As indicated, make a plugin that gets the info and set thresholds
so high they're never likely to ring the red bell.
But in addition to this, I'd set it up with something like nagiosgraph
to generate graphs that you can watch. This will save/show historical
data and show you the norm. Thus you
Actually, that raises an interesting tangential question.
Can/should your nagios server handle mail delivery to
many many people? Unless it's tuned for higher capacity
email delivery, I'd expect SMTP functions to negatively
impact nagios checks.
In other words, why not just use a single address
One way might be to use the check_http plug-in but
have it look for a specific string in each web-page being hosted
and use the warn/critical-time returns for empirical threshold
On Thu, Jul 21, 2011 at 9:02 AM, af.at.w...@gmail.com wrote:
Hey gang, I am having some difficulty getting a single
Nagios is a phenomenal tool but isn't necessarily a silver bullet.
Nagios does try a certain number of times before a soft failure
becomes a hard failure. This can be adjusted. Also, many people
have replaced a very tight check_ping with check_icmp and looser
return times.
But in general, you
To get all checks to run in a shorter amount of time, take a look at reducing
max_service_check_spread and max_host_check_spread.
You might also be interested in the fast startup options in the documentation.
This will also point you to such parameters as use_large_installation_tweaks
On
We have an interesting need. When a particular service goes red on our
Nagios 3.2.1 server, we'd like to be able to click on Acknowledge this
service problem and have that activate a local script. Anyone have any
idea how this can be accomplished?
TIA
I don't know of any plug-in but you can roll your own.
If you're already running samba on a linux box, you can mess with
the output of net status shares
assuming users pick up (mount) their homes when they login to Windows.
I believe this also shows login times
For a more generic solution
Write a plugin. It could search the output of the command, mount
On Tue, Feb 8, 2011 at 6:49 AM, Javier Rivera jriv...@skisa.com wrote:
Good Morning,
Im having a problem with check_nrpe. Im monitoring a partition /mnt/2
f.e. If i dont have this partition mounted, it just returns the
I'd try to write a wrapper for an event handler that checked
the time and kept a log. A morning cron could then check
for presence of the log and if it exists, email to you
On Mon, Jan 31, 2011 at 9:46 AM, Tntteam tntt...@free.fr wrote:
Hello there,
I'm wondering if there is a way to make
Wrapper?
$NAGIOS/libexec/check_tcp -H myhost -p 135
$NAGIOS/libexec/check_tcp -H myhost -p 445
On Wed, Dec 29, 2010 at 1:39 PM, stan st...@panix.com wrote:
I think I need to verify that both port 135, and 445 are avaialble on some
Windows amchines. As I understand it, both of these need to
pitfalls in
my proposed configuration method.
On 9/16/08, dave stern - e-mail.pluribus.unum dit.d...@gmail.com wrote:
Not quite sure what you're trying to do. If you monitor a service,
it's going to be on the nagios
service details pagre regardless, That's the whole point. There are
some
Yes
http://nagiosuser:nagiospassw...@nagioshost.domain
/nagios/cgi-bin/status.cgi\?host=server...
Note the backslash before the question mark.
BTW, I think IE blocks this for what MicroSloth calls security reasons
On Thu, Oct 7, 2010 at 1:48 AM, Steve Kemp
Sounds like a postfix error. You DID configure postfix mail, right?
Take a look at http://www.postfix.org/BASIC_CONFIGURATION_README.html
This isn't a nagios question
On Mon, Aug 16, 2010 at 3:10 PM, i...@toonz it.to...@gmail.com wrote:
Hi all,
We are testing with Fully Automated Nagios.
We've had a lot of luck with check_http with the params -u and -s
-u points to the particular path on the web server and -s is the
return string it expects.
And then there's the hacky way of doing it; wget the page and
search for presence/absence of a particular string
On Fri, Aug 6, 2010 at
As pointed out, normal depends on your environment. Some sites only
check a handful
of things per host; some check more. The bigger issue is if your
server can handle
the number of checks and in a timely fashion. Look at nagios parameters
Max_service_check_spread,
max_host_check_spread
as well as
I've got NSClient++ installed on a number of Windows XP and
W2K servers reporting on such things as disk utilization, memory,
load. This all works fine. I'd like to take it to the next level defining
my own external scripts. But the examples all show using the
command, check_nrpe on the
We're interested in monitoring radius with nagios 3.2.1. We have IAS installed
on two windows boxes providing separate radius aaa authentication. I tried the
check_radius.pl plugin as well as another compilable one but they don't appear
to work despite providing them with user, password and
This may be stretching nagios beyond its designed intention but...
I'm writing a plugin that could produce many many lines of output.
Since this obviously wont fit on a status page, what I'd like to do
is if the output is not exit-status=0, produce a webpage and in the
status information field of
I'd like to specify an additional string within the notification message
for just a few hosts that will be sent to the contactgroup for those
hosts and only those hosts. The string will be static. I presume this
can be done with macros somehow?
TIA
We're looking for a mechanism to allow others access to our nagios
server for viewing only. The kicker is they want to be able to reach it
without any htpasswd so multiple htpasswords and
authorization_for_system_commands is not an option.
I'm thinking the only practical method to accomplish this
It seems that all options will run up against the fact that we have
enabled web authentication
and are not willing to turn it off. I've come up with a couple of ideas however.
1/ create a htpasswd account called guest with no password. At least
that limits the
necessary typing/authentication
I'm trying to check for the size of an nfs share that is accessible
from a windows host.
The windows host is running NSClient++ but there are a number of problems here.
First, the share isn't persistently mapped so at a minimum, I'd have
to write a .bat
warpper that first mapped it to a letter
Ok, everyone agrees event handler can take action to fix a problem but bear in
mind that this comes with caveats. Affectively, nagios event handler is treating
a symptom; the disease goes merely on its way. If a service stops, WHY did
it stop in the first place? Most good sysadmins would tackle
Anyone know of a plug-in or mechanism to log local disk I/O on windows?
My nagios server is currently using check_nt to connect to windows hosts
via nsclient++. I was hoping perhaps COUNTER has something buried
within it to pull down this info.
TIA
I'm looking for a way to see if there are any unacknowledged host/service
problems within nagios.
I've written up my own gosh-awful ugly code that parses nagios
logs to search for hosts that went down, haven't come back up, are
still in the nagios config and have not been acknowledged. Although
We've had great success with devices from avtech eg
http://avtech.com/LP/LP_RMA_1.htm
Added benefit; they also sell industrial grade light bars
(sorta like traffic lights) that can be triggered by nagios
to show overall status
I'm looking for a plugin that I can run from the unix side to connect to a
windows host and at least test for the existance of a file, although checking
for its contents would be even more ideal. I couldn't find anything at
nagiosexchange unless my search string wasn't specific enough.
Anyone?
We have a primary nagios install that monitors hosts/services spread
across multiple
labs (political domains). In one of the labs, I've build a nagios
install monitoring all of
its local hosts and services. It would be interesting to tie the two
together. Ultimatey
what I'd like to see is a 30,000
I'm trying to build nagios plugins (1.4.12) on several 32 bit hosts (suse 10.1).
The configs complete but throw a lot of warnings about 32 vs 64 bit. Running the
plug-ins generate segmentation faults. I tried setting LD_LIBRARY_PATH
but that didn't make a difference. Google shows me one potential
This can be done without NRPE. We have a cluster on a private net.
The head node is visible from nagios so I wrote a plugin that lives
on the head node (The dual homed machine, in your case). On the nagios
side, a plugin ssh'es to the head node, runs the script which talks to
hosts on the private
We're looking for some method to have nagios check the status of our
many backups
run off a backup server. Currently, we have a perl script search for
just one or two specific
lines and pass the results to send_nsca. It's crude and doesn't cover
all instances.
The problem is the nsr logfile has
I've got two nagios installs with snmp giving different responses.
Details:
Host1 is an operational nagios 2.9 install on SuSE 10.1-64bit machine
Host2 is a development machine, Suse 10.2-64bit with newly installed
nagios 3.02 and nagios-plugins-1.4.12. I have all the perl modules necessary
(Net
Anyone know of a dirt-simple analog device that can take ANY red
status's from nagios and
transmit the signal to some visible device?
Details: We have a medium nagios install on a closed network and wish
to have it transmit
a something is wrong message to some device not on the network. We
are
Event handlers can take care of any common task but you need to
be careful with them. Why is a task required to be done so frequently?
Shouldn't we take care of the cause rather than the symptom?
Also, scripts that fix known problems can be very dangerous if
they inadvertantly run at times when a
I had a linux-based nagios 2.9 install with nagiosgraph 0.8. The server recently
shutdown and upon startup, while nagios ran fine and nagiosgraph.log
and perdata.log
both show the correct type of activity, new graphs simply weren't
generated after startup.
The classic: Illegal attempt to update
I'm trying to streamline my nagios config using wildcards.
Unfortunately, not all services I wish to define via wildcard
follows a clean set of rules. Is it possible to define a service
with a host list of something like *,!linux1, !linux2
I suspect the answer is no and what I'd need to do is use
You can do the NSclient installs remotely by staging it on a common
share, have a .cmd file something like this:
mkdir c:\nsclient
@if %ERRORLEVEL%==1 goto :EXIT
copy C:\\temp\nsclient\Win_2k_XP_Bin\* c:\nsclient
@if %ERRORLEVEL%==1 echo error occurred while copying files
Anyone know of a plugin (NRPE, nsclient) that will display back to nagios on
linux a list of users logged into win2000 or winXP? I can do this on the window
side thru scripting and running from the resource kit psloggedon on each host.
My users would find it useful to see which machines are
Is it possible to have hosts in two sets of groups? Currently I've divided
hosts by hostgroup names associated
with their physical location. But it would be quite useful to have
hostgroups based on operating systems. Anyone
have a way to accomplish this allow multiple views?
TIA
We have a legacy setup with tru64 hosts that are currently configured for
rsh, not ssh. I've installed the latest and greatest nagios on a relatively
recent fedora host and have used plugins with ssh in the past. The problem
is that rsh doesn't return status codes properly. Rather, it returns
40 matches
Mail list logo
