Hi,

When I run

sudo newrelic_resque_agent install

I have a message

 /usr/bin/env: ruby: No such file or directory

Hi,
how can you access a Redis that is protect by a password after settings in Redis instance config requirepass ?

how do you access Redis Master instance (have 1 Master and 1 slave) and in front of it have multiple Sentinel ?

Regards

Eric

With the commit to allow multiple instances with one agent (see: 5cb8f2a#diff-d7cd39b75e98d8d6bb7a02b42f9a2574) it appears that you have lost the ability to run this agent simultaneously with other agents. It seems that you are using the agent level of config to drive the additional instances; however I think that's going to cause the resque agent to also pick up parameters for every other agent setup in newrelic_plugin.yml.

For example if you tried to have something like this:

agents:
  varnish:
    -
      name: "Local Varnish"
      vname:

  my_resque_1:
    # Redis connection string: 'hostname:port' or 'hostname:port:db' or 'redis://user:password@hostname:port:db'
    redis: localhost:6379
    # Resque namespace
    namespace:
  my_resque_2:
    # Redis connection string: 'hostname:port' or 'hostname:port:db' or 'redis://user:password@hostname:port:db'
    redis: localhost:6379
    # Resque namespace
    namespace:

Then I believe the resque plugin would attempt to execute:

NewRelic::Plugin::Setup.install_agent varnish, ResqueAgent

There is an example of how to run multiple agents per plugin here, although I'm not sure it would work in this scenario: https://docs.newrelic.com/docs/plugin-dev/multiple-agents-for-plugins-example

It would be nice if this was bundled as a gem.

Setting an alert in newrelic based on the number of background workers alerts when the number of background workers is higher than the set value rather than lower. Shouldn't the alert work the other way around? If background worker count drops below the set value it should alert rather than alert when background worker count is above the set value.

For example: in the haproxy i'd like to monitor the local 'www' haproxy config but would like to have it reported as 'www on server-x-y-z'...
The usual config in other plugins is 'name' (or a derivative)

We have multiple different workers running in the same name space. For example email sending workers and other background jobs. They each have a different queue, How do I configure the plug in to report for each queue ?

Trying to get your Haproxy plugin for New Relic to work with no success.
Here's what's going on:

travis@ip-xxx-xxx-xxx-xxx:/newrelic_haproxy_agent$ bundle install
Using multipart-post (1.2.0)
Using faraday (0.8.6)
Using json (1.7.7)
Using newrelic_plugin (0.2.11) from [email protected]:newrelic-platform/newrelic_plugin.git (at master)
Using bundler (1.0.15)
Your bundle is complete! Use bundle show [gemname] to see where a bundled gem is installed.
travis@ip-xxx-xxx-xxx:/newrelic_haproxy_agent$ ruby newrelic_haproxy_agent.rb
Faraday: you may want to install system_timer for reliable timeouts
newrelic_haproxy_agent.rb:21: undefined method `needs' for main:Object (NoMethodError)
travis@ip-xxx-xxx-xxx-xxx:~/newrelic_haproxy_agent$

Here's my gem env:
RubyGems Environment:

• RUBYGEMS VERSION: 1.8.15
• RUBY VERSION: 1.8.7 (2011-06-30 patchlevel 352) [x86_64-linux]
• INSTALLATION DIRECTORY: /var/lib/gems/1.8
• RUBY EXECUTABLE: /usr/bin/ruby1.8
• EXECUTABLE DIRECTORY: /usr/local/bin
• RUBYGEMS PLATFORMS:
• ruby
• x86_64-linux
• GEM PATHS:
• /var/lib/gems/1.8
• /home/travis/.gem/ruby/1.8
• GEM CONFIGURATION:
• :update_sources => true
• :verbose => true
• :benchmark => false
• :backtrace => false
• :bulk_threshold => 1000
• REMOTE SOURCES:
• http://rubygems.org/

Here's my gem list:
*** LOCAL GEMS ***

faraday (0.8.7, 0.8.6)
fastercsv (1.5.5)
json (1.8.0, 1.7.7)
multipart-post (1.2.0)

I added a haproxy configuration and after some time I noticed that I had values for sessions and throughput, but requests/min were always 0. I believe since this value is 0, I am unable to create an alert condition through the newrelic web ui. It would be nice if you could setup the alerts as soon as the configuration shows up in new relic.

Is there some example on how to check 404, 301 or Googlebot access?
From whaat I've understand I should only put the string I want to check, but some example should be good for a lot of people.

Probably also some screenshot on how the plugin works will be appreciated. I'd like to see what I can do with it before installing.

When I add this entry into the generated newrelic yml file and run the agent i keep getting the error below:

I'd like to make a absolute path to a file, I've tried many different ways, could someone help?

log_path: c:/log.txt
log_path: c:\log.txt
log_path: c:\log.txt
log_path: log.txt

yml file section:
agents:
my_logfile_1:
# Full path to the the log file
log_path: c:/log.log
# Returns the number of matches for this term. Use Linux Regex formatting.
# Default: "[Ee]rror"
term: "[Ee]rror "
# Provide any options to pass to grep when running.
# For example, to count non-matching lines, enter 'v'.
# Use the abbreviated format ('v' and not 'invert-match').
grep_options:

c:\Scripts\newrelic_logwatcher_agent>ruby newrelic_logwatcher_agent.rb
Agent Logwatcher is at version 0.0.2
c:/Scripts/newrelic_logwatcher_agent -> This is the cur directory
c:/log.log
Error occurred in poll cycle: No such file or directory - test -e c:/log.log
Gathered 0 statistics

Hi, I'm trying to parse a log that is something like "blabalbla [blablabla] [error] bla bla bal bla bla bla"
To parse it I'm using the regex ".error." but it seems to be not working because nothing new appears in the Logwacher entry for my site.
Am I missing something or it's bug ? Any advice is much appreciated.

Trying to install the newrelic logwatcher.

ruby newrelic_logwatcher_agent.rb
[Thu Mar 27 19:37:08 UTC 2014] WARN: SSL is disabled by default when using Ruby 1.8.x
newrelic_logwatcher_agent.rb:130: undefined method `keys' for nil:NilClass (NoMethodError)

Here's the failing code: NewRelic::Plugin::Config.config.agents.keys.each do |agent|

Hi,

We installed the HA Proxy plugin and it works ok - although our "Servers" graph is empty.

Hope someone can let me know what could be the problem or how this could be fixed.

Thanks

Allow the ability to monitor multiple proxies from one config file.

Hi,

How do I go about installing the logwatcher plugin?

I am barely familiar with Ruby so the instructions which start by saying 'bundle install' do not make it clear what I need to do first. Will the logwatcher plugin be installed if I type 'bundle install'?

I'm an experienced sysadmin but not in the area of Ruby plugins.

Cheers,
Ian

I'm using the haproxy plugin, it is running but sometimes it stops sending data to newrelic. We have to restart the plugin to start sending data again and there is no information about it in the logs. Could you please help me solving this issue?

Hi dudes!
When i try to monitor HAproxy with this plugin i getting this error in console

root@development:~# newrelic_haproxy_agent run -v
[2013-09-04 05:07:22 UTC] Agent Haproxy is at version 1.0.1
[2013-09-04 05:07:22 UTC] Agent Haproxy is at version 1.0.1
[2013-09-04 05:07:22 UTC] Error occurred in poll cycle: SSL_connect returned=1 errno=0 state=SSLv3 read server certificate B: certificate verify failed
[2013-09-04 05:07:22 UTC] Error occurred in poll cycle: SSL_connect returned=1 errno=0 state=SSLv3 read server certificate B: certificate verify failed
[2013-09-04 05:07:22 UTC] Gathered 0 statistics

How i can disable SSL cert checking ?

https://www.dropbox.com/s/2apa35d7nebbxce/Resque_-_New_Relic_and_1__root_i-6424277d__etc_newrelic__ssh_.png?dl=0

OMG....

After trying to poke around the log watcher script with my non existant ruby and perl skills i'd like to request a new plugin: Postfix :)

I personally was trying to base one off the various munin-plugins (see https://github.com/munin-monitoring/contrib/blob/master/plugins/mail/)

but got stuck before even getting anywhere :) But i do see potential for a postfix plugin in conjunction with the logwatch plugin.
The code would have to 'read' / 'watch' one logfile (on most systems /var/log/mail.info) and do some 'fairly' simple commands:

spooldir = /var/spool/postfix;
cd spooldir
deferred.value = `(test -d deferred && find deferred -type f ) | wc -l`
active.value  = `(test -d active && find active -type f ) | wc -l`
maildrop.value = `(test -d maildrop && find maildrop -type f ) | wc -l`
incoming.value = `(test -d incoming && find incoming -type f ) | wc -l`
corrupt.value = `(test -d corrupt && find corrupt -type f ) | wc -l`
hold.value = `( test -d hold && find hold -type f ) | wc -l`

calculating throuhput (bytes) (as taken from one of the munin plugins)

througput: (/var/log/mail.log)
while ($line =<$LOGFILE>) {
    chomp ($line);
    if ($line =~ /qmgr.*from=.*size=([0-9]+)/) {
        $volume += $1;
        $deliveryCount++;
    } elsif ($line =~ /postfix\/smtpd.*reject: \S+ \S+ \S+ (\S+)/ ||
           $line =~ /postfix\/cleanup.* reject: (\S+)/)
    {
        $rejects->{$1}++;
    }

}

Which is basically the parser from the logwatcher plugin but instead of just counting, just holding some numbers... right?

Anyway, i'd be a happy user if this plugin was there (or if i ever get around to finishing my-not-so-far venture of trying to build one myself)

It would be good if it included an init.d script for starting this as a service.

Hi,

I am watching the logs using Logwatcher but not seeing any options to configure the alerts. Am I missing any configurations? Or is it a bug?

Please see the attached screenshot.

Hello, I've installed using steps from readme, on centos 6:

yum install ruby, rubygems, gcc, ruby-devel -y
gem install newrelic_haproxy_agent

tried to to open "/etc/newrelic/newrelic_haproxy_agent.yml " but there's no such file.

EDIT: then i saw it's my fault and i missed a step "sudo newrelic_haproxy_agent install".
but running this command says some ruby errors, missing libraries? how do i know what else to install, is there a list of needed libraries or something?

Thank you.

P.S. here's the error:
/usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:in gem_original_require': no such file to load -- bundler/setup (LoadError) from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:inrequire'
from /usr/lib/ruby/gems/1.8/gems/newrelic_haproxy_agent-1.0.1/lib/newrelic_haproxy_agent.rb:4
from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:in gem_original_require' from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:inrequire'
from /usr/lib/ruby/gems/1.8/gems/newrelic_haproxy_agent-1.0.1/bin/newrelic_haproxy_agent:5
from /usr/bin/newrelic_haproxy_agent:19:in `load'
from /usr/bin/newrelic_haproxy_agent:19

EDIT: it's ok figured out,
gem install bundler
gem install fastercsv

then worked.
Thanks.

Hi,

Is it possible implement a new feature to show jobs/pending for each queue?

Thanks

I followed the intructions in the README, but it doesn't seem to work.

1. I add the gem
2. I created the config/newrelic_plugin.yml file as follow :

newrelic:
  license_key: <%= ENV['NEW_RELIC_LICENSE_KEY'] %>
  verbose: 0

agents:
  my_resque_agent_name:
    redis: <%= ENV["REDISCLOUD_URL"] %>
    namespace:

3. I added resque_monitoring: newrelic_resque_agent run -c config/newrelic_plugin.yml to my Procfile
4. I scaled my dynos

My ENV['REDISCLOUD_URL'] looks like redis://rediscloud:[email protected]:16503, I'm using the rediscloud heroku adds-on and the database is not specified.

I've been waiting for a wile, but I don't find any resque monitoring on my newrelic interface (nothing in the plugin section).

I am running the haproxy plugin and am getting data for sessions and traffic, but requests are always 0. If I watch the haproxy stats csv the req_rate values are above 0. I am running haproxy 1.4.18 and am running the newest newrelic haproxy plugin from github.

Is possible to run the logwatcher as a daemon or to make it run continuosly?

How the logwatcher graph should be read?
You show the occurence on a per minute base?

I dont seem to be collecting server counts?

Sent 8 metrics to New Relic [https://platform-api.newrelic.com]:
[Mon Jul 01 06:55:52 UTC 2013] Component/Requests[Requests/Minute]: count: 1, total: 0.0, min: 0.0, max: 0.0, sum_of_squares: 0.0
[Mon Jul 01 06:55:52 UTC 2013] Component/Errors/Request[Errors/Minute]: count: 1, total: 0.0, min: 0.0, max: 0.0, sum_of_squares: 0.0
[Mon Jul 01 06:55:52 UTC 2013] Component/Errors/Connection[Errors/Minute]: count: 1, total: 0.0, min: 0.0, max: 0.0, sum_of_squares: 0.0
[Mon Jul 01 06:55:52 UTC 2013] Component/Errors/Response[Errors/Minute]: count: 1, total: 0.0, min: 0.0, max: 0.0, sum_of_squares: 0.0
[Mon Jul 01 06:55:52 UTC 2013] Component/Bytes/Received[Bytes/Seconds]: count: 1, total: 4216.1936466624, min: 4216.1936466624, max: 4216.1936466624, sum_of_squares: 17776288.8661564
[Mon Jul 01 06:55:52 UTC 2013] Component/Bytes/Sent[Bytes/Seconds]: count: 1, total: 180724.703665754, min: 180724.703665754, max: 180724.703665754, sum_of_squares: 32661418515.0747
[Mon Jul 01 06:55:52 UTC 2013] Component/Sessions/Active[Sessions]: count: 1, total: 7.0, min: 7.0, max: 7.0, sum_of_squares: 49.0
[Mon Jul 01 06:55:52 UTC 2013] Component/ProxyUp[Status]: count: 1, total: 1.0, min: 1.0, max: 1.0, sum_of_squares: 1.0

Am i missing something?

follow the instructions, while execute the plugin in commandline:

[root@ip-10-0-3-13 newrelic_logwatcher_agent]# ruby newrelic_logwatcher_agent.rb
newrelic_logwatcher_agent.rb:130:in <module:LogwatcherAgent>': undefined methodkeys' for nil:NilClass (NoMethodError)
from newrelic_logwatcher_agent.rb:66:in `(main)'

The example described newrelic_plugin.yml and the readme indicate you can use the connection the string redis://user:password@hostname:port:db, but it isn't working for me. However, redis://user:password@hostname:port/db does work. The difference is the forward slash between port and db.

The other configurations have worked perfectly for me.

Hi Guys
Not sure if I am missing something but here are my questions

• Why is newrelic_plugin.yml hardcoded to be located in /etc/newrelic makes it harder to deploy to Heroku. I think this is why you require to install the gem with sudo ? It would be great to not have to do that
• Have you considered using environment variables instead ? These are much safer and easier to update without requiring a push for example and also not needing to run
  newrelic_resque_agent install

I'm using SmartOS, where app-specific config files get written to /opt/local/etc instead of /etc. The -c option will work great for loading the config from there, but the "install" command expects hard-coded paths.

There's a pretty simple workaround; I'll just copy the .yml.example manually and stick it there, but it'd be nice if the -c option worked consistently.

Hi, appreciate your work on this plugin. Not being real familiar with ruby/gems, I didn't realize this ominous-looking trace just means that I need to gem install bundler:

$ newrelic_haproxy_agent install    
/usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `gem_original_require': no such file to load -- bundler/setup (LoadError)
    from /usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `require'
    from /var/lib/gems/1.8/gems/newrelic_haproxy_agent-1.0.1/lib/newrelic_haproxy_agent.rb:4
    from /usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `gem_original_require'
    from /usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `require'
    from /var/lib/gems/1.8/gems/newrelic_haproxy_agent-1.0.1/bin/newrelic_haproxy_agent:5
    from /usr/local/bin/newrelic_haproxy_agent:19:in `load'
    from /usr/local/bin/newrelic_haproxy_agent:19

same with fastercsv. Is there any way to have them as dependencies of the newrelic_haproxy_agent gem? Or mention in your README that they are required and may need to be installed separately? In my case, it was needed with Ubuntu 12.04. Thanks!

I get this error when attempting to get newrelic_logwatcher_agent by running the following:

rvmsudo gem install newrelic_logwatcher_agent

NOTE: when I run: rvmsudo gem install newrelic_resque_agent - it works fine.

So, it looks like newrelic_logwatcher_agent isn't in the repository?

Please help...
Thanks... Clark

In attempting to run the plugin from a command line I get the following error:

[ /usr/local/newrelic_mpstat_agent]# ruby newrelic_mpstat_agent.rb
Agent Mpstat is at version 0.0.2
Error occurred in poll cycle: Couldn't parse output. Make sure you have mpstat installed. undefined method `split' for nil:NilClass
Gathered 0 statistics

I have verified that /bin/mpstat is working and updated the yml file as follows:

Agent Configuration:

agents:
mpstat:
# The command used to display MP statistics
command: /bin/mpstat
# Report current usage as the average over this many seconds.
interval:

I do not know if this is my fault for not knowing ruby or an issue with the packages.

Got the same issue on both packages, installed Ruby and now running 1.9.1.

Issue with package, or any advice?

root@server:~# newrelic_haproxy_agent install
/usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require': cannot load such file -- bundler/setup (LoadError)
    from /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require'
    from /var/lib/gems/1.9.1/gems/newrelic_haproxy_agent-1.0.1/lib/newrelic_haproxy_agent.rb:4:in `<top (required)>'
    from /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require'
    from /usr/lib/ruby/1.9.1/rubygems/custom_require.rb:36:in `require'
    from /var/lib/gems/1.9.1/gems/newrelic_haproxy_agent-1.0.1/bin/newrelic_haproxy_agent:5:in `<top (required)>'
    from /usr/local/bin/newrelic_haproxy_agent:19:in `load'
    from /usr/local/bin/newrelic_haproxy_agent:19:in `<main>'

root@server:~# newrelic_resque_agent install
/usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `gem_original_require': no such file to load -- bundler/setup (LoadError)
    from /usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `require'
    from /var/lib/gems/1.8/gems/newrelic_resque_agent-1.0.1/lib/newrelic_resque_agent.rb:4
    from /usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `gem_original_require'
    from /usr/lib/ruby/vendor_ruby/1.8/rubygems/custom_require.rb:36:in `require'
    from /var/lib/gems/1.8/gems/newrelic_resque_agent-1.0.1/bin/newrelic_resque_agent:5
    from /usr/local/bin/newrelic_resque_agent:19:in `load'
    from /usr/local/bin/newrelic_resque_agent:19

Following the available documentation: step 1

root@myservername:~# ln -s /usr/bin/gem1.9.1 /usr/bin/gem
root@myservername:~# gem list

*** LOCAL GEMS ***


root@myservername:~# gem install newrelic_haproxy_agent
Building native extensions. This could take a while...
To get started with this plugin, do
newrelic_haproxy_agent -h
to find out how to install and run the plugin agent.
Successfully installed multipart-post-2.0.0
Successfully installed faraday-0.9.0
Successfully installed json-1.8.1
Successfully installed newrelic_plugin-1.0.3
Successfully installed newrelic_haproxy_agent-1.0.1
5 gems installed
Installing ri documentation for multipart-post-2.0.0...
Installing ri documentation for faraday-0.9.0...
Installing ri documentation for json-1.8.1...
Installing ri documentation for newrelic_plugin-1.0.3...
Installing ri documentation for newrelic_haproxy_agent-1.0.1...
Installing RDoc documentation for multipart-post-2.0.0...
Installing RDoc documentation for faraday-0.9.0...
Installing RDoc documentation for json-1.8.1...
Installing RDoc documentation for newrelic_plugin-1.0.3...
Installing RDoc documentation for newrelic_haproxy_agent-1.0.1...

root@myservername:~# gem list

*** LOCAL GEMS ***

faraday (0.9.0)
json (1.8.1)
multipart-post (2.0.0)
newrelic_haproxy_agent (1.0.1)
newrelic_plugin (1.0.3)

Trying step 2:

root@myservername:~# newrelic_haproxy_agent install
-bash: newrelic_haproxy_agent: command not found

Command not found, so obviously the newrelic_haproxy_agent is not in $PATH

root@myservername:~# locate newrelic | grep bin
/var/lib/apt/lists/apt.newrelic.com_debian_dists_newrelic_non-free_binary-amd64_Packages
/var/lib/gems/1.9.1/bin/newrelic_haproxy_agent
/var/lib/gems/1.9.1/gems/newrelic_haproxy_agent-1.0.1/bin
/var/lib/gems/1.9.1/gems/newrelic_haproxy_agent-1.0.1/bin/newrelic_haproxy_agent

Looks like ithe gem was installed:

root@myservername:~/.gem# /var/lib/gems/1.9.1/bin/newrelic_haproxy_agent install
<internal:lib/rubygems/custom_require>:29:in "require": no such file to load -- bundler/setup (LoadError)
from <internal:lib/rubygems/custom_require>:29:in "require"
from /var/lib/gems/1.9.1/gems/newrelic_haproxy_agent-1.0.1/lib/newrelic_haproxy_agent.rb:4:in "<top (required)>"
from <internal:lib/rubygems/custom_require>:29:in "require"
from <internal:lib/rubygems/custom_require>:29:in "require"
from /var/lib/gems/1.9.1/gems/newrelic_haproxy_agent-1.0.1/bin/newrelic_haproxy_agent:5:in "<top (required)>"
from /var/lib/gems/1.9.1/bin/newrelic_haproxy_agent:19:in "load"
from /var/lib/gems/1.9.1/bin/newrelic_haproxy_agent:19:in "<main>"

I have no experience with ruby, so I'm stuck at step 2.

For example, if I have two HAproxy servers with identical configs, I want to see the "www" frontend proxy statistics aggregated into a single graph on New Relic. I tried simply using the same API key, but doesn't seem to be working.

Hi.

I can't get the haproxy plugin to run on the following setup:

CentOS 6.4 x86_64
haproxy v 1.5-dev17
Agent Haproxy v 1.0.1

It connect fine to haproxy and Newrelic website but not data goes to the site other than:
Servers Active (works)
Proxy status (works)
No data from Requests/Errors/Traffic or Sessions gets to the site

Can someone confirm that haproxy 1.5-dev train and haproxy agent 1.0.1 work together?

Hi,

Just recently installed the plugin to monitor our resque stack. Noticed that no alerts have been configured as of yet. Are there any plans to do so? Can I take a crack at building some alerts for the plugin?

Would love to have an alert when the number of our workers drops below a certain threshold, indicating that some have failed.

Would also love for the plugin to be able to track individual queues.

Thanks for the great plugin!
Dan

Hi,

We are using newrelic_haproxy_agent newrelic plugin for haproxy.. We have multiple haproxy instances.. We need to group them in one single group so we can see the stats for all in one group.. How can we achieve this ??

How do I delete them?

Will they delete themselves?

Hi All,

I've installed the plugin and all is working but the alert. I have 3 log files where I monitor the [DONE] term. If the [DONE] is not there I need an alert to notify me.

But whatever I try it's not working and I think because it's always greather then a number ?

[edit]
Found out that it's only working if the amount of times something happens in the log is increasing instead of decreasing. So when errors occur and they occur more than 10 time a minute you can set the alert to 10. This is working but for my purpose I need when the amount of times something occurs in the log decreases. Any work around for me?
[/edit]

Can we use the logwatcher plugin for Windows.If yes, can you please let me know the steps to configure it.

Thanks

It would be good if we could put our config in a standard location such as /etc/newrelic/newrelic_resque_agent.yml and it would check there before any other locations.

How do I download the resque plugin (by Railwares Inc) for New Relic?

The other plugins have "Download" buttons. (sorry - I'm pretty new with New Relic) I couldn't find one, or an equivalent, for this plugin.

Thanks...
Clark (email: [email protected])

We have installed newrelic-haoxy-agent in a linux server, but when we run the agent it is not running. See the output below. What could be reason behind this?

[root@pm-app03 ~]# newrelic_haproxy_agent run
[Thu Jan 16 14:34:23 UTC 2014] Agent Haproxy is at version 1.0.1
[Thu Jan 16 14:34:23 UTC 2014] Agent Haproxy is at version 1.0.1
[Thu Jan 16 14:34:23 UTC 2014] Error occurred in poll cycle: Connection refused - connect(2)
[Thu Jan 16 14:34:23 UTC 2014] Error occurred in poll cycle: Connection refused - connect(2)
[Thu Jan 16 14:34:23 UTC 2014] Gathered 0 statistics

Regards
Sikat

The HAProxy plugin uses stot to determine the number of HTTP requests per minute, but this is actually the number of cummulative connections. req_rate should be used instead.

Getting this error after it says everything installed properly.