perfSONAR User Q&A and Other Discussion

["Mike Cruzz" <>]

Hi Ivan

I have been checking this further and I have a theory of why the alerts don't
come through.

At the point when I run the nagios check_ping_loss.pl script, when I
disconnect the VM nic, I get the following output.

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS UNKNOWN - Unable to find any tests with data in the given
time range where source is 2.2.2.2 and destination is 3.3.3.3

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS UNKNOWN - Unable to find any tests with data in the given
time range where source is 2.2.2.2 and destination is 3.3.3.3

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS UNKNOWN - Unable to find any tests with data in the given
time range where source is 2.2.2.2 and destination is 3.3.3.3


This of course is accurate as there is no data in the esmond archive.

When I re-enable the nic the data starts to populate.


[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS OK - Average loss is 0.00% | Count=1;; Min=0;; Max=0;;
Average=0;; Standard_Deviation=0;;

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS OK - Average loss is 0.00% | Count=1;; Min=0;; Max=0;;
Average=0;; Standard_Deviation=0;;

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS OK - Average loss is 0.00% | Count=1;; Min=0;; Max=0;;
Average=0;; Standard_Deviation=0;;

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS OK - Average loss is 0.00% | Count=1;; Min=0;; Max=0;;
Average=0;; Standard_Deviation=0;;

[root@perf7 ~]# /usr/lib64/nagios/plugins/check_ping_loss.pl -u
https://1.1.1.1/esmond/perfsonar/archive -s 2.2.2.2 -d 3.3.3.3 -r 100 -c 0.1 -
w 0.001 -t 60
Subroutine JSON::PP::Boolean::("" redefined at /usr/share/perl5/overload.pm
line 49.
Subroutine JSON::PP::Boolean::(eq redefined at /usr/share/perl5/overload.pm
line 49.
PS_CHECK_PING_LOSS OK - Average loss is 0.00% | Count=1;; Min=0;; Max=0;;
Average=0;; Standard_Deviation=0;;

I guess I will need to use an additional script as a filter where if it sees
"PS_CHECK_PING_LOSS UNKNOWN" for a 60 second interval then flag it as ping
loss 100%.

I don't suppose you can share any other elegant way to do this?

Thanks again.