ranl/monitor-utils

Monitoring scripts for Zabbix / Nagios

monitor-utils's Issues

check-roomalert reports wrong, when temperature below 10 degrees C

The current calculation from centidegrees(?) to degrees contains an error.

When temperature is below 10 degrees the raw value is only 3 digits long, so by allways taking the first 2 digits, the calculated value ends up as eg 99 degreess instead of 9.9 degrees

I've fixed locally by using this line instead
$internal = $internal / 100.0;

check-cisco.pl: snmp return status are not checked

For example if the remote host is not responding, I get:

PS: OK - 0 PS are running all good | total=0 err=0

For the fans checks, I get a misleading message:

fan: CRIT - No Fans are running 

I was expected a message like "No reponse from ...", as the message returned by snmpget.

I'm working on a fix.

check-netapp-ng: Not compabitle with NetApp 8.3 (8.x?)

Hi Ran,

Great plugin however unfortunately it doesn't work with NetApp 8.3.
I tried it and it doesn't find the OID's - maybe NetApp has made some drastic changes in the MIB?
NetApp release:

snmpwalk -v 2c -c public netappvip 1.3.6.1.4.1.789.1.1.2
iso.3.6.1.4.1.789.1.1.2.0 = STRING: "NetApp Release 8.3.2P1: Thu Apr 14 15:58:25 UTC 2016"

Some examples:

./check-netapp-ng.pl -H netappvip -V 2c -C public -T GLOBALSTATUS
CRIT: GLOBALSTATUS  noSuchObject noSuchObject | globalstatus=noSuchObject

./check-netapp-ng.pl -H netappvip -V 2c -C public -T FAN
OK: FAN noSuchObject | failedfans=noSuchObject

Would be great if you could make it compatible with the newer NetApp release. Maybe detect the version as a first step and according to the version (if version > 8) load different OID's?

add aggregate usage

Hesslo i would be nice to check alse the aggregate allocation.

check_netapp_ng.pl -T SNAPSHOTAGE ---- look for the youngest snapshot

Hi ,
I am using check_netapp_ng.pl -H xxx -T SNAPSHOTAGE -C public -V 1 – c 2 (it works fine)
and we have from each /vol/*** ca. five snapshot
The check gives me a CRITICAL if the oldest snapshot is older then 2 days.
But I want a CIRTICAL if the youngest snapshot is older then 24 hours
Can someone help me ?

missing examples/guide about interfaces mode

Hi,
I'm trying to use your check-cisco.pl to monitor a cisco router, almost all checks are clear and simple to me, great job! Anyway I don't know how to perform the "int" check, I don't know wich is the format of the parameter I should provide in -i and where I can find the right OID.
For examples if I give as OID .1.3.6.1.2.1.2.2.1.1.1 which should be the index of first interface I get :

./check-cisco.pl -H XXX.XXX.XXX.XXX -C <my community> -t int -o .1.3.6.1.2.1.2.2.1.1.1
SNMP error: No response from remote host 'XXX.XXX.XXX.XXX'

If I use the same OID using snmpwalk I get :

$ snmpwalk -v 1 XXX.XXX.XXX.XXX -c <my community> .1.3.6.1.2.1.2.2.1.1.1
IF-MIB::ifIndex.1 = INTEGER: 1

Can you please provide some examples ? It's not so clear to understand which is the right way to pass the parameter and which parameter I should pass

Thanks in advance
magowiz

Missing lib location for default Nagios path

I've got this error when installing this plugin:

(No output on stdout) stderr: Can't locate utils.pm in @INC (@INC contains: /usr/lib64/nagios/plugins /usr/lib64/nagios/libexec /usr/lib/nagios/plugins /usr/lib/nagios/libexec /usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 .) at /usr/local/nagios/libexec/check-netapp-ng.pl line 43.

Please add this use lib "/usr/local/nagios/libexec"; to the script to support default Nagios location when installed from source

UPTIME after reboot no function

Hello,

we have a FAS8200 and after a reboot the uptime check give following message.

**ePN /usr/local/icinga/libexec//check_netapp_ng.pl: plugin did not call exit()

The reason ist for the first day after a reboot the uptime output from snmp is

4 hours, 44:54.86

The check expected something like 4 days, 7:14:35:17.12

I have no experience with github so i do not know how i upload the fixed code.
Here my recommended code.

### UPTIME ###
} elsif("$opt{'check_type'}" eq "UPTIME") {
        my $check = _get_oid_value($snmp_session,$snmpUpTime);
        $msg = "$opt{'check_type'}: $check";
        if ($check =~ /hours/){
                $check =~ m/^\s*(\d+)\s+hours,\s+(\d+):(\d+).*$/;
                $perf = "uptime=" . ($1*3600 + $2*60 + $3) . "s";
        }else{
                $check =~ m/^\s*(\d+)\s+days,\s+(\d+):(\d+):(\d+).*$/;
                $perf = "uptime=" . ($1*86400 + $2*3600 + $3*60 + $4) . "s";
        }

check-netapp-ng.pl: timeout not handled properly

If the check timeouts, no useful message is provided, and also the exit code is not correct:

[root@nagios hardware]# ./check-netapp-ng.pl -H 192.168.0.10 -C public -T UPTIME
Alarm clock
[root@nagios hardware]# echo $?
142

The check should return 2 (critical), and provide an input like "no reponse from XXXX"

I'll try to fix this.

Warning and Critical parameters always required

They shouldn't be required in the firmware and model modes.

check-netapp-ng.pl -- SHELF subcommand too verbose

Hello,

I noticed a problem with the subcommand SHELF of check-netapp-ng.pl:
the output is split on multiple lines and this make Nagios display the wrong message (the first line instead of the mast one). The following patch fix this issue:

@@ -368,7 +368,6 @@                                                                                                                                                       
                $shelf{'VoltUnderWarn'} = _get_oid_value($snmp_session,"$snmpEnclTableVoltUnderWarn.$oid");                                                               

                foreach my $subkey ( keys %shelf) {                                                                                                                       
-                       print "$subkey\t$shelf{$subkey}\n";                                                                                                               
                        if($shelf{$subkey}) { push(@shelf_err,"$addr $subkey,") }                                                                                         
                }                                                                                                                                                         
                if($#shelf_err != -1) {                                                                                                                                   

Makes it sense to you?
Thanks

OK status on failure getting HA mode

The check times out and then exits with OK and status 0.

[root@hlv]~# time ./check-paloalto-A500.pl -H 123.123.123.123 -C communitystring -t ha -w asd -c asd
OK: High Availablity Mode: , Local: , Peer: 


real    0m30.086s
user    0m0.047s
sys 0m0.007s
[root@hlv]~# echo $?
0
[root@hlv]~# 

Should probably exit with UNKNOWN or CRITICAL.

Spelling Mistake in help for -d argument

-d = number of days that th ethernet interface hasn't change state, default is 14 (only for -t freeint)

"the" is missing the "e"