Hello,We are interested in collecting puppet profile data from the puppet master in the real world. This data will be used to help us find a direction to go to increase the throughput of the puppet master master.
I've created a script that will create a yaml file that will have various system facts and profile data that you will need to generate from a master.To generate this data follow the steps listed below.If you would like to do so send the apache log files for the time that the puppet master was running as well, it will help us see how the agent acts during the profile. We realize that some may not be comfortable doing this so it is completely optional.
- Download the profiler script from https://github.com/demophoon/puppet-master-profiler
- Run your puppet master with the `--debug`, `--profile`, and `--logdest=<path_to_a_blank_file>` flags
- Wait for a few agents to check-in. You can tail the log file to see agents checking in to the master.
- From the repository run `profile.rb <path_to_log_file>`. The log file will the the same path as the one specified above.
- Answer the questions on screen and email the result file to me at br...@puppetlabs.com.
If you have any questions or concerns please feel free to ask them here!Thank you,Britt Gresham
--
You received this message because you are subscribed to the Google Groups "Puppet Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-dev/5c2ec333-d92a-441b-98a8-8dc62e51795d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
On 29 September 2014 18:20, Andy Parker <an...@puppetlabs.com> wrote:
> We haven't gotten any responses to this yet. :( Getting this information
> will really help in allowing us to target the right places for performance
> improvements.
Quite a few of us on this list run puppet in masterless mode so we
might not be the ideal target market. If you put a blog post up about
it I'm sure some of us will be happy to retweet/link to it for wider
viewing.
> If you aren't able to run this in a production environment, that is fine as
> well. We just need some information from installations that have some load
> and are configured in the way it is normally used.
As a side note I did have a look through the script to see if it'd be
something I would run in prod if I had a master and the length of the
code, most of which seems to be vendoring an optparser, is going to
deter quite a few admins from reviewing and running it (IMHO).
Dean
--
Dean Wilson http://www.unixdaemon.net
Profanity is the one language all programmers understand
--- Anon
--
You received this message because you are subscribed to the Google Groups "Puppet Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-dev/CAFbDO0e8HdHQCdUNvh054sx%3D5DFyL14-wwTS0DjZeMGyQwEjFg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
Britt, I had a similar concern about vendoring the code and I think Dean
is right. The script should be made as straightforward, simple, and small
as possible so that people can understand what it is doing at a glance.
Quite a few of us on this list run puppet in masterless mode so we
might not be the ideal target market. If you put a blog post up about
it I'm sure some of us will be happy to retweet/link to it for wider
viewing.