Re: [Nagios-users] check_ntp offsets
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 19/08/09 01:15 AM, Chris wrote:
On Wed, Aug 19, 2009 at 1:09 PM, Thomas Guyot-Sionnestderm...@aei.ca wrote:
On 18/08/09 10:13 PM, Onotsky, Steve x55328 wrote:
-Original Message-
From: Chris [mailto:atst...@gmail.com]
Sent: August 18, 2009 21:56
To: nagios-users ML
Subject: [Nagios-users] check_ntp offsets
I am running check_ntp plugin and I have setup an offset threshold of
-w 200 -c 300 but Nagios is still showing Offset -102.234564 secs
(Warning)
Running check_ntp from command line shows NTP OK
(/usr/lib/nagios/plugins/check_ntp -H hostname -w 200 -c 300)
Output: NTP OK: Offset -102.234564
secs|offset=-102.234564s;200.00;300.00;
Here is what I have added:
edit /etc/nagios/objetcs/command.cfg and add:
define command{
command_name check_ntp
command_line $USER1$/check_ntp -H $HOSTADDRESS$ -w $ARG1$ -c
$ARG2$
}
edit /etc/nagios/objetcs/server.cfg and add:
define service{
use critical-service
host_namehostname
service_description NTP
check_command check_ntp!200!300
}
According to your check definition, the result is absolutely correct.
You asked it to warn you if the offset is more than 200 seconds, and go
critical if over 300 seconds.
What you were probably hoping for were thresholds in milliseconds. In
that case, you'll want to set -w and -c to fractional values (-w 0.2 -c
0.3, or whatever values you actually want as thresholds).
The offset and thresholds are both in seconds:
$ ./check_ntp -H pool.ntp.org -w 0.001 -c 0.01
NTP WARNING: Offset 0.008824706078 secs |offset=0.008825s;0.001000;0.01;
I have tried all the suggestions but still a no go.
I still get offset=-105.293049s;60.00;120.00;
and Offset -105.293049 secs
According to that result above you're using the default thresholds
(60.00;120.00), meaning that you're not passing the arguments
correctly.
Verify your definitions (especially look for typos - you may have two
very-similar check_ntp commands defined), or use the echo trick to see
which what command you're actually running. Use:
command_line echo $USER1$/check_ntp -H $HOSTADDRESS$ -w $ARG1$ -c $ARG2$
This will skip the check and print the command that normally runs instead.
- --
Thomas
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFKi6sb6dZ+Kt5BchYRAsfIAKCosd5iB9nhDFXLZVZxls9DlWmHiwCgpQ+D
CsXIrBr63mx7axlJ5/+p2aY=
=PXRE
-END PGP SIGNATURE-
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
[Nagios-users] SSL Handshake errors
Hi all Before I went home last night all my servers were responding to Nagios, this morning when I get into work there is one server that is reporting Could not complete SSL handshake, I have asked around but it appears that nobody has made any changes to the server. How can this be working one minutes and then report SSL problems the next, do you think a change has been made somewhere if so could you point me to where the possible changes were made. Thanks all Martyn -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] organizing config files with templates
On Wed, Aug 19, 2009 at 3:31 AM, Terry td3...@gmail.com wrote:
Hello,
I have a medium-sized installation (10,000) services) in a manged
services/data center environment. I am curious how others in these
types of multi-tenant environments organize their configuration files.
There are multiple things to consider:
- templates
- cfg_dirs
- grouping
I am considering cfg_dirs in this fashion:
/etc/nagios/objects/hosts/customer.cfg
/etc/nagios/objects/services/customer/host.cfg
I am considering templates in this fashion:
/etc/nagios/objects/services/templates/sql.cfg (will contain
microsoft sql type services)
This template will look like this:
define service{
use standard-server
hostgroup_name sql
service_description microsoft sql server service
check_command check_nrpe_win_service_mssql
}
Then I can just add hosts to the sql hostgroup to give them this service.
What am I missing? What do others do that might make my life easier?
Sounds a lot like what I use... I also use host-templates for example:
define host{
use solaris,dns,an-other-app
address...
}
the solaris host-template applies standard checks like disk, cpu, SMF.
the solaris-10-global template applies checks such as zone-checks
the same can be applies to a windows / rhel template
our rhel and solaris templates actually use basic *nix templates too.
this was each hosts config is kept to a minimum. also keeping each host in
its own .cfg file helps eg .../customer1/server-name1.cfg then any host
specific checks (eg url checks) can be added to that one config file too.
Ritchie
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus
on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when
reporting any issue.
::: Messages without supporting info will risk being sent to /dev/null
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] nagios host dependencies (strange) issue--it does not works as expected
Hi Marc,
thank you for mail. I do not understand if I have in dependencies.cfg
define hostdependency{
host_name FIREWALL
dependent_host_name SERVER
notification_failure_criteria d,u
}
and nagios does not report any problem when restarted ( what means
conf. is correct )
and if I push FIREWALL host to be DOWN - and this state is correcly
stated via nagios
web interface, why host SERVER is still up and green. I expect if
FIREWALL is down
that means all hosts depndent on it should be represented different
that OK ( down, unrechable... )
In nagios.cfg I have enabled
enable_predictive_host_dependency_checks=1
enable_predictive_service_dependency_checks=1
The same I have when I set up ( commneting above hostdependency )
in dependencies.cfg
define servicedependency{
host_name FIREWALL
service_description PING
dependent_host_name SERVER
dependent_service_description Remote Desktop
execution_failure_criteria w,c,u
notification_failure_criteria c,w,u
}
and then disable ping to FIREWALL, Remote Desktop service on SERVER
is still UP, nothing is changed
Any comment is welcome
Thank you in advance,
Nice regards,
Arlytex
On Tue, Aug 18, 2009 at 8:42 PM, Marc Powellm...@ena.com wrote:
On Aug 18, 2009, at 1:25 PM, arly arly wrote:
2009/8/18 Marc Powell m...@ena.com
Did the *host* status change to critical or just the service status?
Host status changed, when I browse it over web host detail ---
host it is
represented as DOWN.
The 'firewall' host showed down, right? If that's the case then I
would expect the host dependency to suppress host notifications about
the 'host_behind_firewall'. I don't use dependencies though so there
may be some obscure criteria I don't remember. Again, I think using
the 'parents' directive is probably more appropriate for your stated
purposes.
--
Marc
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] organizing config files with templates
m...@catsnest.co.uk wrote:
On Wed, Aug 19, 2009 at 3:31 AM, Terry td3...@gmail.com
mailto:td3...@gmail.com wrote:
Hello,
I have a medium-sized installation (10,000) services) in a manged
services/data center environment. I am curious how others in these
types of multi-tenant environments organize their configuration files.
There are multiple things to consider:
- templates
- cfg_dirs
- grouping
I am considering cfg_dirs in this fashion:
/etc/nagios/objects/hosts/customer.cfg
/etc/nagios/objects/services/customer/host.cfg
I am considering templates in this fashion:
/etc/nagios/objects/services/templates/sql.cfg (will contain
microsoft sql type services)
This template will look like this:
define service{
use standard-server
hostgroup_name sql
service_description microsoft sql server service
check_command check_nrpe_win_service_mssql
}
Then I can just add hosts to the sql hostgroup to give them this
service.
What am I missing? What do others do that might make my life easier?
Sounds a lot like what I use... I also use host-templates for example:
define host{
use solaris,dns,an-other-app
address...
}
the solaris host-template applies standard checks like disk, cpu, SMF.
the solaris-10-global template applies checks such as zone-checks
Hmm, could you eventually explain how the solaris host-template applies
standard checks (which seem to me like services) ? How do you do this ?
Regards,
the same can be applies to a windows / rhel template
our rhel and solaris templates actually use basic *nix templates too.
this was each hosts config is kept to a minimum. also keeping each host
in its own .cfg file helps eg .../customer1/server-name1.cfg then any
host specific checks (eg url checks) can be added to that one config
file too.
Ritchie
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008
30-Day
trial. Simplify your report design, integration and deployment - and
focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
mailto:Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when
reporting any issue.
::: Messages without supporting info will risk being sent to /dev/null
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
--
DI Simon Kainz
Graz, University of Technology
Department Computing
Phone: ++43 (0) 316 / 873 6885
signature.asc
Description: OpenPGP digital signature
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
[Nagios-users] Monitoring linux quotas
Hello all. I'm considering setting up quotas on our RHEL-servers, and were wondering if there are any Nagios plugins that can monitor quotas. Btw, we're running Nagios 3.0.3. Regards, Kenneth Holter -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
[Nagios-users] high load on nagios server due to status.cgi
Hello, *Problem:* We are facing the problem while browsing the servers from Nagios web portal. Rendering of the server details is too slow that we need to wait around 2 min to get details about the server. We are continuously seeing that server load is always high and it is due to processing of status.cgi and extinfo.cgi Here is the o/p of top PID USER PR NI %CPUTIME+ %MEM VIRT RES SHR S COMMAND *14857 apache25 0 88 0:12.46 0.8 35588 30m 16m R status.cgi* In status.cgi there are no of entries for service comments,acknowledgement associated with each host and services. Even if the problem associated with the service is resolved that service comment is still in status.cgi. Due to which size of the status.cgi file is keep on increasing and which in turn resulting this high time rendering issue and also huge load on server. *Please suggest how to resolve this problem.* * * Our technical details: We are using nagios v 3.0.6 on red hat Linux 2.6.9-67.ELsmp to monitor our all servers like (windows/Linux/Solaris) Total 905 hosts and 7985 services are being monitored by our server. Technical spec of server: Processor:2* AMD Opteron(tm) Processor 254 with 2.8 GHZ cpu speed RAM:4 GB Please let me know in case of any more information required. Rakesh Kakde -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] Servicegroups with stats
2009/8/18 Alex Huth a.h...@tmr.net: Hi! I am really new to Nagios and read the doc and tested some things. Now i want to have all webservices (45) grouped together. Not so difficult, but i want to see also the stats in this servicegroup. The main reason is to have a quick overview over the performance of the webservices. I don't want to scroll every time in the servicedetails to the webserver to see them. Thanks Alex An excellent way to make graphs of the performance stats is to use pnp4nagios http://www.pnp4nagios.org/pnp/start Once you have created some graphs in pnp4nagios, you can group graphs together in to one either by using pnp's own special templates ( http://www.pnp4nagios.org/pnp/tpl_special ) or you can use the tool drraw ( http://web.taranis.org/drraw/ ) to create graphs from multiple sources and web dashboards showing multiple graphs on one page. I confess I haven't tried pnp's special templates for combining graphs myself yet, but I do use drraw all the time for bringing graphs from various sources together in to one graph or one dashboard. The version I use is a pnp-specific fork of drraw at http://github.com/perldork/drraw-pnp/tree/master . hth, Jim -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] Monitoring linux quotas
2009/8/19 Kenneth Holter kenneho@gmail.com: Hello all. I'm considering setting up quotas on our RHEL-servers, and were wondering if there are any Nagios plugins that can monitor quotas. Btw, we're running Nagios 3.0.3. There's a plugin check_quota on Nagios Exchange http://exchange.nagios.org which reports on the output from the command repquota -a -s if that's what you're after. If you give it a try, please post a reivew on Nagios Exchange to help anyone else who might find it useful. Unfortunately, the author says almost nothing about it on NagiosExchange so it's difficult to see at first glance whether it's relevant or not. Cheers, Jim -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] high load on nagios server due to status.cgi
2009/8/19 rakesh kakde rakesh.ka...@gmail.com: Problem: We are facing the problem while browsing the servers from Nagios web portal. Rendering of the server details is too slow that we need to wait around 2 min to get details about the server. We are continuously seeing that server load is always high and it is due to processing of status.cgi and extinfo.cgi In status.cgi there are no of entries for service comments,acknowledgement associated with each host and services. Even if the problem associated with the service is resolved that service comment is still in status.cgi. Due to which size of the status.cgi file is keep on increasing and which in turn resulting this high time rendering issue and also huge load on server. Make sure you read http://nagios.sourceforge.net/docs/3_0/tuning.html , but to be honest, those instructions are more to do with performance of the Nagios scheduler, not the .cgis. Some screens do take a long time to render on a busy server. I have removed one or two of them from /usr/local/nagios/share/side.html to prevent users from displaying them and using up excessive CPU. If you have configured the Nagios EXtended Status Map (NEXSM) you will find that while that is running it puts quite a heavy load on the web server. You might want to consider increasing the value refresh_rate in cgi.cfg to reduce the frequency that the cgi will automatically refresh. I don't know if any of these suggestions will entirely fix your problem, but hopefully I've given you a few ideas. Cheers, Jim -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] check_ntp offsets
On Aug 18, 2009, at 8:55 PM, Chris wrote: I am running check_ntp plugin and I have setup an offset threshold of -w 200 -c 300 but Nagios is still showing Offset -102.234564 secs (Warning) If your configured thresholds used to be lower, verify that you don't have multiple nagios daemons running at the same time. -- Marc -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] nagios host dependencies (strange) issue--it does not works as expected
On Aug 19, 2009, at 4:46 AM, Arlen Drina wrote:
Hi Marc,
thank you for mail. I do not understand if I have in dependencies.cfg
define hostdependency{
host_name FIREWALL
dependent_host_name SERVER
notification_failure_criteria d,u
}
and nagios does not report any problem when restarted ( what means
conf. is correct )
and if I push FIREWALL host to be DOWN - and this state is correcly
stated via nagios
web interface, why host SERVER is still up and green. I expect if
FIREWALL is down
that means all hosts depndent on it should be represented different
that OK ( down, unrechable... )
This is a misconception. This will only be the case if the
check_command for those dependent hosts says they are non-OK. The
status of a dependent host will not change because of a dependency
definition. If the check_command for the dependent host still returns
OK, then that's it's status. Dependencies only suppress notifications
or stop checks entirely; they never change the status of a host or
service. Why does the check_command for 'SERVER' not properly reflect
the status of 'SERVER' when 'FIREWALL' is down? It seems to me that's
what you should be trying to determine. I assume you're pinging it.
Why is it still pingable from the nagios host if the firewall is down?
--
Marc
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] high load on nagios server due to status.cgi
On Aug 19, 2009, at 6:24 AM, rakesh kakde wrote: Problem: We are facing the problem while browsing the servers from Nagios web portal. Rendering of the server details is too slow that we need to wait around 2 min to get details about the server. I'm not sure what you've done differently but my experience is that this isn't normal. I've not done anything special to my install (other than use large install tweaks setting), running on comparable hardware to you with 8683 services and none of the status.cgi screens take longer than 15 seconds to pull up, even over the Internet. Those that do take the longest (show all services and hostgroup grid) are slow primarily due to the time it's taking the browser to render the tables... We are continuously seeing that server load is always high and it is due to processing of status.cgi and extinfo.cgi What specific status.cgi view is causing this? Here is the o/p of top PID USER PR NI %CPUTIME+ %MEM VIRT RES SHR S COMMAND 14857 apache25 0 88 0:12.46 0.8 35588 30m 16m R status.cgi Use strace to see what it's doing. Where does top show the utilization (i.e. is it IO wait? could you have slow disks or a problem disk?) In status.cgi there are no of entries for service comments,acknowledgement associated with each host and services. Even if the problem associated with the service is resolved that service comment is still in status.cgi. I'm not sure what you're saying here. There are no service comments but if the problem is resolved there are still no service comments? If you're talking about the comment you add when acknowledging, and mean to say that it does exist, this is expected if you've selected 'Persistent Comment'. Comments added outside of Acknowledgements hang around until you remove them. Due to which size of the status.cgi file is keep on increasing and which in turn resulting this high time rendering issue and also huge load on server. 12M status.dat and objects.cache here... -- Marc -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] organizing config files with templates
On Wed, Aug 19, 2009 at 4:52 AM, Simon Kainzsimon.ka...@tugraz.at wrote:
m...@catsnest.co.uk wrote:
On Wed, Aug 19, 2009 at 3:31 AM, Terry td3...@gmail.com
mailto:td3...@gmail.com wrote:
Hello,
I have a medium-sized installation (10,000) services) in a manged
services/data center environment. I am curious how others in these
types of multi-tenant environments organize their configuration files.
There are multiple things to consider:
- templates
- cfg_dirs
- grouping
I am considering cfg_dirs in this fashion:
/etc/nagios/objects/hosts/customer.cfg
/etc/nagios/objects/services/customer/host.cfg
I am considering templates in this fashion:
/etc/nagios/objects/services/templates/sql.cfg (will contain
microsoft sql type services)
This template will look like this:
define service{
use standard-server
hostgroup_name sql
service_description microsoft sql server service
check_command check_nrpe_win_service_mssql
}
Then I can just add hosts to the sql hostgroup to give them this
service.
What am I missing? What do others do that might make my life easier?
Sounds a lot like what I use... I also use host-templates for example:
define host{
use solaris,dns,an-other-app
address...
}
the solaris host-template applies standard checks like disk, cpu, SMF.
the solaris-10-global template applies checks such as zone-checks
Hmm, could you eventually explain how the solaris host-template applies
standard checks (which seem to me like services) ? How do you do this ?
Regards,
the same can be applies to a windows / rhel template
our rhel and solaris templates actually use basic *nix templates too.
this was each hosts config is kept to a minimum. also keeping each host
in its own .cfg file helps eg .../customer1/server-name1.cfg then any
host specific checks (eg url checks) can be added to that one config
file too.
Ritchie
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008
30-Day
trial. Simplify your report design, integration and deployment - and
focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
mailto:Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when
reporting any issue.
::: Messages without supporting info will risk being sent to /dev/null
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
--
DI Simon Kainz
Graz, University of Technology
Department Computing
Phone: ++43 (0) 316 / 873 6885
I think he is doing this:
define service{
use server-1-hour-retry-emailonly-noticket
hostgroup_name windows
service_description cpu
check_command check_nrpe_win_cpu!80!90
}
define service{
use server-servicehours
hostgroup_name windows
service_description commit charge
check_command check_nrpe_win_memory_page!90%!95%
}
define service{
use server
hostgroup_name windows
service_description uptime
servicegroups uptime
check_command check_nrpe_win_uptime!1h
retry_check_interval70
}
So, all hosts in the windows host_group will get the uptime,cpu, and
commit charge services.
--
Let Crystal Reports handle the reporting -
Re: [Nagios-users] check ping unusual issue
I think I have found the problem, Please see the following theread
http://article.gmane.org/gmane.network.nagios.plugins.devel/6738
On Wed, Aug 19, 2009 at 9:54 AM, shadih rahman shadhi...@gmail.com wrote:
I am using nagios-3.0.6 and nagios plugins-1.4.13
As suggested I have read the check_ping man page. However having critical
warning is not the cause of the problem. We chose to keep both warning
and critical same because we want all ping issues to become critical. Below
is an example like you suggested. The question is how can you have 75%
packet loss, when you are sending one packet. Please advise on this.
Thanks
check_ping -H HostA -t 45 -4 -w 200.0,30% -c 300.0,40% -p 1
PING CRITICAL - Packet loss = 75%, RTA = 2.17
ms|rta=2.168000ms;200.00;300.00;0.00 pl=75%;30;40;0
On Tue, Aug 18, 2009 at 8:37 PM, Morris, Patrick patrick.mor...@hp.comwrote:
On Tue, 18 Aug 2009, shadih rahman wrote:
All,
My check ping result is not making sense. I have the following in
commands.cfg
define command{
command_namecheck-host-alive
command_line$USER1$/check_ping -H $HOSTADDRESS$ -t 45 -4 -w
3000.0,100% -c 3000.0,100% -p 1
}
However I see the following output
Mon Aug 17 13:06:34 2009];HOST ALERT: HostA;UP;SOFT;2;PING OK - Packet
loss = 87%, RTA = 2.09 ms
[Mon Aug 17 13:09:24 2009];HOST ALERT: Host B;DOWN;SOFT;2;PING CRITICAL
- Packet loss = 94%, RTA = 4001.56 ms
[Mon Aug 17 13:14:44 2009];HOST ALERT: Host C;UP;SOFT;3;PING OK -
Packet loss = 93%, RTA = 1.15 ms
[Mon Aug 17 13:55:14 2009];HOST ALERT: Host D;UP;SOFT;2;PING OK -
Packet loss = 83%, RTA = 2.19 ms
How can I have 87 % packet loss when I am sending only on packet. Can
someone please explain this to me. Thanks in advance.
Which version of the plugins are you using, on which version of Nagios?
There have been bugs in some versions of the plugin that might cause
results like that, but there's not enough information here to know if
that's the case with your issue.
--
Cordially,
Shadhin Rahman
--
Cordially,
Shadhin Rahman
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] organizing config files with templates
Sounds a lot like what I use... I also use host-templates for example:
define host{
use solaris,dns,an-other-app
address...
}
the solaris host-template applies standard checks like disk, cpu, SMF.
the solaris-10-global template applies checks such as zone-checks
Hmm, could you eventually explain how the solaris host-template applies
standard checks (which seem to me like services) ? How do you do this ?
Yes checks = services
Sorry i should be more careful with my terminology
...
I think he is doing this:
define service{
use
server-1-hour-retry-emailonly-noticket
hostgroup_name windows
service_description cpu
check_command check_nrpe_win_cpu!80!90
}
define service{
use server-servicehours
hostgroup_name windows
service_description commit charge
check_command check_nrpe_win_memory_page!90%!95%
}
define service{
use server
hostgroup_name windows
service_description uptime
servicegroups uptime
check_command check_nrpe_win_uptime!1h
retry_check_interval70
}
So, all hosts in the windows host_group will get the uptime,cpu, and
commit charge services.
Thats right, linking the services to hostgroups
and just using a host-template to add the host to a host_group
eg, using the above example a (simplified) host-template would look like
define host{
name windows
hostgroup+windows
register 0
}
I actually then 'use' another host-template in the above, to set default
options such as check_command.
Ritchie
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
[Nagios-users] Connection refused or timed out
What should I look for when troubleshooting a Connection refused or timed out when checking via NRPE to a Linux machine? I'm having this problem on two targets. So far, I've installed NRPE on 5 machines, and 3 worked without a hitch, and these two are giving me headaches. Some info about the targets. 10.5.10.50 is my Nagios server, dfslit1 10.5.10.90 is the target client, dfslrdb01. The second problem machine produces identical output to this one, aside from a different CentOS version. Both (well, all 5) are running this version of NRPE: NRPE - Nagios Remote Plugin Executor Copyright (c) 1999-2008 Ethan Galstad (nag...@nagios.org) Version: 2.12 Client OS versions are: CentOS release 5.3 (Final) CentOS release 5.2 (Final) From /var/log/messages when I restart the NRPE: Aug 19 13:51:08 dfslrdb01 nrpe[13249]: Starting up daemon Aug 19 13:51:08 dfslrdb01 nrpe[13249]: Listening for connections on port 5666 Aug 19 13:51:08 dfslrdb01 nrpe[13249]: Allowing connections from: 127.0.0.1,10.5.10.50 Aug 19 13:51:53 dfslrdb01 nrpe[13249]: Caught SIGTERM - shutting down... Aug 19 13:51:53 dfslrdb01 nrpe[13249]: Cannot remove pidfile '/var/run/nrpe.pid' - check your privileges. Aug 19 13:51:53 dfslrdb01 nrpe[13249]: Daemon shutdown Aug 19 13:51:53 dfslrdb01 nrpe[13266]: Starting up daemon Aug 19 13:51:53 dfslrdb01 nrpe[13266]: Listening for connections on port 5666 Aug 19 13:51:53 dfslrdb01 nrpe[13266]: Allowing connections from: 127.0.0.1,10.5.10.50 (10.5.10.50 is my Nagios server) I look in /var/run/nrpe.pid and the file is gone when NRPE stops, and is re-created when it starts. When it's running, permissions are: [r...@dfslrdb01 ~]# ll /var/run/nrpe.pid -rw-r--r-- 1 root root 6 Aug 19 14:09 /var/run/nrpe.pid Checking to see if the port's open: [mi...@dfslrdb01 ~]$ netstat -l | grep 5666 tcp0 0 *:5666 *:* LISTEN Command definition from /etc/nagios/nrpe.cfg: command[check_disk]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /dev/sda1 Running the command on the target machine: [mi...@dfslrdb01 ~]$ sudo su - nagios -sh-3.2$ /usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /dev/sda1 DISK OK - free space: / 294574 MB (45% inode=99%);| /=358494MB;550890;619751;0;688613 Running it at the command line from the server; result is the same with or without -n: nag...@dfslit1:~$ /usr/lib/nagios/plugins/check_nrpe -H 10.5.10.90 -c check_disk Connection refused or timed out nag...@dfslit1:~$ tcptraceroute 10.5.10.90 5666 Selected device eth0, address 10.5.10.50, port 36013 for outgoing packets Tracing the path to 10.5.10.90 on TCP port 5666, 30 hops max 1 10.5.10.90 2.965 ms !C 0.841 ms !C 0.744 ms !C nag...@dfslit1:~$ telnet 10.5.10.90 5666 Trying 10.5.10.90... telnet: Unable to connect to remote host: No route to host I installed from Dag's repository: http://dag.wieers.com/rpm/FAQ.php#B [r...@dfslrdb01 ~]# iptables -L Chain INPUT (policy ACCEPT) target prot opt source destination RH-Firewall-1-INPUT all -- anywhere anywhere Chain FORWARD (policy ACCEPT) target prot opt source destination RH-Firewall-1-INPUT all -- anywhere anywhere Chain OUTPUT (policy ACCEPT) target prot opt source destination Chain RH-Firewall-1-INPUT (2 references) target prot opt source destination ACCEPT all -- anywhere anywhere ACCEPT icmp -- anywhere anywhereicmp any ACCEPT esp -- anywhere anywhere ACCEPT ah -- anywhere anywhere ACCEPT udp -- anywhere 224.0.0.251 udp dpt:mdns ACCEPT udp -- anywhere anywhereudp dpt:ipp ACCEPT tcp -- anywhere anywheretcp dpt:ipp ACCEPT all -- anywhere anywherestate RELATED,ESTABLISHED ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:smtp ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:nfs ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:ssh ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:https ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:http ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:sunrpc ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:terabase ACCEPT udp -- anywhere anywherestate NEW udp dpt:terabase ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:newoak ACCEPT udp -- anywhere anywherestate NEW udp dpt:newoak ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:pxc-spvr-ft ACCEPT udp -- anywhere anywherestate NEW udp dpt:pxc-spvr-ft ACCEPT tcp -- anywhere anywhere
[Nagios-users] check_mk - thoughts
Anyone using check_mk? I've been playing around with it today to see if it would be a good move away from NRPE. Good concept but I am having issues trying to introduce contact group granularity into the mix. I am probably missing something that can be put into main.mk. -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] check_mk - thoughts
On Wed, Aug 19, 2009 at 3:59 PM, Terrytd3...@gmail.com wrote: Anyone using check_mk? I've been playing around with it today to see if it would be a good move away from NRPE. Good concept but I am having issues trying to introduce contact group granularity into the mix. I am probably missing something that can be put into main.mk. Awesomefound my answer after I sent my email. You can implement this with host tags apparently: http://mathias-kettner.de/checkmk_hosttags.html -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] check_mk - thoughts
Terry wrote: Anyone using check_mk? I've been playing around with it today to see if it would be a good move away from NRPE. Good concept but I am having issues trying to introduce contact group granularity into the mix. I am probably missing something that can be put into main.mk. Hi, Terry. I just looked at it quickly. Do you see any great benefit to switch from NRPE to this? Briefly looking over the docs, it's seems a bit more complicated to setup and use compared to NRPE. What's your opinion on playing with it so far? Regards, Max -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
[Nagios-users] Initial Nagios Startup Notification Delay
Hi, I'm running Nagios 3.0.3 and am wondering if there is a way I can delay notifications out of Nagios by x minutes from initial application startup? If the service/host were still down after this initial startup non-notification period, I would expect to receive a notification though. It would be quite handy, as I send alerts to cell phones, and receiving a very large amount of TXT messages can be frustrating, especially in a meeting. Is there a function like this available? I haven't seen anything in the documentation that would suggest it is (although could have missed it). Regards, Andrew Bruce -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
[Nagios-users] external_command_buffer_slots??
Hello everyone, I joined the list just for this! A google search of external_command_buffer_slots doesn't provide much beyond the official documentation. Can someone please clarify for me what this directive is actually doing. Is this buffer internal to Nagios? The name seems unfortunate because it reminds me of the external command file which obviously can't be increased beyond 4096 without hacking fifo sizes in kernel source. What downside is there to setting this buffer to say.. 8192 or ? Can someone describe where in the chain of events check results reside in this buffer? Thanks Everyone! Ryan Bowlby -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] New perl module: Nagios::Plugin::WWW::Mechanize
Hi Tom,
If you add this subroutine to the end of your module, any methods in
WWW::Mechanize that do not exist as wrapped methods will have method
proxies created for them in your wrapper module IF they exist in
WWW::Mechanize so that users can use the full range of methods
WWW::Mechanize offers in their plugins without you having to move to
multiple inheritence OR having to manually write wrapper calls for
every method.
Once the method proxy is created, AUTOLOAD will not be called again
for that method as the proxy will then exist :p.
If the method does not exist in WWW::Mechanize, normal perl behavior
for undefined methods will be used and the proxy code will be skipped.
- Max
sub AUTOLOAD {
my @args = @_;
# Subroutine the user requested from us that does not exist in our namespace
my $wanted = $Nagios::Plugin::WWW::Mechanize::AUTOLOAD;
$wanted =~ s/Nagios::Plugin::WWW::Mechanize:://ms;
# We were not passed a reference to ourselves, so exit.
if (ref($args[0]) ne 'Nagios::Plugin::WWW::Mechanize') {
return;
}
my $self = shift @args;
my $mech = $self-mech();
# If WWW::Mechanize knows this method, we create a proxy for it.
if ($mech-can($wanted)) {
my $func = EOF;
*Nagios::Plugin::WWW::Mechanize::$wanted = sub {
my (\$self, \...@args) = \...@_;
return \$self-mech-$wanted(\...@args);
};
EOF
# New scope to limit effect of 'no warnings'
{
no warnings 'redefine';
eval $func;
$self-die(Can't autocreate WWW::Mechanize::$wanted wrapper: $@)
if $@;
# Call directly via mech this time; future calls will invoke
# the proxy method.
$mech-$wanted(@args);
}
}
}
sub AUTOLOAD {
# Subs we proxy to WWW::Mechanize
my @args = @_;
# Subroutine the user requested from us that does not exist in our namespace
my $wanted = $Nagios::Plugin::WWW::Mechanize::AUTOLOAD;
my $wanted = $Nagios::Plugin::WWW::Mechanize::AUTOLOAD;
$wanted =~ s/Nagios::Plugin::WWW::Mechanize:://ms;
# We were not passed a reference to ourselves, so exit.
if (ref($args[0]) ne 'Nagios::Plugin::WWW::Mechanize') {
return;
}
my $self = shift @args;
my $mech = $self-mech();
# If WWW::Mechanize knows this method, we create a proxy for it.
if ($mech-can($wanted)) {
my $func = EOF;
*Nagios::Plugin::WWW::Mechanize::$wanted = sub {
my (\$self, \...@args) = \...@_;
return \$self-mech-$wanted(\...@args);
};
EOF
# New scope to limit effect of 'no warnings'
{
no warnings 'redefine';
eval $func;
$self-die(Can't autocreate WWW::Mechanize::$wanted wrapper: $@)
if $@;
# Call directly via mech this time; future calls will invoke
# the proxy method.
$mech-$wanted(@args);
}
}
}
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
Re: [Nagios-users] New perl module: Nagios::Plugin::WWW::Mechanize
I am using this in a plugin I am writing, example code:
$plugin-form_with_fields(qw(user passwd));
print R: . $plugin-field('r') . \n;
print S: . $plugin-field('s') . \n;
exit;
The field method is created as a proxy the first time it is called so
I can access it without having to write proxies in your module.
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Nagios-users mailing list
Nagios-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting
any issue.
::: Messages without supporting info will risk being sent to /dev/null
