Collectd write_http

Add-Ons Monitoring
1

I have a splunk instance in my network and I wanted to have the collectd service running in ipfire forward the collected metrics to that splunk instance. I have edited the file /etc/collectd.custom with the following ( the token has been changed to protect the innocent)

LoadPlugin write_http
    <Plugin write_http>
    	<Node "ipfire">
    		URL "http://10.0.0.131:8080/services/collector/raw"
                    Header "Authorization: Splunk 3999999-e585-49bb-ab5c-9681b5e18a14"
    	  	Format "JSON"
    		Metrics true
    		StoreRates true
    	</Node>
    </Plugin>

When I run the command /etc/init.d/collectd restart I’m greeted with this one log statement

Jan  3 10:42:38 ipfire collectd[26083]: collectd: Stopping 1 read threads.
Jan  3 10:42:38 ipfire collectd[26083]: ping plugin: Shutting down thread.
Jan  3 10:42:38 ipfire collectd[26083]: rrdtool plugin: Shutting down the queue thread. This may take a while.
Jan  3 10:42:41 ipfire collectd[30323]: write_http plugin: Invalid configuration option: Node.
Jan  3 10:42:41 ipfire collectd[30324]: cpufreq plugin: Found 8 CPUs
Jan  3 10:42:41 ipfire collectd[30324]: Initialization complete, entering read-loop.

specifically
Jan 3 10:42:41 ipfire collectd[30323]: write_http plugin: Invalid configuration option: Node.

I’ve reviewed the docs on collectd write_http and my entry seems to be ‘valid’. On another server the entry is the same and appears to work…

I did notice that the version of collectd we are running is pretty old. 4.10.9 - From April 7 2013 and looking at the source code it doesn’t look like the add http headers functionality was added until 3/24/2016

So is there any other way to get the collectd information off the box? Thanks

2

I am not sure what you can do do get it done without the headers option, but we could not update to collectd 5 because there are some incompatible changes in it.

We are maintaining a fork of the older version. Potentially you can patch that.