I'm struggling to figure out how to do this as I just don't know what
to search for...
What we have: e1 - e9 (e1 being dev, e9 being prod, middle
environments being test machines of various degrees).
All environments have a fixed set of websites, call them eX-website1,
eX-website2, etc...
On so
Jeremy Hansen writes:
> How do I dictate the order of how things get executed? Right now it's
> installing the package first, then doing a usermod/groupmod to set the
> uid/gid and it's trying to run the init commands before the package is
> installed, etc.
You can do it from either direction,
If there's a section in the documentation that highlights this, please
feel free to point me in the right direction. I am definitely a beginner.
I'm trying to build a recipe to install splunk clients on all our
hosts. Splunk is a commercial log aggregation utility. The RPMs splunk
distribut
are you using 0.25 beta on purpose? if its production env you should use
0.24-8.
Cheers,
Ohad
On Fri, May 15, 2009 at 2:25 AM, Marlow, Christopher (Ballston Spa) <
christopher.mar...@sgs.com> wrote:
> Hmmm.. I’m still getting the same errors as before, even using a
> secondary server to connect
Hmmm.. I’m still getting the same errors as before, even using a secondary
server to connect. It looks like the error is here:
Thu May 14 14:23:27 2009] [info] [client 10.108.111.87] Connection closed to
child 7 with standard shutdown (server spainfradev.spa.sgstestcom.com:8140)
*** Except
I have encountered a repeatable and odd problem. If I run the
following manifest snippet yumrepo over writes the
rpmfusion-free-updates.repo file after the RPM is installed:
yumrepo{rpmfusion-free-updates:
priority =>10,
enable =>1,
gpgcheck =>0,
require => Exec["r
Revocation isn’t something I’ve used before… I assume you’re talking about SSL
Certificate revocation? I have commented out the “SSLCARevocationFile” below,
but still receive the same error. I’ll bring up another test box to see if
that works better… Thanks,
SSLEngine on
Maybe related maybe not... but if you have revocation enabled in apache, you
cant use puppetd to connect to the same machine(anyone knows why btw?).
according to your logs, it seems that revocation is enabled.
cheers,
Ohad
On Thu, May 14, 2009 at 9:45 PM, Marlow, Christopher (Ballston Spa) <
ch
one rather rough solution I've used for this before, is a separate
monitoring daemon, that polls the primary server on a regular basis, if
this falls over, the daemon then can use 1 of 2 approaches
1) Rewrite the IP for the hostname of the puppetmaster in the /etc/hosts
file (really rough, but doe
I've been running Puppet using the default puppetmaster server for a few
months now, and as the sphere of influence is expanding throughout our
network, I thought I'd try my hand at upgrading the server to use
Passenger instead. I went through all the instructions, and didn't get
any error messag
Hrmmm... Local proxies on each client may introduce a fair amount of
overhead, but may actually work. The only problem with scripting and
then calling the right --server option is that it would require a
restart of the client to fail the service over to the alternate
puppetmaster - it wouldn't hap
On Thu, May 14, 2009 at 00:35, Mike Renfro wrote:
> I may be naive here, since I've never written my own types, but I'm
> wondering if a puppet definition for mysql_user would be better than
> writing a native puppet type. Something like:
>
> mysql_user { "u...@localhost": password_hash => "has
Either use a local proxy or load balancer or use a wrapper script that
checks which pm is alive and then execute puppet with the correct
--server option
Ohad
On 5/14/09, Greg wrote:
>
> Hi guys,
>
> I have a multi-site setup that I am configuring to manage via Puppet.
> At present
> there is on
13 matches
Mail list logo
