Hi Curt, I realize this is 3 year s later, but do you have any idea how one
debugs such an issue in 2012?
Cheers,
Virgil
On Friday, July 3, 2009 2:13:14 PM UTC-7, Curt Micol wrote:
>
> On Fri, Jul 3, 2009 at 12:05 PM, Martin
> Wheldon>
> wrote:
> >
> > Hi Curt,
> >
> > Did you manage to resolv
On Fri, Jul 3, 2009 at 12:05 PM, Martin Wheldon wrote:
>
> Hi Curt,
>
> Did you manage to resolve this issue? If not I think it may be the
> format of your yaml output.
>
>>---
>>classes:
>>- custom
>>- monitoring::base
>
> Should be this
>
> ---
> classes:
> - custom
> - monitoring::base
>
>
Hi Curt,
Did you manage to resolve this issue? If not I think it may be the
format of your yaml output.
>---
>classes:
>- custom
>- monitoring::base
Should be this
---
classes:
- custom
- monitoring::base
Note the indentation. Having said that I have been unable to persuade
the python y
2009/6/29 Nicolas Szalay :
> Le lundi 29 juin 2009 à 01:09 -0700, Curt Micol a écrit :
> Did you try running it as the "puppet" user ? permissions on NODEFILE &
> LOGFILE could be source of errors.
Yes, sorry, I get the same results with running as the puppet user.
Just tested once more to verify
Le lundi 29 juin 2009 à 01:09 -0700, Curt Micol a écrit :
> When I run the script directly:
> [r...@vpsadmins ~]# python /etc/puppet/tools/external_nodes.py dc2-
> vps1-400b.example.com
> ---
> classes:
> - custom
> - monitoring::base
> - ntpd
> - puppetd
> - rpms
> - sshd
> - yum
> - crontab
> -
This is not my original error, I have fixed that. My original error did not
output to stderr (from the script execution) because it was a permissions
issue with the puppet user.
I created this error as an example to show that error outputs from script
execution are being suppressed at trace level.
Le mercredi 24 juin 2009 à 10:28 +0200, Dan Bode a écrit :
> # /etc/puppet/conf/node_classifier HOST.com
> /etc/puppet/conf/node_classifier:26:in `initialize': No such file or
> directory - /etc/puppet/conf/inventory.csv (Errno::ENOENT)
> from /etc/puppet/conf/node_classifier:26:in `new'
>
I just had to solve it the old fashion way. Write statements to a log file,
and find the broken line. The issue was that a file did not have proper read
permissions for the puppet user.
trace is not very helpful
here is the server on trace:
#puppetmasterd --no-daemonize --verbose --debug --trace
Le mercredi 24 juin 2009 à 10:02 +0200, Dan Bode a écrit :
> Hi all,
>
> How can I debug external node script errors? I cant seem to figure out
> what is wrong with my script. It appears to work from the command
> line, but not from puppet.
Run your master with --debug --trace
--~--~-
