allowed_hosts is only applied if nrpe is running as a daemon. If you're running
under a service framework such as inetd/xinetd and apparently SMF according to
your experience, it's ignored.
The Solaris 10 servers I was monitoring were decommissioned last week so I
can't help much w
chance and see if I could monitor since I
> was receiving some data from the command line, just this one error.
> And sure enough, can. Now my question is WHY this is like this?
What's the value of the allowed_hosts directive in your nrpe.cfg?
--
Phil
t's still not working, post back with the
>relevant section of inetd.conf and the output of
svcs svc:/network/nrpe/tcp:default
and
inetadm -l svc:/network/nrpe/tcp:default
--
Phil Costelloe
Foundation IT, Hermitage Berkshire RG18 9SE
-
> -Original Message-
> From: Hari Sekhon [mailto:[EMAIL PROTECTED]
>
> Phil Costelloe wrote:
> >> I'm also looking forward to this. I'm a big fan of OpmonAgent as a
> >> replacement
> >> for Nsclient on Windows systems, it installs and
uot;.
I'm also looking forward to this. I'm a big fan of OpmonAgent as a replacement
for Nsclient on Windows systems, it installs and runs without any fuss at
all. I like things that Just Work. :)
--
Phil Costelloe
Foundation IT, Hermitage Berkshire RG18 9SE
-
The times on the two error messages did seem a bit distant, they may relate to
different ways you tried to start nrpe. Is it working yet?
Phil
> -Original Message-
> From: Canhua [mailto:[EMAIL PROTECTED]
> Sent: 06 March 2008 00:13
> To: Phil Costelloe
> Subject: Re
omething other than root. I usually set up a nagios user and group with the
least login and permission rights I can get away with. Also in nrpe.cfg, add
10.240.0.54 and 127.0.0.1 to allowed_hosts. Restart nrpe and see what
difference that makes.
Phil
k=view&id=54&Itemid=54.
If anything it's easier to install and maintain than NSClient and It Just
Works which is always good.
Phil
--
Phil Costelloe
Technical Consultant
mailto:[EMAIL PROTECTED]
http://www.foun
27
to /etc/hosts.allow which will permit localhost access to the nrpe service.
Phil
--
Phil Costelloe
Technical Consultant
mailto:[EMAIL PROTECTED]
http://www.foundation-it.com/
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +
_nt on the Nagios server in order to
query performance counters on the Windows servers. A Google for "check_nt
performance counters" should pull up some useful information. SNMP is also a
possibility for IIS and many other services if you prefer to go down that route.
--
Phil Costell
the OP wants if somebody writes (or has already written) an
http-to-gnokii gateway, but then that wouldn't really be a hardware
solution as requested. I've never run across anything that implements
the sending interface and modem hardware in one product but it may be
out there.
--
Phil
DDRESS$ -p $ARG1$ -v COUNTER -l
"\\SMTP Server(_Total)\\Remote Queue Length","SMTP Remote queue length %.0f" -w
$ARG2$ -c $ARG3$ -s $ARG4$
}
Looks like some of the characters have got mangled in yours.
--
Phil Costelloe
Foundation IT, Hermitage, Berkshire RG18 9SE
tual mem SNMP OK - Virtual Memory at 44% with 1,376 of 2,473 MB
free
MemuseMemory usage: total:2473.56 Mb - used: 1097.36 Mb
(44%) - free: 1376.20 Mb (56%)
--
Phil Costelloe
Foundation IT, Hermitage Berkshire RG18 9SE
-
Oracle client installed for full functionality or
InstantClient to use a restricted (but still useful) set of checks. Make
sure ORACLE_HOME is set, preferably in a script rather than relying on
an
environment variable, and you should be good to go.
--
Phil Costelloe
Foundation IT,
m,
> [ memused, GAUGE, $1],
> [ memwarn, GAUGE, $2],
> [ memcrit, GAUGE, $3] ];
I agree with your analysis, it probably is just a matter of configuring
the graph tool to recognise the output. But the OP mentioned
NagiosGrapher and that's a
sted
and return whatever that outputs.
Phil
----
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +44 (0) 1635 203700
Helpdesk: +44 (0) 8
ect System as the Performance Object and System Up
Time as the Counter, click Add then Close. If you select System Up Time
in the list of counters below the graph, do the counter statistics show
numbers (increasing by one every second) or big fat zeroes?
--
Phil Costelloe
Foundation IT, Hermitage Berksh
g on the monitored server is wrong, but I don't know
> what. Any ideas on what I can check?
What agent are you using on the server (e.g. NSClient, nc_net, etc.)? Do
you run any other agent checks against that server and do they work?
--
"/bin/ps axwo 'stat uid pid ppid vsz rss pcpu comm
args'"
On Solaris, it gives:
#define PS_COMMAND "/usr/local/nagios/libexec/pst3"
Phil
From: M V Ajay (vMoksha) [mailto:[EMAIL PROTECTED]
Sent: 24 May 2007 17:04
To
os are you using? What version of
the plugins? What version of nrpe? How did you compile/install the
plugins that are on the remote server, specifically were they compiled
either on that server or on a server running the same OS?
--
Phil Costelloe
Foundation IT, Hermitage Berkshir
Hugo van der Kooij wrote:
> On Mon, 30 Apr 2007, Navdeep Sidhu wrote:
>
>> I have already installed PNP for graphical output.
>
> Just to be sure you meant PHP here?
http://www.ederdrom.de/doku.php/nagios/pnp_en
--
Phil Costelloe
Foundation IT, Hermitage
l but I'd guess there's no group called
root and that's what's causing the error. Assuming that is the case, I
don't know how you'd fix it programmatically (there's probably an option
to configure that will do it) but you could manually change the line
INI
the usual tools - check_mssql.sh in
the standard plugin distribution for checking connectivity or by
querying the performance counters for more detailed info.
I'm not aware of any Sharepoint specific plugins.
--
Phil Costelloe
Foundation IT, Hermitage Berkshire
Is nrpe running as a standalone daemon or under inetd. If the latter,
can you post the relevant line from inetd.conf?
Phil
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
decode the SQL
logic without a bit more investigation.
Of course all these require you to install the DBD and DBI::Oracle Perl
modules so it would still be nice to have a shell script that does the
right thing.
--
Phil Costelloe
Foundation IT, Hermitage Berkshire RG18 9SE
t; that if I can't interrogate the tablespaces, I can't connect to the
> databases either. I'm not sure how the tablespace check would work
> with auto-extending tablespaces if you use them.
I'm curous about that myself. Anybody know?
--
Phil Costelloe wrote:
> Jim Avery wrote:
>> I confess I haven't used these yet myself (probably ought to though).
>> The ones I would immediately pick are:
>>
>> check_oracle --tns and
>> check_oracle --tablespace
>>
>> I probably wouldn
Phil Costelloe wrote:
> Rusty Hall wrote:
>> Has anyone tried accessing the performance counter files for SQL
>> Server 2000 Enterprise on a Windows 2003 server.
[snip]
> I monitor both of those (amongst many others) and get sensible
> results back (SQL Server 2000 Stand
SER1$/check_nt -H $HOSTADDRESS$ -p $ARG1$ -v COUNTER
-l "\\SQLServer:General Statistics\\User Connections","User connections
%.0f" -w $ARG2$ -c $ARG3$ -s $ARG4$
}
--
Phil Costelloe
Foundation IT, Hermitage Berkshire RG18 9SE
es up in sync with the
number of messages held in the queue. If I freeze the Local Delivery
queue in System Manager, the Nagios check "Exchange SMTP local queues"
(\\SMTP Server(_Total)\\Local Queue Length) goes up in sync with the
number of messages held in the queue.
Phil
---
H $HOSTADDRESS$ -p $ARG1$ -v COUNTER
-l "\\MSExchangeIS Public(_Total)\\Send Queue Size","IS Public send
queue size %.0f" -w $ARG2$ -c $ARG3$ -s $ARG4$
define command{
command_namecheck_nt_msexc_ispub_recq
command_line$USER1$/check_nt -H $HOSTADDRESS$ -p $ARG1$ -v COU
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +44 (0) 1635 203700
Helpdesk: +44 (0) 800 0121099
DDI: +44 (0) 1635
nexpectedly.
1067 is just a general "starting the service didn't work" error so
doesn't help much.
I'm confused as to whether you're installing nrpe or nsclient, can you
clarify the steps you've taken so far? I think you've probably missed
something out.
Ph
d"
will turn up plenty of general info.
Phil
----
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +44 (0) 1635 203700
Helpdesk: +44 (0) 800 0121099
DDI: +44 (0) 1635 20
tical) and
not a specific metric concerning inodes.
Phil
----
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +44 (0) 1635 203700
Helpdesk: +44 (0) 800 0121099
DDI: +44 (0) 1635 20
/sbin/init.d
directory called nrpe. Here is the text of the script:
>
>/usr/local/nagios/bin/nrpe -n -c /usr/local/nagios/nrpe.cfg
-d
I'd say that you either want to run as a daemon or run under inetd, not
both.
Phil
--------
P
inname or http://192.168.13.2
> they would see the same permission denied error.
Not on a default Fedora install, which is what makes it difficult to
diagnose. Fedora serves a test page (the equivalent of the Apache It
Worked! page) but with a 403 HTTP status.
--
Phil Costelloe
Foundation IT, Hermi
Andy Shellam (Mailing Lists) wrote:
> I've noticed a few visitors have used Konqueror - are any of these
> people on this list? How did it look? Can anyone feedback how it
> behaves in Opera?
It looks and acts the same in Opera 9.02 on Windows as it does in
Firefox 1.5.
you need to create an index.html in /var/www/html. Apache on
Fedora returns 403 if there's no index.html. The Test Page is still
somehow automagically served if you use a browser but all Nagios sees is
the error.
Phil
----
Phil Cost
= c_idfp;
se->found =0;
se->found_len = 0;
*pathtail = se;
Presumably a fix of some kind is in the current CVS version.
Phil
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshir
o use the
Windows native kill program from the W2K Resource Kit.
Phil
----
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +44 (0) 1635 203700
Helpdesk:
rformance viewer that's installed with the Administrative Tools but
the definitive source (for Windows 2003 at least) seems to be
http://technet2.microsoft.com/WindowsServer/en/library/3fb01419-b1ab-4f5
2-a9f8-09d5ebeb9ef21033.mspx?mfr=true
Phil
------
d!foobar
etc etc
}
Untested but should just about work.
Phil
----
Phil Costelloe
Technical Consultant
[EMAIL PROTECTED]
http://www.foundation-it.com
Foundation IT
Foundation Court
Old Street
Oare
Hermitage Berkshire RG18 9SE
Switch: +44 (
't think you can read Docbook documents directly, I build mine to
HTML or PDF. A lot of Linux distributions come with all the necessary
tools to set up a build environment but I'm not sure how well they work
together out of the box.
Phil
nd of change?
Won't the plugin be using OpenSSL to do the certificate handling and
hence won't the authority checking be handled at the library layer?
I've just done a quick test and can "check_http -S" a site with a
self-signed certificate so the behaviour you seem to be
Sandeep Narasimha Murthy wrote:
>
> So, does load average: 0.35, 0.47, 0.53, mean there are only 0.53
> jobs running and waiting over the last 15 minutes ? executing TOP in
> the shell shows me at least 15 different processes..
The important distinction here is "running", i.e. actually on the C
Sand Philipp wrote:
> Thanks for the hint with snmp, but I think those are only the
> counters, you can also check via performance counter.
> I'll ask my Mailserver Admins, if these values are adequate.
>
> Otherwise, how do you monitor you Exchange Servers? Am I the first
> one, who wants to do
47 matches
Mail list logo
