From 0ea1c74717f81df222d894d353f648cae6d28776 Mon Sep 17 00:00:00 2001
From: "Hahn Axel (hahn)" <axel.hahn@unibe.ch>
Date: Thu, 27 Jul 2023 12:24:19 +0200
Subject: [PATCH] update check_dns_responsetime; add doc page
---
check_dns_responsetime | 90 +++++++++++++++++-------
docs/20_Checks/_index.md | 2 +-
docs/20_Checks/check_dns_responsetime.md | 80 +++++++++++++++++++++
3 files changed, 147 insertions(+), 25 deletions(-)
create mode 100644 docs/20_Checks/check_dns_responsetime.md
diff --git a/check_dns_responsetime b/check_dns_responsetime
index e78903f..6e86510 100755
--- a/check_dns_responsetime
+++ b/check_dns_responsetime
@@ -11,24 +11,22 @@
# test tcp 53 first
# check result depends on 1st nameserver only
# 2022-10-24 v1.2 <axel.hahn@unibe.ch> shell fixes; remove pipe char in output
+# 2023-07-27 v1.3 <axel.hahn@iml.unibe.ch> shell fixes; add help page
# ======================================================================
-. $(dirname $0)/inc_pluginfunctions
+. $( dirname $0 )/inc_pluginfunctions
+
+self_APPNAME=$( basename $0 | tr [:lower:] [:upper:] )
+self_APPVERSION=1.3
+
tmpfile=/tmp/check_netstat_out_$$
infofile=/tmp/check_netstat_out_2_$$
myHost=$( hostname -f )
-echo $myHost | cut -f 3- -d "." | grep "\." >/dev/null
-if [ $? -ne 0 ]; then
- ph.setStatus unknown
- ph.status "DNS check for [$myHost] - SKIP: hostname -f returned a FQDN with less than 2 dots"
- ph.exit
-fi
-
# set default / override from command line params
-typeset -i iWarnLimit=$( ph.getValueWithParam 300 w "$@")
-typeset -i iCriticalLimit=$( ph.getValueWithParam 1000 c "$@")
+typeset -i iWarnLimit; iWarnLimit=$( ph.getValueWithParam 300 w "$@")
+typeset -i iCriticalLimit; iCriticalLimit=$( ph.getValueWithParam 1000 c "$@")
rm -f $tmpfile $infofile 2>/dev/null
@@ -38,23 +36,67 @@ typeset -i iTime=0
typeset -i iCounter=0
typeset -i iNotReachable=0
+# show help
+function showHelp(){
+ _self=$(basename $0)
+cat <<EOF
+______________________________________________________________________
+
+$self_APPNAME
+v$self_APPVERSION
+
+(c) Institute for Medical Education - University of Bern
+Licence: GNU GPL 3
+______________________________________________________________________
+
+Loop over all defined dns servers (in /etc/resolv.conf) and check
+each:
+ - detect of port 53 is available (DNS service)
+ - 5x check time of a nslookup for current host (${myHost})
+
+The a warning / critical response will be reached if the maximum time
+of nslookup on the first dns server reaches a limit.
+
+The critical response will be returned if the first of the
+nameservers is not available.
+
+An unknown response will be returned if the current hostname
+(hostname -f) is no FQDN.
+
+SYNTAX:
+$_self [ -w VALUE -c VALUE -h ]
+
+OPTIONS:
+
+ -w VALUE warning level (default: 300)
+ -c VALUE critical level (default: 1000)
+ -h or --help Show this help.
+
+EOF
+}
+
# ----------------------------------------------------------------------
# MAIN
# ----------------------------------------------------------------------
# --- check param -h
-if [ "$1" = "-h" ]; then
- echo "
- usage: $0 [ -w value -c value -h ]
-
- -w Warning level
- -c Critical level
- -h this help
- "
- exit 0
+case "$1" in
+ "--help"|"-h")
+ showHelp
+ exit 0
+ ;;
+ *)
+esac
+
+if ! echo $myHost | cut -f 3- -d "." | grep "\." >/dev/null; then
+ ph.setStatus unknown
+ ph.status "DNS check for [$myHost] - SKIP: hostname -f returned a FQDN with less than 2 dots"
+ ph.exit
fi
+ph.require nslookup
+
for mydns in $(grep ^nameserver /etc/resolv.conf | awk '{ print $2 } ' )
do
iCounter=$iCounter+1
@@ -63,13 +105,12 @@ do
echo "---------- $iCounter - $mydns " >>$infofile
# todo loop
- >/dev/tcp/${mydns}/53 >>$infofile 2>&1
- if [ $? -ne 0 ]; then
+ if ! >/dev/tcp/${mydns}/53 >>$infofile 2>&1; then
iNotReachable=$iNotReachable+1
test $iCounter -eq 1 && ph.setStatus critical
echo "ERROR: ${mydns} is not reachable on tcp 53" >>$infofile
else
- for i in `seq 5`
+ for i in $(seq 5)
do
(time nslookup ${myHost} $mydns) >$tmpfile 2>&1
@@ -84,7 +125,7 @@ do
# --- set status
test $iCounter -eq 1 && ph.setStatusByLimit $iSrvMax $iWarnLimit $iCriticalLimit
- label=$(echo $mydns | sed "s#\.#-#g" )
+ label=${mydns//\./-}
ph.perfadd "response-$label" "${iSrvMax}"
test $iSrvMax -gt $iMax && iMax=$iSrvMax
fi
@@ -93,7 +134,8 @@ do
done
-ph.status "DNS check for $myHost - found maximum was $iMax ms - $iNotReachable of $iCounter nameservers not reachable"
+test $iNotReachable -eq 0 && ph.status "DNS check for $myHost - found maximum was $iMax ms - OK: all nameservers are reachable"
+test $iNotReachable -ne 0 && ph.status "DNS check for $myHost - found maximum was $iMax ms - INFO: $iNotReachable of $iCounter nameservers not reachable"
cat $infofile
rm -f $tmpfile $infofile
diff --git a/docs/20_Checks/_index.md b/docs/20_Checks/_index.md
index 4cb4b73..80d492c 100644
--- a/docs/20_Checks/_index.md
+++ b/docs/20_Checks/_index.md
@@ -18,7 +18,7 @@ There is one include script used by all checks:
* [check_cpu](check_cpu.md)
* [check_cronstatus](check_cronstatus.md)
* [check_disk-io](check_disk-io.md)
-* check_dns_responsetime
+* [check_dns_responsetime](check_dns_responsetime.md)
* [check_eol](check_eol.md)
* check_haproxy_health
* check_haproxy_status
diff --git a/docs/20_Checks/check_dns_responsetime.md b/docs/20_Checks/check_dns_responsetime.md
new file mode 100644
index 0000000..438566e
--- /dev/null
+++ b/docs/20_Checks/check_dns_responsetime.md
@@ -0,0 +1,80 @@
+# CHECK_DNS_RESPONSETIME
+
+## Introduction
+
+**check_dns_responsetime** check dns response time of all dns servers found in local /etc/resolv.conf
+
+## Requirements
+
+* `nslookup` query Internet name servers
+
+## Syntax
+
+```txt
+______________________________________________________________________
+
+CHECK_DNS_RESPONSETIME
+v1.3
+
+(c) Institute for Medical Education - University of Bern
+Licence: GNU GPL 3
+______________________________________________________________________
+
+Loop over all defined dns servers (in /etc/resolv.conf) and check
+each:
+ - detect of port 53 is available (DNS service)
+ - 5x check time of a nslookup for current host (www.example.com)
+
+The a warning / critical response will be reached if the maximum time
+of nslookup on the first dns server reaches a limit.
+
+The critical response will be returned if the first of the
+nameservers is not available.
+
+An unknown response will be returned if the current hostname
+(hostname -f) is no FQDN.
+
+SYNTAX:
+check_dns_responsetime [ -w VALUE -c VALUE -h ]
+
+OPTIONS:
+
+ -w VALUE warning level (default: 300)
+ -c VALUE critical level (default: 1000)
+ -h or --help Show this help.
+
+```
+
+### Parameters
+
+(none)
+
+## Examples
+
+`$ check_disk-io -m io` returns
+
+```txt
+./check_dns_responsetime
+OK: DNS check for www.example.com - found maximum was 46 ms - OK: all nameservers are reachable
+
+---------- 1 - 8.8.4.4
+8.8.4.4 #1 >>> 36 ms
+8.8.4.4 #2 >>> 28 ms
+8.8.4.4 #3 >>> 24 ms
+8.8.4.4 #4 >>> 30 ms
+8.8.4.4 #5 >>> 31 ms
+max: 36 ms
+ ^
+ :
+ +--- 1st nameserver is relevant for total status of the check. Limits are warning=300 and critical=1000
+
+
+---------- 2 - 8.8.8.8
+8.8.8.8 #1 >>> 39 ms
+8.8.8.8 #2 >>> 35 ms
+8.8.8.8 #3 >>> 33 ms
+8.8.8.8 #4 >>> 46 ms
+8.8.8.8 #5 >>> 43 ms
+max: 46 ms
+
+ |response-8-8-4-4=36;; response-8-8-8-8=46;;```
--
GitLab