check_icmp: keep performance data order in case of none-reachable hosts

For responding hosts nothing changes:

    ./check_icmp -H 10.0.0.1
    OK - 10.0.0.1 rta 0.336ms lost 0%|rta=0.336ms;200.000;500.000;0; rtmax=0.471ms;;;; rtmin=0.000ms;;;; pl=0%;40;80;0;100

In case the host is down, the performance data now contain 'U' according to:
https://www.monitoring-plugins.org/doc/guidelines.html#AEN201

Old:
    ./check_icmp -H 10.0.0.99
    CRITICAL - 10.0.0.99: Host unreachable @ 10.0.1.138. rta nan, lost 100%|pl=100%;40;80;0;100

Old:
    ./check_icmp -H 10.0.0.99
    CRITICAL - 10.0.0.99: Host unreachable @ 10.0.1.138. rta nan, lost 100%|rta=U;;;; rtmax=U;;;; rtmin=U;;;; pl=100%;40;80;0;100

This confuses some performance graphing tools, ex.: pnp.

Signed-off-by: Sven Nierlein <sven@consol.de>

1 files changed, 3 insertions, 2 deletions

diff --git a/plugins-root/t/check_icmp.t b/plugins-root/t/check_icmp.t
index 4f9db86..de1d88d 100644
--- a/plugins-root/t/check_icmp.t
+++ b/plugins-root/t/check_icmp.t
@@ -12,7 +12,7 @@ my $allow_sudo = getTestParameter( "NP_ALLOW_SUDO",
         "no" );
 
 if ($allow_sudo eq "yes" or $> == 0) {
-        plan tests => 39;
+        plan tests => 40;
 } else {
         plan skip_all => "Need sudo to test check_icmp";
 }
@@ -57,7 +57,8 @@ $res = NPTest->testCmd(
         "$sudo ./check_icmp -H $host_nonresponsive -w 10000ms,100% -c 10000ms,100% -t 2"
         );
 is( $res->return_code, 2, "Timeout - host nonresponsive" );
-like( $res->output, '/100%/', "Error contains '100%' string (for 100% packet loss)" );
+like( $res->output, '/pl=100%/', "Error contains 'pl=100%' string (for 100% packet loss)" );
+like( $res->output, '/rta=U/', "Error contains 'rta=U' string" );
 
 $res = NPTest->testCmd(
         "$sudo ./check_icmp -w 10000ms,100% -c 10000ms,100%"