ... and its probably a long
> The RPM for the puppet client came from the Puppet Labs website. I
> downloaded the source for 2.7.3, and had to make a few changes to the
> SPEC file
(I keep asking this but ...) What OS are you running and version?
Where did you get the RPM's for 0.25?
Also ... why don't you try these RPM's f
> I don't know what pbcopy is locate can't find it. Never heard of
> it,
Yeah never mind - its a convenient Mac OS X tool for copying something
into the clipboard. Not critical.
> but, after removing the rpm and wiping /var/lib/puppet and running
> 'locate puppet', I get:
(you really should
> What version of facter are you running btw?
You already answered that - never-mind :-).
ken.
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group,
I'm guessing you get nothing when you try:
facter domain
?
What version of facter are you running btw?
Can you show the results of the following commands:
hostname
dnsdomainname
cat /etc/resolv.conf
Cheers.
ken.
On Tue, Sep 13, 2011 at 1:33 PM, Sans wrote:
> I know it's not directly relate
Incidentally, my puppet master runs "facter-1.6.0-2.el5" and here
> "facter fqdn" returns the correct value.
> Cheers!!
>
>
> On Sep 13, 2:32 pm, Ken Barber wrote:
>> I'm guessing you get nothing when you try:
>>
>> facter domain
>>
I could be wrong about the cause actually ... I think its still a bug
though :-).
Let me take a closer look at the code and see if I can work it out.
ken.
On Tue, Sep 13, 2011 at 3:29 PM, Ken Barber wrote:
> Yep - that looks like a bug. The change was here:
>
> https://github.com/p
in your full domain - it won't fall
back to checking dnsdomainnname or resolv.conf.
ken.
On Tue, Sep 13, 2011 at 3:36 PM, Ken Barber wrote:
> I could be wrong about the cause actually ... I think its still a bug
> though :-).
>
> Let me take a closer look at the code and see if
Since this is urgent and we are in RC, I've raised a bug for you Sans:
http://projects.puppetlabs.com/issues/9457
ken.
On Tue, Sep 13, 2011 at 3:43 PM, Ken Barber wrote:
> Yeah okay I was close though :-).
>
> if name = Facter::Util::Resolution.exec('hostname')
>
Yeah this doesn't seem to be old versions of Puppet.
So from my other email ... can you show us the code where you are
doing your comparison for $puppetversion? I have a feeling I might
know what it is ... although I'm probably wrong ...
In fact - grep for 'puppetversion' in all of your puppet co
So that is the only case where you are using the variable throughout
your entire code? I mean - ALL your code ... not just the one line you
are printing to the screen ...
ken.
On Tue, Sep 13, 2011 at 8:30 PM, Douglas Garstang
wrote:
> On Tue, Sep 13, 2011 at 11:01 AM, Ken Barber wrote:
>
n.
On Tue, Sep 13, 2011 at 8:34 PM, Ken Barber wrote:
> So that is the only case where you are using the variable throughout
> your entire code? I mean - ALL your code ... not just the one line you
> are printing to the screen ...
>
> ken.
>
> On Tue, Sep 13, 2011 at 8:30 PM, D
Even using:
notify{"${::puppetversion} on ${::fqdn}":}
Instead would be of interest ...
ken.
On Tue, Sep 13, 2011 at 8:46 PM, Ken Barber wrote:
> The reason why I say this - is because I can replicate this problem myself
> with:
>
> node default {
> $puppetver
otify statement ...
ken.
On Tue, Sep 13, 2011 at 9:01 PM, Douglas Garstang
wrote:
> On Tue, Sep 13, 2011 at 12:46 PM, Ken Barber wrote:
>> The reason why I say this - is because I can replicate this problem myself
>> with:
>>
>> node default {
>> $puppetversio
.rb" file that you were talking about?
>
> Cheers,
> San
>
> On Sep 13, 5:38 pm, Ken Barber wrote:
>> Since this is urgent and we are in RC, I've raised a bug for you Sans:
>>
>> http://projects.puppetlabs.com/issues/9457
>>
>> ken.
>
> --
Good to hear. That fix should be in the next rc.
ken.
On Tue, Sep 13, 2011 at 10:46 PM, Sans wrote:
> Done!!
> Work like a charm. thanks.
>
> -Santanu
>
>
> On Sep 13, 9:25 pm, Ken Barber wrote:
>> Yeah - try this copy instead if you can:
>>
>> https://
Sep 13, 2011 at 3:21 PM, Douglas Garstang
>> wrote:
>>> On Tue, Sep 13, 2011 at 1:06 PM, Ken Barber wrote:
>>>> Hmm ... well can you try using ${::puppetversion} ...?
>>>
>>> Adding this:
>>>
>>> notify{"xxx = ${::puppetversion}
capture from the client of facts to
rule out the client ... but I'm not 100% on the technique as its SSL.
I'll do some research - but maybe someone else can chime in.
ken.
On Wed, Sep 14, 2011 at 12:33 AM, Douglas Garstang
wrote:
> On Tue, Sep 13, 2011 at 4:27 PM, Ken Barber wrote:
> I think this may be totally related. We're using mongrel... AND,
> running the puppet master in standalone mode as you suggested yields
> this now on the client:
>
> Sep 13 16:49:06 hproxy11 puppet-agent[27311]:
> (/Stage[main]/Puppet::Setup/Notify[xxx = 2.7.3 ...]/message) defined
> 'message' as
> Thanks for your help. Unfortunately, I can't tell you much about the
> setup. I've only ever done this with passenger as well, and the person
> who did set it up has
> left the company.
I'm not really sure if you should raise a new ticket on this one - or
add an addendum to this:
http://project
Hi Alex,
What happens when you run 'ip addr list' instead?
ken.
On Tue, Sep 20, 2011 at 6:20 PM, Alex L. Demidov
wrote:
> I have Gentoo host where `ifconfig -a` prints long interface names
> truncated to 9 chars (there is closed bug report [1]).
>
> Unfortunately, `facter` uses `ifconfig -a` ou
20, 2011 at 6:31 PM, Alex L. Demidov
wrote:
> On Tue, Sep 20, 2011 at 06:24:40PM +0100, Ken Barber wrote:
>> Hi Alex,
>>
>> What happens when you run 'ip addr list' instead?
>
> It shows interface names properly and not truncated.
>
>>
>> ken.
>&
1 - 100 of 558 matches
Mail list logo
