5.02021-11-21T21:33:16ZTemplatesApp e2guardian - ActiveApp e2guardian - Active## Description
https://github.com/e2guardian/e2guardian/blob/master/notes/dstats_format USER.MACRO={$E2GUARDIAN_DSTATS} UserParameter=e2guardian[*],tail -1 $1 | cut -f $2
## Overview
Introduction
============
This is my approach for a performance monitoring Template of [E2guardian](http://e2guardian.org/).
This Template was created on and for Zabbix 3.4 and E2Guardian 4.1.x on Debian 9 Stretch.
Installation
------------
Please check the GitHub Page for latest version and instructions.
Changes
-------
* February 2018: initial commit.
* 03 April 2018: Switch to Zabbix agent (active) & converted s
## Author
Marco Hofmann
Templatese2guardian- e2guardian - timestamp of dstats dumpZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},1]5m7d180dunixtime1. Time - in unix formate2guardian
- e2guardian - httpworkers totalZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},2]5m7d180d2. httpw - Number of httpworker threads runninge2guardian
- e2guardian - httpworkers busyZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},3]5m7d180d3. busy - Number of httpworker threads busy handling connections at current timee2guardian
- e2guardian - httpworkers queueZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},4]5m7d180d4. httpwQ - Number of connections waiting in queue for a worker thread.
In normal use this will be 0 (and on occasion 1 or 2) but will increase once
all worker threads are busy. Lowish numbers are OK for transitory peaks.
High numbers indicate an overloaded system. If there is still memory and
cpu available, then httpworkers may be increased.e2guardian{min(30m)}>0e2guardian - connections waiting in queue for a worker threadWARNING
- e2guardian - log queueZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},5]5m7d180d5. logQ - Number of messages waiting to be logged to disk. In normal use 0.
An increase in number may indicate system overloading, but should not
otherwise affect performance.e2guardian{min(30m)}>0e2guardian - messages waiting to be logged to diskWARNING
- e2guardian - total sum connections each 5 minutesZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},6]5m7d180d6. conx - number of connections handled since last stats line.e2guardian
- e2guardian - connections per secondZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},7]5m7d180d7. conx/s - average connections per sec over period since last stats line.e2guardian
- e2guardian - requests since last stats dumpZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},8]5m7d180d8. reqs - indication of number of requests handled since last stats line.
Only an indication as requests made over https tunnels (i.e. non-MITM) cannot
be counted, so each https tunnel is counted as one request.e2guardian
- e2guardian - requests per secondZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},9]5m7d180d9. reqs/s - average requests per sec over period since last stats line.e2guardian
- e2guardian - maximum number file descriptor in useZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},10]5m7d180d10. maxfd - Indication of maximum number file descriptor in usee2guardian
- e2guardian - Number of List Option Containers in useZABBIX_ACTIVEe2guardian[{$E2GUARDIAN_DSTATS},11]5m7d180d11. LCcnt - Number of List Option Containers in use. A new List Option Container
is created at start up and on gentle restart. Old containers are deleted
once all connections still using it have finished. Normally figure will be
one, but may rise after a gentle restart(s) and should return to one after a
period of time.e2guardian
{$E2GUARDIAN_DSTATS}/var/log/e2guardian/dstats.loge2guardian - httpworkers1200300FIXEDITEMApp e2guardian - Activee2guardian[{$E2GUARDIAN_DSTATS},2]1A7C11- App e2guardian - Activee2guardian[{$E2GUARDIAN_DSTATS},3]