Hi all,
When I start puppet, I do it like:
puppet9343 1 0 11:24 ?00:00:00 /usr/bin/ruby
/usr/sbin/puppetmasterd --manifest=/etc/puppet/manifests/site.pp
--servertype=mongrel --logdest=/var/log/puppet/puppetmaster.log
--servertype=mongrel --masterport=18140
--pidfile=/var/run
On Wed, 11 Feb 2009 13:40:56 -0800
Paul Lathrop wrote:
>
> Arnau,
Hi Paul,
> Please don't be offended by my response, as you requested we not be
> offended by your post.
I won't, don't worry. I hope you or other developers are not angry
with me... I'm just explaining my experience with this
Hi all,
I'm facing many problems for last days with my puppet server/client.
Those who read mail list may know that.
I'm not sure if it's just my and my poor knowledge on puppet (I must
say in my defense that I have read many wiki docs and I have bough the
book, ok, it's not enough, if I'm a ba
Hi all,
puppet server is not able to find signed certificates:
]# puppetca --clean td011.pic.es
Could not find client certificate or request for td011.pic.es
But the cert is installed:
# locate td011.pic.es|grep pem
/var/lib/puppet/ssl/ca/signed/td011.pic.es.pem
And, If I go to the node, I rein
I've cahnged to the first apache conf exmaple and seems that now it's
balancing:
[r...@gridinstall conf.d]# netstat -puta|grep ruby|grep -c 18143
12
[r...@gridinstall conf.d]# netstat -puta|grep ruby|grep -c 18142
10
[r...@gridinstall conf.d]# netstat -puta|grep ruby|grep -c 18141
21
[r...@grid
I forgot to add a netstat:
[r...@gridinstall ~]# netstat -puta|grep ruby
tcp0 0 localhost.localdomain:18140 *:*
LISTEN 2848/ruby
tcp0 0 localhost.localdomain:18141 *:*
LISTEN 2866/ruby
tcp
Hi all,
I've followed http://reductivelabs.com/trac/puppet/wiki/UsingMongrel
for configuring my puppet with mongrel.
Al seems to work fine, except that, after a reinstall of 40 nodes atone
time, I got many kind of errors like:
On Mon, 9 Feb 2009 12:58:36 -0800 (PST)
Larry Ludwig wrote:
> On Feb 9, 1:31 pm, Paul Lathrop wrote:
> > Are you using stored configs?
nop.
> Also what version of the puppetmaster?
# rpm -qa|grep puppet
puppet-0.24.7-4.el5
puppet-server-0.24.7-4.el5
sorry, I tought I already set it in OP.
Hi all,
following http://reductivelabs.com/trac/puppet/wiki/UsingMongrel it
says that hostkey/cert must be owned by puppet:puppet.
I tried to keep original path (and also owner) of both files and seems
to puppet still works...
what problems could it cause to my conf?
# grep lib mongrel.conf
S
Hi,
anyone could explain why every time that puppet compiles a catalog for a
host, its compilation time increases?
Hosts share same "type", so, from my understanding, compiling catalog
time should decrease... (ok, from my poor understanding).
Feb 9 11:33:26 gridinstall puppetmasterd[356]: Com
I've created alittle script with yum install / remove and done
something liek:
exec {"lhcb_patch":
command => "/usr/local/sbin/lhcb.sh",
timeout => "-1",
unless => "file /opt/d-cache/dcap/lib/libdcap.so|grep 32-bit
>/dev/null",
}
Hi,
I have a distro problem with a couple of pacakges (I will be fixed in
next release).
Package A.i586 and A.x86_64 do install libraries in same place. So
libraries are 64 bits or 32 depending in the order of A package
installation (both belong to a yum group).
Now seems that metpackage (yum g
On Tue, 03 Feb 2009 09:22:20 -0500
Frank Sweetser wrote:
> > So the file changes after first start.
> > If I keep the first line (createUser one), everytime puppet runs
> > will notice that file has changed and will redo this step.
> > If I keep the end-state file, snmp does not work... I'm not
On Tue, 03 Feb 2009 07:25:38 -0500
Frank Sweetser wrote:
> It's hard to say for certain, since you didn't actually post what the
> errors are, but I see a couple issues here.
Ups, sorry, the error is that the file sometimes is created sometimes
not.
> First, based on the command line you're u
Hi all,
I have a little problem when configuring snmp.
I have to create the user, and for so I must add a line in :
/var/net-snmp/snmpd.conf
like:
createUser nagios MD5 DES XXXx
and tehn restart snmp.
I'm doing it like:
class snmp {
package { "net-snmp": }
Thanks to both for your replies.
I'll take a look and come back with my decision.
Cheers,
Arnau
--~--~-~--~~~---~--~~
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
On Thu, 22 Jan 2009 10:32:14 -0600
Mike Renfro wrote:
>
> On 1/22/2009 10:11 AM, Arnau Bria wrote:
>
> > Yep, I was thinking of inherits, but I'll have to control host group
> > like:
> > node "nodeA" "nodeB" {
> > }
> >
>
On Thu, 22 Jan 2009 08:16:48 -0700
Zach Buckholz wrote:
>
>
> Has anyone put together a puppet proposal for deploying puppet in your
> environment?
I don't understand you, sorry.
> Zach
Arnau
--~--~-~--~~~---~--~~
You received this message because you are sub
On Thu, 22 Jan 2009 09:34:11 -0600
Mike Renfro wrote:
>
> On 1/22/2009 9:10 AM, Arnau Bria wrote:
>
> > Is there ant way for grouping node in nodes.pp?
>
> The following should work for what you described. I use a similar
> model to have a parent class for all compute
Hi,
maybe it's in front of me but I can't see it, maybe I'm looking for
something in existent, but sounds strange to me.
Is there ant way for grouping node in nodes.pp?
I mean, now I have:
# cat nodes.pp
node "td234.pic.es" , "td065.pic.es", "td006.pic.es" [...]{
include worker_node
On Wed, 14 Jan 2009 12:29:03 -0600
Luke Kanies wrote:
[...]
> This directory needs to be owned by puppet/puppet.
>
> >
> > 4 drwxr-xr-x 7 puppet puppet 4096 Jan 9 12:03 ssl
> > 4 drwxr-xr-t 2 root root 4096 Jan 9 12:03 state
> > 4 drwxr-x--- 4 puppet puppet 4096 Jan 9 12:10 yaml
Thanks
Hi,
following http://reductivelabs.com/trac/puppet/wiki/ReportsAndReporting
I'm trying to confgiure reports in my frsh puppet install.
But as my puppet version has some problems creating its own dirs (I've
already opened a bug), rrd repots do not work:
Jan 14 12:36:08 gridinstall puppetmaster
On Tue, 13 Jan 2009 09:42:50 -0500
Joe McDonagh wrote:
Hi,
> Is the service already running before you run it by hand?
no.
> Can you run a ps aux | grep puppet to see if puppetmasterd does
> indeed go to the background?
[r...@gridinstall ~]# !ps
ps -ef|grep puppet
root 27312 27281 0 22:51
Hi all,
I'm with
# rpm -qa|grep puppet
puppet-server-0.24.6-1.el5
puppet-0.24.6-1.el5
Instaled from epel 5 repos.
When trying puppet with debug option with or without --no-daemonize,
puppetmasterd still goes to background:
[r...@gridinstall modules]# /usr/sbin/puppetmasterd --debug --no-
On Mon, 12 Jan 2009 17:41:11 -0600
Luke Kanies wrote:
> I'll update the FAQ; trac is being phased out, and the current
> buglist is on redmine:
>
> http://reductivelabs.com/redmine/
Thank you.
Cheers,
Arnau
--~--~-~--~~~---~--~~
You received this message because
Hi all,
I have a custom fact which determines what interface is primary in
future bonding:
It works fine when I run puppet for second time:
# /usr/bin/ruby /usr/sbin/puppetd --server=gridinstall.pic.es
--logdest=/var/log/puppet/puppet.log --test
info: Retrieving plugins
notice: /File[/var/lib
On Fri, 9 Jan 2009 10:10:37 -0600
Carl Caum wrote:
>
> Try adding an exec statement that runs at the beginning of your
> manifest. I'd have to know your manifest to suggest the best way to
> do it. I personally have an essential package resource require it.
>
> exec {"mkdir -p /var/lib/pu
On Fri, 9 Jan 2009 04:59:57 -0800 (PST)
Rodney Quillo wrote:
>
> >parent directory /var/lib/puppet does not exist
>
> This directory seems dont exists?
Are you sure? :-)
I finally created it by hand and the other ones too, but I was asking
if this error is normal in a fresh install.
> Rodney
Hi all,
I'm doing a new install of my puppet server and I'm doing it like:
1.-) adding epel repo:
http://fedoraproject.org/wiki/EPEL/FAQ#howtouse
2.-) yum -y install puppet-server
3.-) rm -rf /etc/puppet
4.-) copy my old puppet conf (from puppet-0.24.5 to 0.24.6)
mv /etc/puppet.old /etc/pupp
Hi all,
sorry for breaking the threat, I don't know where the OP is.
anyway, Tim Harper provided yum_plus in order to be able to pass args
to yum, in his example, -x option.
http://groups.google.com/group/puppet-users/browse_thread/thread/234e84e569a40f61?hl=en#
I was looking for something sim
Hi all,
I think I must redo my question:
I must do a package clean before a yum updaet because of the package
itself:
"to update/install the CAs. Occasionally, in case the yum cache is not
updated properly one might need to perform manual cleaning with the command:"
not because yum as I said
Hi all,
When trying to update a package from:
package { "lcg-CA":
ensure => '1.25-1',
}
to
package { "lcg-CA":
ensure => '1.26-1',
}
which is the best way of managing yum update? Create an exec
yum_update/yum_clean and make
Hi all,
I'm runnig puppetd on client like:
/usr/sbin/puppetd --server=puppet.server.domain
--logdest=/var/log/puppet/puppet.log
It's doing changes but client does not log them
in /var/log/puppet/puppet.log.
For example, I wanted to install some pakages, so I add them in a
module and puppet in
Hi,
After installing a new host and running puppet for first time, I always
get this error in client:
Wed Dec 03 14:33:51 CET 2008 Puppet (info): Creating state file
/var/lib/puppet/state/state.yaml
Wed Dec 03 14:33:58 CET 2008 Puppet (err): Could not call puppetreports.report:
#
Wed Dec 03 14
On Mon, 1 Dec 2008 14:19:19 -0800 (PST)
Mark Drayton wrote:
>
> Hi
Hi Mark,
> That's correct. When you run facter on the command line it doesn't
> include facts in Puppet's factpath. I use Puppet to export the
> following to all hosts so command-line facter picks up my facts:
>
> [EMAIL PROTE
Hi,
I'm following
http://reductivelabs.com/trac/puppet/wiki/PluginsInModules in order to
use a custom fact.
My modules looks like:
# ls manifests/modules/network/
manifests plugins
# ls manifests/modules/network/plugins/facter/primaryint.rb
manifests/modules/network/plugins/facter/primaryint.
On Fri, Nov 28, 2008 at 9:33 PM, Mark Drayton <[EMAIL PROTECTED]> wrote:
>
> Hi
Hi,
[...]
thanks you very much.
>
> Follow the instructions at
> http://reductivelabs.com/trac/puppet/wiki/AddingFacts
> to distribute facts to your clients.
http://reductivelabs.com/trac/puppet/wiki/PluginsInModu
On Fri, 28 Nov 2008 12:18:05 +0100
Arnau Bria wrote:
I'll do it using an external script.
After reading :
http://reductivelabs.com/trac/puppet/wiki/LanguageTutorial#variable-expressions
I've seen I'm on version 0.24.5 not 0.24.6 so, I suppose I can't do it
Hi all,
I'm configuring bonding on my hosts. I have no problem on
copying/modifying netwrok-scripts/ifcfg-*, but I have to add some lines
to modprobe.conf and modify the primary interface in function of its ip.
So, if host has an odd IP, primary interface must be eth1, if not, it
must be eth0:
Hi all,
I have a couple of files that puppet is trying to copy each time it
runs.
The first one is a plain text file with a "1" inside:
client:
# cat /opt/localconf/gLite3.1/yaim/pic/configure
1
# md5sum /opt/localconf/gLite3.1/yaim/pic/configure
b026324c6904b2a9cb4b88d6d61c81d1 /opt/localcon
Hi,
I'm running client with --logdest=/var/log/puppet.log just to get a
clean puppet log and don't mix it with messages.
# grep . /etc/sysconfig/puppet |grep -v "^#"
PUPPET_SERVER=gridinstall01.pic.es
PUPPET_LOG=/var/log/puppet.log
But I only get this message in new log:
# cat /var/log/puppet
On Wed, 19 Nov 2008 14:08:51 -0600
Mike Renfro wrote:
>
> On 11/19/2008 4:42 AM, Arnau Bria wrote:
>
> > Yep, after reading a little more (puppet is really complex) I have
> > modified things and now I have dependencies between types.
> > All my modules are, now,
On Wed, 19 Nov 2008 13:35:12 +0100
Arnau Bria wrote:
> Hi,
>
> after reading Paul's reply to one of my questions:
> http://groups.google.com/group/puppet-users/browse_thread/thread/4753569f04aee731?hl=en
>
> I tried to redefine all types dependencies and trasnfor
Hi,
after reading Paul's reply to one of my questions:
http://groups.google.com/group/puppet-users/browse_thread/thread/4753569f04aee731?hl=en
I tried to redefine all types dependencies and trasnform them
into definiton dependencies (trying to simplify all my conf).
So, my wn_class looks like:
Hi all,
I've seen this script which "convert a specified passwd file into the
user::virtual class ".
I've noticed that it does not take in count secondary groups. I've a
perl script which does:
--
#!/usr/bin/
On Tue, 18 Nov 2008 16:47:51 -0800
Paul Lathrop wrote:
Hi Paul,
first, thanks for your reply,
> On Mon, Nov 17, 2008 at 7:56 AM, Arnau Bria <[EMAIL PROTECTED]> wrote:
[...]
> You must define dependencies at the "resource" level (though you can
> also define dependen
On Wed, 19 Nov 2008 00:27:24 +1300
Aj Aj wrote:
>
> You'll want to use the membership parameter to achieve this, check
> the type ref.
I've done. But I don't know how membership could help me
I've tried with both values: inclusive, minimum.
user { "arnaubria":
ensure => present,
On Tue, 18 Nov 2008 11:32:32 +0100
Thomas Bellman wrote:
>
> Arnau Bria wrote:
>
> > In the example I have notify and subscribe, but I've been playing
> > with only one, with only the other and both, and ALWAYS exec
> > "yaim_conf" is executed.
>
&
On Tue, 18 Nov 2008 11:22:40 +0100
Peter Meier wrote:
>
> Hi
[...]
> it does. However exec execs every time puppet runs, unless you specify
> certain conditions. Like unless, onlyif or refreshonly.
Ok, I tried to define a unless condition, but I had no good ideas for
that (it's not as easy as r
Hi all,
Hi have something like:
file { "/opt/localconf/gLite3.1/yaim":
ensure => directory,
recurse => "true",
owner => "root",
group => "root",
mode=> "755",
source =>
"puppet://gri
On Mon, 17 Nov 2008 17:24:27 -0600
Luke Kanies wrote:
Hi Luke,
> > Is delete / recreate the only solution?
>
> What do you mean?
# id arnaubria
uid=63003(arnaubria) gid=31401(grid) groups=31401(grid)
user { "arnaubria":
ensure=> present,
gid => "31401",
home
Hi all,
I've been playing with puppet for a couple of weeks and that's how I
have configured it:
# ls /etc/puppet/manifests/
modules site.pp
# ls /etc/puppet/manifests/modules/
autofs_home fstab local_conf local_groups local_users packages pbs_client
repos yaim
and each "module" is
Hi all,
I'm using "users/group" type for user/group creation.
But how to remove user secondary groups?
Is delete / recreate the only solution?
Cheers,
Arnau
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"Puppe
Solved:
]# puppetca --list
Could not load confine test 'operatingsystem': No such file to load --
puppet/provider/confine/operatingsystem Could not load confine test
'operatingsystem': No such file to load --
puppet/provider/confine/operatingsystem Could not load confine test
'operatingsystem': N
*Sorry for breaking the treat
Andrew Shafer said:
> The error is telling you that the certificate isn't trusted.
> You just rebuilt the server, but the client is the same?
no, I rebuilt both... but the ssl dir was the same.
> You should blow away the certs on the client and get
Hi all,
last week I was able to configure client-server with a CentOs 5.2 and
Scientifi Linux 4 (RH ent 4).
Now, I must change my server to SL4, and my repos give me version 0.22,
and I found some problems with rrdtool, again.
So, I'd like to install version 0.24 using epel repos:
http://downlo
On Fri, 7 Nov 2008 10:06:17 -0600
Evan Hisey wrote:
[...]
> Okay simple check, as you did not install by Yum, do you rrdtool
> installed? I would think you do but wanted to check.
> Also the current
> version of puppet for Centos/RHEL is 0.24.5, not 0.22. I really
> suggest you upgrade. I suspec
On Fri, 7 Nov 2008 09:09:02 -0600
Evan Hisey wrote:
Hi Evan,
> I already did a rebuild. I ran in to this a while ago setting up
> puppetview. Here is the package I built for Centos and Puppet:
> http://unix.eng.ua.edu/~barnowl/RubyRRDtool-0.6.0-1.i386.rpm
Yep, I followed your treat:
> > http:
Hi all,
well, I think I have some things clear now.
I still have same error:
warning: RRD library is missing; cannot store metrics
But I think I have the source of the problem, maybe it's a development
issue.
My scenario:
# cat /etc/redhat-release
CentOS release 5.2 (Final)
# rpm -qa|grep pu
On Thu, 6 Nov 2008 16:43:31 +0100
Arnau Bria wrote:
> err: Report rrdgraph failed: Could not create RRD
> file /var/lib/puppet/rrd/td234.pic.es/time.rrd: invalid argument -
> Bignum, expected T_STRING or T_FIXNUM on index 2
>
I've installed http://rubyforge.org/projects/r
> logs say nothing about rrd or something similar, where may I find some
> info about what's happening?
I got the error when debbuging:
err: Report rrdgraph failed: Could not create RRD file
/var/lib/puppet/rrd/td234.pic.es/time.rrd: invalid argument - Bignum, expected
T_STRING or T_FIXNUM on
On Wed, 5 Nov 2008 14:58:49 +0100
Arnau Bria wrote:
Hi,
I solved some reporting problems:
[...]
bad:
> [puppetmasterd]
> reports = rrdgraph,store,true
> reportdir = /tmp/puppet/
> rrddir = /var/www/html/rdd/
> rrdin
On Wed, 05 Nov 2008 18:48:12 +0100
Peter Meier wrote:
[...]
> http://reductivelabs.com/trac/puppet/wiki/FrequentlyAskedQuestions#how-can-i-manage-whole-directories-of-files-without-explicitly-listing-the-files
my god, I did some search on the wiki, but did not find that link.
many thanks!
> gr
On Wed, 05 Nov 2008 17:37:45 +0100
Peter Meier wrote:
>
> Hi
Hi pete,
> > how may I do something like:
> > file { "/usr/local/sbin/*":
> > ensure => file,
> > owner => "root",
> > group => "root",
> > mode => "700",
> >
Hi,
how may I do something like:
file { "/usr/local/sbin/*":
ensure => file,
owner => "root",
group => "root",
mode => "700",
source => "puppet://gridinstall01.pic.es/files/usr/local/sbin/",
}
Hi again,
I'm trying to configure puppetview
(http://www.devco.net/pubwiki/Puppet/PuppetView) cause is one
interesting thing I'd like to evaluate before moving to puppet.
I'm following the above link but now I have some problems when sending
reports from client to server.
It says that I must
Thanks for your replies.
Arnau
--~--~-~--~~~---~--~~
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, send email to [E
Hi all,
I'm starting my puppet tests... I've done some basci conf (auto-fs,
fstab...) but now, I'm stopped in a yum problem.
First, how do you manage yum updates?¿ exec? I've seen some apt-example (
http://reductivelabs.com/trac/puppet/wiki/Recipes/Apt_Repositories) and
there they use exec, but as
On Thu, Oct 30, 2008 at 6:12 PM, Andrew Shafer <[EMAIL PROTECTED]>wrote:
> Arnau,
Hi Andrew,
>
> The groupinstall is specific behavior to yum, it isn't an abstraction
> supported by Package resource.
Yep,
*yum*: Support via yum. Required binaries: python, rpm, yum. Default for
operatingsystem
On Thu, Oct 30, 2008 at 5:39 PM, RijilV <[EMAIL PROTECTED]> wrote:
> > So, I'm not using "package" here, I'm doing it with exec, telling that if
> > the package is foobar, execute "yum groupinstall"... but then, it will be
> > logged out of package type section... is it?
>
> Honestly I wouldn't u
On Wed, Oct 29, 2008 at 8:48 PM, RijilV <[EMAIL PROTECTED]> wrote:
>
> 2008/10/29 Arnau <[EMAIL PROTECTED]>:
[...]
> Sure, you should be able to do that. For the yum groupinstall you
> might have to run that as an exec (someone else here might know more),
> but in general you'd have something
Hi all,
my name is Arnau Bria and I'm a sys admin in a center where we
must deal with hundred hosts. We're currently working with quattor,
but it's too complex for our purposes, so I'm looking for new admin
tool.
I've been playing with CFengine for few days (2
201 - 272 of 272 matches
Mail list logo
