check_snmp CPU Load strange result
Pascal Wessel
pascal.wessel at media-online.ch
Tue Dec 3 10:18:20 CET 2002
You asked: Please post the version of the plugin/os/net-snmp
Hummm... Ok, I say it again: (I copy-past the needed info from my first
mail. Btw it's at the very end of this thread I agree)
---snip---
> My Nagios system installation is as follows:
>
> System Intel i686, Mandrake 9.0, Kernel 2.4.19-16
> NAGIOS: Nagios 1.0b6
> Plugins: nagios-plugins-200211131100
> Check_snmp: Revision: 1.17
> SNMP:
> libsnmp0-4.2.3-4mdk
> ucd-snmp-4.2.3-4mdk
> ucd-snmp-utils-4.2.3-4mdk
---snip---
I had this:
[libexec]# ldd check_snmp
libutil.so.1 => /lib/libutil.so.1 (0x40023000)
libc.so.6 => /lib/i686/libc.so.6 (0x40026000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
check_snmp treats bare numbers in warning and critical as upper bounds,
so
-w'60,69' is interpreted as warn on 0-60 for the first and 0-69 for the
second...
Very interesting... I did't catch it at first.
I made some tests with different values for warning and/or critical
thresholds... Very stange, I only have WARNING status returning from
these ones. To clarify a bit I made tests with only one OID (The 1 min
CPU load), with same result: WARNING Always returns.
By reading check_snmp --help I saw that I can use absolute values (like
-w '0:30') if colons are used.
Ranges are inclusive and are indicated with colons. When specified as
'min:max' a STATE_OK will be returned if the result is within the
indicated
range or is equal to the upper or lower bound. A non-OK state will be
returned if the result is outside the specified range.
Then....some tests again....
Eg (with the 1 min CPU load OID):
./check_snmp -v -t 10 -H 192.168.1.1 -o .1.3.6.1.4.1.9.2.1.57.0 -C
publicro -w '50:69' -c'70:99'
enterprises.9.2.1.57.0 = 4
SNMP WARNING - *4*
That's ok... 4% is not in the range 50:69... Then it's a "non-OK"
result, thus the WARNING for the -w switch.
./check_snmp -v -t 10 -H 192.168.1.1 -o .1.3.6.1.4.1.9.2.1.57.0 -C
publicro -w '1:9' -c'10:99'
enterprises.9.2.1.57.0 = 7
SNMP WARNING - *7*
7% is within the range 1:9... I should have a STATE_OK.
GrrrRRRrrrRRRRRR %#*$& !!! Why do I have WARNING ?????
Well, any help appreciated. Thanks in advance.
Warm regards,
Pascal
-sg
On Mon, 2 Dec 2002, Pascal Wessel wrote:
> Nagios gives me warning when snmp_check 'ing for Cisco 3640 CPU load /
> IOS is (C3640-IK9O3S-M), Version 12.2(10a) but the CPU load is below
> my Warning threshold.
>
> When launched from the command-line with verbose output:
>
> [libexec]# ./check_snmp -v -t 10 -H 192.168.1.1 -o
> .1.3.6.1.4.1.9.2.1.57.0,.1.3.6.1.4.1.9.2.1.58.0 -C publicro -w
> '60,69', -c '70,80' -l 'CPU usage 1min/5min' -D ' / '
> /usr/bin/snmpget -m ALL -v 1 -c publicro 192.168.1.1:161
> .1.3.6.1.4.1.9.2.1.57.0 .1.3.6.1.4.1.9.2.1.58.0
> enterprises.9.2.1.57.0 = 4
> enterprises.9.2.1.58.0 = 3
>
> CPU usage 1min/5min WARNING - *4* / *3*
>
> As you can see.. (and if I understood the syntax)
> Warning status should be triggered when the CPU load is between 60 and
> 69% Critical status should be triggered when the router CPU is between
> 70 to 80%
>
> #----
> My question is: why this check reports WARNING as my router CPU load
> (4% last minute and 3% last 5 min) is below the WARNING threshold ?
> #----
>
> My Nagios system installation is as follows:
>
> System Intel i686, Mandrake 9.0, Kernel 2.4.19-16
> NAGIOS: Nagios 1.0b6
> Plugins: nagios-plugins-200211131100
> Check_snmp: Revision: 1.17
> SNMP:
> libsnmp0-4.2.3-4mdk
> ucd-snmp-4.2.3-4mdk
> ucd-snmp-utils-4.2.3-4mdk
>
> Below a snip of my "cfg file
>
> #--- hosts.cfg for myrouter
>
> define host {
> name generic-host
> notifications_enabled 1 ; Host
notifications
> are enabled
> event_handler_enabled 1 ; Host event
handler is
> enabled
> flap_detection_enabled 1 ; Flap
detection is
> enabled
> process_perf_data 1 ; Process
performance
> data
> retain_status_information 1 ; Retain status
> information across program restarts
> retain_nonstatus_information 1 ; Retain
non-status
> information across program restarts
> max_check_attempts 10
> register 0 ; DONT REGISTER THIS
> DEFINITION - ITS NOT A REAL HOST, JUST A TEMPLATE!
> }
>
> define host {
> use generic-host
; Name
> of host template to use
> host_name myrouter
> alias Router Gva Coulou -6
> address 192.168.1.1
> check_command check-host-alive
> notification_interval 60
> notification_period 24x7
> notification_options d,u,r
> }
>
> #--- services.cfg
> define service {
> name generic-service ;
> active_checks_enabled 1 ; Active service checks
are
> enabled
> passive_checks_enabled 1 ; Passive service
checks are
> enabled/accepted
> parallelize_check 1 ; Active service checks
should
> be parallelized
> obsess_over_service 1 ; We should obsess over
this
> service (if necessary)
> check_freshness 0 ; Default is to NOT
check
> service 'freshness'
> notifications_enabled 1 ; Service notifications
are
> enabled
> event_handler_enabled 1 ; Service event handler
is
> enabled
> flap_detection_enabled 1 ; Flap detection is
enabled
> process_perf_data 1 ; Process performance
data
> retain_status_information 1 ; Retain status
information
> across program restarts
> retain_nonstatus_information 1 ; Retain non-status
information
> across program restarts
> normal_check_interval 5
> retry_check_interval 2
> notification_period 24x7
> notification_options u,c,r
> register 0 ; DONT REGISTER
THIS
> DEFINITION
> }
>
> define service{
> use generic-service
> host_name myrouter
> service_description CPU
> is_volatile 0
> check_period 24x7
> max_check_attempts 3
> retry_check_interval 1
> contact_groups router-admins
> notification_interval 120
> notification_period 24x7
> check_command
> check_cisco_cpu!publicro!60!69!70!80
> }
>
>
>
> #--- checkcommands.cfg
> # 'check_snmp' generic command definition
> define command{
> command_name check_snmp
> command_line $USER1$/check_snmp -t 10 -H $HOSTADDRESS$ -C $ARG1$
> $ARG2$ $ARG3$ $ARG4$ $ARG5$ $ARG6$ $ARG7$ $ARG8$ $ARG9$
> }
> # check_cisco_cpu: checks router CPU-usage
> # Syntax
> !Hostname!Community!WARN-1min-%!WARN-5min-%!CRIT-1min-%!CRIT-5min-%
> define command{
> command_name check_cisco_cpu
> command_line $USER1$/check_snmp -t 10 -H $HOSTADDRESS$
> -o.1.3.6.1.4.1.9.2.1.57.0,.1.3.6.1.4.1.9.2.1.58.0 -C $ARG1$ -w
> :$ARG2$,:$ARG3$ -c : $ARG4$,:$ARG5$ -l 'CPU usage 1min/5min' -D ' / '
> }
>
>
>
> Btw, by looking at the code in check_snmp.c I'm wondering . Is there a
> problem with : #define mark(a) ((a)!=0?"*":"") in check_snmp.c ??? Or
> are my parms so bad ? :-o
>
> Thanks for your kind help.
> Warm regards,
> Pascal
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Nagios-users mailing list
> Nagios-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-users
>
--
-------------------------------------------------------
This SF.net email is sponsored by: Get the new Palm Tungsten T
handheld. Power & Color in a compact size!
http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0002en
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
-------------------------------------------------------
This SF.net email is sponsored by: Get the new Palm Tungsten T
handheld. Power & Color in a compact size!
http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0002en
More information about the Users
mailing list