<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1126" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT size=2>To me, one of Nagios' main strengths is its ability to be
deployed in many different types of configurations. A central server
configuration with one Nagios agent polling information from other servers via
NRPE is extremely powerful because of the centralized view the single agent
gives via the web interface. Adding parameter support to NRPE would represent a
significant step forwards for Nagios because its current implementation prevents
central configuration. I believe the security implications raised by Ethan would
be adequately addressed by his proposal. Overall in terms of priorities with
Nagios, I would rank this fairly high.</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>I would suggest that strong consideration also be given to
creating a port of this next version of NRPE to the NT/WIN2K/XP platforms. It
seems that NSClient has been orphaned and a single model for remote checking of
all servers would ease installation, configuration and on-going maintenance of a
Nagios monitoring environment. Plugins would also of course have to follow for
these platforms but I believe users would be quick to develop them.</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>While I probably have as many issues with the NT/WIN2K/XP
platforms as other users of Nagios I still have to deal with the presence of
these platforms in my environment and need to monitor them. </FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>What issues would be involved in such a port, are there any
real show-stoppers (technical) and has this been discussed in the
past?</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>Tim Shouldice</FONT></DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=nagios@nagios.org href="mailto:nagios@nagios.org">Ethan Galstad</A>
</DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A
title=nagios-users@lists.sourceforge.net
href="mailto:nagios-users@lists.sourceforge.net">nagios-users</A> </DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Tuesday, January 07, 2003 12:48
AM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> RE: [Nagios-users] NRPE
enhancement</DIV>
<DIV><BR></DIV>I've thought about this for a while now and have decided that I
will <BR>allow arguments to be passed to plugins in future versions of
NRPE. <BR>I've gotten a lot of requests for this feature and I
understand that <BR>it makes configuring things on the central server much
easier. <BR>However, because of the security implications this will
have, I will <BR>also be doing the following:<BR><BR>1. Incorporate encryption
similar to NSCA that gives the NRPE daemon <BR>some assurance of a trust level
with the client<BR><BR>2. Strip all potentially dangerous shell metacharacters
from the <BR>arguments before they're passed to the plugin for
exection<BR><BR>Incorporating crypto into NRPE will affect performance a bit,
but it <BR>will still probably scale better than check_by_ssh. If the
user <BR>decides to disable native crypto support (and tunnel traffic through
<BR>stunnel, etc. instead), the NRPE daemon will refuse to accept
<BR>arguments for plugins unless the user specifically supplies a
--dont-<BR>blame-nrpe option to the configure script before compiling.
If the <BR>daemon gets compiled with this option, it will loudly complain in
the <BR>logs and elsewhere that is running in an insecure mode.<BR><BR>Does
this approach sound reasonable to people?<BR><BR><BR><BR>On 30 Dec 2002 at
16:43, Carroll, Jim P [Contractor] wrote:<BR><BR>> If you have 500
machines, and among those machines there are no two disks<BR>> alike, then
I can only imagine the grief ahead of you. To take a page out<BR>> of
<A href="http://www.infrastructures.org">www.infrastructures.org</A>, it's
desireable to maintain a convergence, not<BR>> divergence, among the
various systems. Homogeneity is A Good Thing ฎ.<BR>> <BR>> Having
said that, rather than maintaining 500 config files, why not maintain<BR>>
a single config file containing all the similar and dissimilar config<BR>>
particulars? Edit that one file, and either push it out from a trusted
host<BR>> ("gold server", in Infrastructures.org parlance). Better
still, set up an<BR>> rsync server (or some other server that you can live
with, and set up a cron<BR>> job to pull down the latest nrpe.cfg
file. (As has been emphasized on their<BR>> list, pull, never
push. But take your pick. :)<BR>> <BR>> In our location, I
maintain a single nrpe.cfg file, complete with all the<BR>> unique disk
definitions. I'm not sure if it matters to you, but for the<BR>> most
part, I've elected to use the percentage option of check_disk. 5%
free<BR>> on the root partition of an 18GB disk isn't going to be a whole
lot<BR>> different from 5% free on root of a 36GB disk. Sure,
mathematically they're<BR>> different, so if exacting differences is
critical in your environment,<BR>> create unique definitions for the
different root partitions. (I'm also<BR>> specifying mount points,
not device filenames.) On a related note, we have<BR>> partitions on
our database servers which follow a fairly straightforward<BR>> naming
scheme for the mount points, and since I don't expect those<BR>> partitions
to change, I can either exclude them, or I can set a trivial<BR>> threshold
for each of the warning and critical, but in those cases I specify<BR>>
kbytes. (I do the latter; this lets anyone with the appropriate
permissions<BR>> at the web interface view all the defined
partitions.)<BR>> <BR>> [Sidebar: I noticed at one point that
there's a limit to the length of the<BR>> command token in nrpe.cfg.
As a result, I simplified all the names. Lesson<BR>> learned:
Be economical with the keystrokes.]<BR>> <BR>> If you like, I can send
you the nrpe.cfg file that I install on all our<BR>> hosts.<BR>>
<BR>> You might find that instead of one master nrpe.cfg file, that you'd
rather<BR>> manage a handful of dissimilar files. Might still want to
use the<BR>> gold/config server, as above.<BR>> <BR>> Food for
thought.<BR>> <BR>> jc<BR>> <BR>> > -----Original
Message-----<BR>> > From: Dave Viner
[mailto:dviner@yahoo-inc.com]<BR>> > Sent: Monday, December 30, 2002
2:43 PM<BR>> > To: Carroll, Jim P [Contractor]; 'nagios-users'<BR>>
> Subject: RE: [Nagios-users] NRPE enhancement<BR>> > <BR>> >
<BR>> > Ok, so check_by_ssh doesn't scale well, and nrpe is scalable
<BR>> > cpu-wise, but<BR>> > has problems
configuration-wise. (Imagine 500 machines all <BR>> > of which
have 5<BR>> > disks, and different check_disk arguments for each.
That's 2500<BR>> > configuration lines that need to be maintained over
500 <BR>> > machines in 500<BR>> > files.)<BR>> > <BR>>
> Perhaps there is room for something more secure than the <BR>> >
current nrpe, but<BR>> > more scalable than check_by_ssh. Does
that seem reasonable? <BR>> > If so, does<BR>> > anyone have
suggestions for low cost security implementations <BR>> > that
could<BR>> > enhance the security of nrpe without the cost of
ssh?<BR>> > <BR>> > dave<BR>> > <BR>> > <BR>> >
-----Original Message-----<BR>> > From: <A
href="mailto:nagios-users-admin@lists.sourceforge.net">nagios-users-admin@lists.sourceforge.net</A><BR>>
> [mailto:nagios-users-admin@lists.sourceforge.net]On Behalf Of
Carroll,<BR>> > Jim P [Contractor]<BR>> > Sent: Monday, December
30, 2002 10:49 AM<BR>> > To: 'nagios-users'<BR>> > Subject: RE:
[Nagios-users] NRPE enhancement<BR>> > <BR>> > <BR>> > The
general consensus is that check_by_ssh isn't a solution <BR>> > which
will scale<BR>> > well, due to the nature of the number-crunching crypto
beast. <BR>> > This is where<BR>> > NRPE has the
advantage.<BR>> > <BR>> > check_by_ssh has the advantage when it
comes to punching <BR>> > through a firewall,<BR>> > assuming that
an appropriate port is open. Sure, you could <BR>> > open port
5666<BR>> > for NRPE, but has has been discussed, that doesn't quite
<BR>> > leave the same warm<BR>> > fuzzies.<BR>> > <BR>>
> HTH.<BR>> > <BR>> > jc<BR>> > <BR>> > >
-----Original Message-----<BR>> > > From: Dave Viner
[mailto:dviner@yahoo-inc.com]<BR>> > > Sent: Monday, December 30,
2002 12:14 PM<BR>> > > To: 'Ethan Galstad'; 'nagios-users'<BR>>
> > Subject: RE: [Nagios-users] NRPE enhancement<BR>> >
><BR>> > ><BR>> > > Let me ask a second question that
might help me understand<BR>> > > more clearly the<BR>> > >
situation. Check_by_ssh allows for the passing of arbitrary<BR>> >
> arguments to<BR>> > > arbitrary command from the centralized
Nagios server to any<BR>> > > remote machine<BR>> > > which
has sshd running. NRPE allows for executing a specific<BR>> > >
command with<BR>> > > specific arguments on any remote machine which
has nrpe running.<BR>> > ><BR>> > > As someone setting up
monitoring, when should I use<BR>> > > check_by_ssh and when<BR>>
> > should I use nrpe?<BR>> > ><BR>> > > dave<BR>>
> ><BR>> > ><BR>> > > -----Original
Message-----<BR>> > > From: <A
href="mailto:nagios-users-admin@lists.sourceforge.net">nagios-users-admin@lists.sourceforge.net</A><BR>>
> > [mailto:nagios-users-admin@lists.sourceforge.net]On Behalf
Of<BR>> > > Dave Viner<BR>> > > Sent: Monday, December 30,
2002 9:07 AM<BR>> > > To: Tom Welsh; 'Ethan Galstad';
'nagios-users'<BR>> > > Subject: RE: [Nagios-users] NRPE
enhancement<BR>> > ><BR>> > ><BR>> > > I don't
think that my enhancement allows an arbitrary command to be<BR>> > >
executed. I think the code that I wrote will only allow one<BR>> >
> of the commands<BR>> > > already listed in the nrpe.cfg file to
be executed. The<BR>> > > arguments passed<BR>> > >
are arbitrary, but not the command. (The code even checks to<BR>> > >
ensure that the<BR>> > > command requested, without any arguments,
exists before<BR>> > > executing it to<BR>> > > prevent
malicious usage of arguments.)<BR>> > ><BR>> > >
dave<BR>> > ><BR>> > ><BR>> > ><BR>> > >
-----Original Message-----<BR>> > > From: Tom Welsh
[mailto:twelsh@square-box.com]<BR>> > > Sent: Saturday, December 28,
2002 5:02 PM<BR>> > > To: 'Dave Viner'; 'Ethan Galstad';
'nagios-users'<BR>> > > Subject: RE: [Nagios-users] NRPE
enhancement<BR>> > ><BR>> > ><BR>> > > In my humble
opinion an option that allows an arbitary command to be<BR>> > >
executed and which by "default" is switched off is an <BR>> > accident
waiting<BR>> > > to happen.<BR>> > ><BR>> > > It
only takes 1 security breach via a plugin to completely <BR>> > destroy
the<BR>> > > good name Nagios and its associated plugins
have.<BR>> > ><BR>> > > There is a good truism that states
"good news travels fast,<BR>> > > but bad news<BR>> > >
travels even faster"<BR>> > ><BR>> > > I for one would not
be too happy having a command available on my<BR>> > > network,
trusted or not that would allow commands to be executed<BR>> > >
remotely pon a box. For one. It's the kind of thing im <BR>> > always
looking<BR>> > > for when im "playing on my " networks.<BR>> >
><BR>> > > Well that's my two cents worth<BR>> >
><BR>> > > Cheers<BR>> > ><BR>> > > Tom
Welsh<BR>> > > <A
href="mailto:twelsh@square-box.com">twelsh@square-box.com</A><BR>> >
><BR>> > > -----Original Message-----<BR>> > > From: <A
href="mailto:nagios-users-admin@lists.sourceforge.net">nagios-users-admin@lists.sourceforge.net</A><BR>>
> > [mailto:nagios-users-admin@lists.sourceforge.net] On Behalf Of
Dave<BR>> > > Viner<BR>> > > Sent: 28 December 2002
21:46<BR>> > > To: Ethan Galstad; nagios-users<BR>> > >
Subject: RE: [Nagios-users] NRPE enhancement<BR>> > ><BR>> >
> These are excellent arguments for not incorporating the<BR>> > >
enhancement I am<BR>> > > suggesting. However, I suspect that
there are lots of<BR>> > > installations of<BR>> > > Nagios
and NRPE that run on completely trusted network. (Or<BR>> > >
the risk of<BR>> > > network intrusion through NRPE is worth the
benefit of reduced<BR>> > > configuration<BR>> > >
management.)<BR>> > ><BR>> > > What do you think about
incorporating this enhancement but have it<BR>> > > turned<BR>>
> > off by default, and enabled only at configure time?<BR>> >
><BR>> > > dave<BR>> > ><BR>> > ><BR>> >
> -----Original Message-----<BR>> > > From: <A
href="mailto:nagios-users-admin@lists.sourceforge.net">nagios-users-admin@lists.sourceforge.net</A><BR>>
> > [mailto:nagios-users-admin@lists.sourceforge.net]On Behalf Of
Ethan<BR>> > > Galstad<BR>> > > Sent: Friday, December 27,
2002 7:38 PM<BR>> > > To: nagios-users<BR>> > > Subject: RE:
[Nagios-users] NRPE enhancement<BR>> > ><BR>> > ><BR>>
> > There are several reasons why I have not added support for
arguments<BR>> > > to checks in NRPE. Most have been touched on
in the past on the<BR>> > > list, but I'll reiterate them here.
The main issue is not <BR>> > overruning<BR>> > > the 2K packet
that the check_nrpe plugin and NRPE daemon pass back<BR>> > > and
forth - that can be easily avoided...<BR>> > ><BR>> > >
1.<BR>> > > Users connecting to the NRPE are not authenticated.
Sure, you can<BR>> > > restrict connection based on IP address using
TCP wrappers, but they<BR>> > > are still not authenticated.
Also, I am not too familiar with IP<BR>> > > spoofing, but I'm sure
its possible for someone to fake the<BR>> > > originating address of
the connection and get the NRPE daemon to<BR>> > > accept the packet
and execute the necessary plugin without too much<BR>> > >
trouble.<BR>> > ><BR>> > > 2.<BR>> > > Some plugins
(like check_dhcp) may (have to) be installed suid root.<BR>> > >
Regardless of what user the NRPE daemon is running as, these plugins<BR>>
> > will be executed with higher privs.<BR>> > ><BR>> >
> 3.<BR>> > > Plugins can be made to segfault under the right
conditions. <BR>> > Sure, we<BR>> > > can try and
eliminate this possibility, but it will probably always<BR>> > >
exist to some extent since many plugins call system commands to get<BR>>
> > their data.<BR>> > ><BR>> > > ---<BR>> >
><BR>> > > Most remote exploits rely on buffer overflows/segfaults
to get their<BR>> > > work done, so allowing unauthenticated users to
pass arbitrary<BR>> > > arguments/data to plugins that might be
running suid commands is a<BR>> > > very bad idea indeed.<BR>>
> ><BR>> > > Stunnel would provide some security, but there is
no guarantee that<BR>> > > everyone would use it. There would
undoubtably be many people that<BR>> > > would put off implementing
it until they finished "testing" <BR>> > NRPE. In<BR>> >
> the worst case, they might never get around to implementing
stunnel<BR>> > > at all. In the likely best case scenario,
there is at <BR>> > least a window<BR>> > > of
opportunity. I just don't want to be responsible for <BR>> > the
possible<BR>> > > carnage that happens at that point. :-)<BR>>
> ><BR>> > > Also, incorporating native encryption into NRPE
involves reinventing<BR>> > > the wheel called "check_by_ssh", so I'm
really interested in doing<BR>> > > that.<BR>> > ><BR>>
> ><BR>> > ><BR>> > > On 27 Dec 2002 at 13:46,
Carroll, Jim P [Contractor] wrote:<BR>> > ><BR>> > > >
I'm not a C programmer by profession, so I defer your <BR>> > query to
those<BR>> > > who<BR>> > > > have a strong background,
both in C code and system/network<BR>> > > security.<BR>> >
> It<BR>> > > > does presume that every other link in the chain
is bulletproof.<BR>> > > [Insert<BR>> > > > ObRef to
Bugtraq here.]<BR>> > > ><BR>> > > > At any rate, I'm
curious to hear why Ethan didn't choose<BR>> > > that
approach<BR>> > > to<BR>> > > > begin with.<BR>> >
> ><BR>> > > > jc<BR>> > > ><BR>> > >
> > -----Original Message-----<BR>> > > > > From: Dave
Viner [mailto:dviner@yahoo-inc.com]<BR>> > > > > Sent: Friday,
December 27, 2002 12:49 PM<BR>> > > > > To: <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > Subject: RE: [Nagios-users] NRPE enhancement<BR>> >
> > ><BR>> > > > ><BR>> > > > > This
sounds interesting, but I have a question about the<BR>> > > >
> security implications of this code. I'm not a security<BR>> >
> > > expert, so please excuse the somewhat basic question.
The<BR>> > > > > struct packet as defined in common/common.h
has an argv<BR>> > > > > member which is a character array of
length 2048. I believe<BR>> > > > > this means that
if the incoming packet has an argv member<BR>> > > > > whose
length is greater than 2048 chars, then the<BR>> > > > >
rc=recvall(sock,(char<BR>> > > > >
*)&receive_packet,&bytes_to_recv,socket_timeout);<BR>> > >
> > should fail, should it not?<BR>> > > > ><BR>> >
> > > However, I think your suggestions regarding stunnel,
and<BR>> > > > > encryption are good ones, regardless of the
inclusion of<BR>> > > this code.<BR>> > > > ><BR>>
> > > > thanks<BR>> > > > ><BR>> > > >
> dave<BR>> > > > ><BR>> > > > ><BR>> >
> > > -----Original Message-----<BR>> > > > > From:
Carroll, Jim P [Contractor]<BR>> > >
[mailto:jcarro10@sprintspectrum.com]<BR>> > > > > Sent: Friday,
December 27, 2002 10:20 AM<BR>> > > > > To: 'Dave Viner'; <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > Subject: RE: [Nagios-users] NRPE enhancement<BR>> >
> > ><BR>> > > > ><BR>> > > > > I think
it's a good idea, but with the following provisions:<BR>> > > >
><BR>> > > > > - This should not be enabled by
default.<BR>> > > > ><BR>> > > > > - The
configure script, the Makefile and any/all NRPE docs<BR>> > > >
> should explicitly<BR>> > > > > state the security risks in
forcing the non-default<BR>> > > (added feature)<BR>> > >
> > behaviour.<BR>> > > > ><BR>> > > > > -
If the daemon is compiled with this option, anytime the<BR>> > > >
> daemon starts, it<BR>> > > > > should briefly mention that
it has been compiled for this<BR>> > > > > behaviour, and
a<BR>> > > > > quick remark about the increased risks.
(Sent to stderr if<BR>> > > > > standalone, else<BR>> >
> > > sent to syslog if running under (x)inetd). It should
scream<BR>> > > > > loud and clear<BR>> > > > >
if it's started under root; preferably it would simply not<BR>> > >
> > run as root, full<BR>> > > > > stop.<BR>> >
> > ><BR>> > > > > - Perhaps a reference to
implementing NRPE with stunnel (and<BR>> > > > > only
permitting<BR>> > > > > connections from localhost, as defined
in nrpe.cfg) would be<BR>> > > > > desireable.<BR>> >
> > ><BR>> > > > > I'm not a security guru, but it
seems to me that facilitating<BR>> > > > > this feature<BR>>
> > > > would open oneself up to a buffer overflow attack.
If you're<BR>> > > > > on a trusted<BR>> > > > >
network, it's a non-issue.<BR>> > > > ><BR>> > > >
> On a related note, I'd be much more comfy with this feature<BR>> >
> > > if there were a<BR>> > > > > facility to enforce
some level of native encryption, such as<BR>> > > > > what NSCA
uses.<BR>> > > > > If you don't have the keys to the house, you
get dropped on<BR>> > > > > the floor. (I<BR>> >
> > > have a similar wish for NSClient.)<BR>> > > >
><BR>> > > > > Food for thought.<BR>> > > >
><BR>> > > > > jc<BR>> > > > ><BR>> >
> > > > -----Original Message-----<BR>> > > > >
> From: Dave Viner [mailto:dviner@yahoo-inc.com]<BR>> > > >
> > Sent: Friday, December 27, 2002 11:48 AM<BR>> > > > >
> To: <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > > Subject: RE: [Nagios-users] NRPE enhancement<BR>>
> > > > ><BR>> > > > > ><BR>> > >
> > > In order to clarify the idea that I'm proposing, I've made
a<BR>> > > > > > patch to the nrpe source that implements
what I'm describing.<BR>> > > > > > This patch is made
against the nrpe-1.5.tar.gz from<BR>> > > sourceforge.<BR>> >
> > > ><BR>> > > > > > Essentially, these
changes allow us to specify in the<BR>> > > > > > nrpe.cfg
file lines like this:<BR>> > > > > >
command[check_disk_gen]=/usr/local/libexec/nagios/check_disk<BR>> > >
> > ><BR>> > > > > > Then when invoking check_nrpe,
you can invoke it like this:<BR>> > > > > >
./check_nrpe 127.0.0.1 -V 2 -c check_disk_gen -a "-w 50000<BR>> > >
> > > -c 10000 -p /dev/ad0s1e"<BR>> > > > >
><BR>> > > > > > And the effect is that
/usr/local/libexec/nagios/check_disk<BR>> > > > > > is
invoked with the -w 50000 -c 10000 -p /dev/ad0s1e as the<BR>> > >
> > > argument string. For example:<BR>> > > > >
><BR>> > > > > >
~/nagios/nrpe-1.5.new/src>./check_nrpe 127.0.0.1 -V 2 -c<BR>> > >
> > > check_disk_gen -a "-w 50000 -c 10000 -p /dev/ad0s1e"<BR>>
> > > > > DISK OK - [1484108 kB (9%) free on
/dev/ad0s1e]<BR>> > > > > >
~/nagios/nrpe-1.5.new/src><BR>> > > > > ><BR>> >
> > > > I think this is really useful and would greatly reduce
the<BR>> > > > > > size of the nrpe.cfg and, more
importantly, would reduce the<BR>> > > > > > number of times
you'd need to modify that configuration file.<BR>> > > > >
> Instead the modifications would occur on the centralized<BR>>
> > > > > Nagios server's configuration file.<BR>> > >
> > ><BR>> > > > > > What does everyone
think? Should we add this to the main<BR>> > > > > >
source for NRPE-1.6?<BR>> > > > > ><BR>> > > >
> > dave<BR>> > > > > ><BR>> > > > >
><BR>> > > > > > -----Original Message-----<BR>> >
> > > > From: <A
href="mailto:nagios-users-admin@lists.sourceforge.net">nagios-users-admin@lists.sourceforge.net</A><BR>>
> > > > > [mailto:nagios-users-admin@lists.sourceforge.net]On
Behalf Of<BR>> > > > > > Dave Viner<BR>> > > >
> > Sent: Monday, December 23, 2002 8:51 AM<BR>> > > > >
> To: Naios Users<BR>> > > > > > Subject: RE:
[Nagios-users] NRPE enhancement<BR>> > > > > ><BR>> >
> > > ><BR>> > > > > > Hi Rue,<BR>> > >
> > > Security is a great reason for limiting the commands<BR>>
> > > > > that NRPE is able to execute. But my suggested
enhancement<BR>> > > > > > wouldn't allow NRPE to execute
any command that isn't listed<BR>> > > > > > in the cfg
file. That is, the NRPE would still need to find<BR>> > > >
> > the path to the executable in the nrpe.cfg file, then use
any<BR>> > > > > > remaining information as arguments passed
to the executable.<BR>> > > > > > It is true that this is
less secure that forcing the entire<BR>> > > > > > command
line (executable and arguments) in the config file.<BR>> > > >
> > But, so long as the executables are well authored and handle<BR>>
> > > > > unexpected arguments well, I think this enhancement
would not<BR>> > > > > > significantly decrease
security. Do you think that<BR>> > > > > > specifying
arguments would make NRPE significantly <BR>> > less secure?<BR>>
> > > > ><BR>> > > > > ><BR>> > >
> > > Dave<BR>> > > > > ><BR>> > > >
> ><BR>> > > > > > -----Original Message-----<BR>>
> > > > > From: <A
href="mailto:nagios-users-admin@lists.sourceforge.net">nagios-users-admin@lists.sourceforge.net</A><BR>>
> > > > > [mailto:nagios-users-admin@lists.sourceforge.net]On
Behalf Of<BR>> > > > > > Rue Turner<BR>> > > >
> > Sent: Friday, December 20, 2002 1:33 PM<BR>> > > > >
> To: Naios Users<BR>> > > > > > Subject: Re:
[Nagios-users] NRPE enhancement<BR>> > > > > ><BR>> >
> > > ><BR>> > > > > > dave,<BR>> > >
> > ><BR>> > > > > > I think the reson for this
choice of configuration is<BR>> > > > > security. If
the<BR>> > > > > > nrpe was allowed to run whatever it was
asked it would<BR>> > > be easy to<BR>> > > > > >
compromise your machines. This way although your configs are<BR>> > >
> > > hefty (mine<BR>> > > > > > have almost a
hundred lines in) you can only ask it to run<BR>> > > > > >
commands from<BR>> > > > > > this library.<BR>> > >
> > ><BR>> > > > > > rue<BR>> > > >
> ><BR>> > > > > ><BR>> > > > > > On
Fri, 2002-12-20 at 17:35, Dave Viner wrote:<BR>> > > > > >
> Hi,<BR>> > > > > > > I'd like to suggest an
enhancement to NRPE, and if<BR>> > > > > > people think this
is a<BR>> > > > > > > good idea, I'll try to make a patch
to support my<BR>> > > > > > suggestion. Currently
the<BR>> > > > > > > nrpe.cfg file specifies all the
commands in this fashion:<BR>> > > > > > ><BR>> >
> > > > command[check_disk1]=/usr/local/nagios/libexec/check_disk
80<BR>> > > > > > 95 /dev/hda1<BR>> > > > >
> > As result of this design is that if you want to check<BR>> >
> > > something like<BR>> > > > > > > /dev/hda1
and /dev/hdb1, you need two seperate lines in the<BR>> > > > >
> nrpe.cfg file.<BR>> > > > > > > So, I'd like to
propose that we extend NRPE to allow<BR>> > > > > > for the
arguments to a<BR>> > > > > > > command to be specified
by the central Nagios server<BR>> > > > > > instead of in
the<BR>> > > > > > > nrpe.cfg. The idea is that the
nrpe.cfg would have one<BR>> > > > > > command line
which<BR>> > > > > > > maps a key, 'check_disk', to a
local executable,<BR>> > > > > > >
'/usr/local/nagios/libexec/check_disk'. The rest would be<BR>> >
> > > > specified from<BR>> > > > > > > the
central Nagios server in some manner.<BR>> > > > > > > I
think this would great simplify the nrpe.cfg files,<BR>> > > >
> > and reduce a lot of<BR>> > > > > > > redundant
command definitions that differ only in the<BR>> > > > >
arguments they<BR>> > > > > > > require. Also, it
would mean that you'd need to update<BR>> > > > > > your
nrpe.cfg very<BR>> > > > > > > rarely. In fact,
you'd only need to update it when you add<BR>> > > > > > a
new plugin.<BR>> > > > > > > I don't have a concrete
suggestion for implementing<BR>> > > > > > this yet, because
I<BR>> > > > > > > want to see if the community is
interested in this idea<BR>> > > > > > first. Has
this<BR>> > > > > > > idea been suggested
previously? Is anyone currently<BR>> > > > > >
interested in the idea<BR>> > > > > > > or would I be the
only consumer of such a service?<BR>> > > > > > ><BR>>
> > > > > > thanks<BR>> > > > > > >
dave<BR>> > > > > > ><BR>> > > > > >
><BR>> > > > > > ><BR>> > > > > >
> -------------------------------------------------------<BR>> > >
> > > > This SF.NET email is sponsored by: The Best
Geek<BR>> > > Holiday Gifts!<BR>> > > > > > >
Time is running out! Thinkgeek.com has the coolest <BR>> > gifts
for<BR>> > > > > > > your favorite geek. Let
your fingers do the <BR>> > typing. Visit<BR>> > >
Now.<BR>> > > > > > > T H I N K G E E K . C O
M <BR>> > <A
href="http://www.thinkgeek.com/sf/">http://www.thinkgeek.com/sf/</A><BR>>
> > > > > >
_______________________________________________<BR>> > > > >
> > Nagios-users mailing list<BR>> > > > > > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > > > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> > > > ><BR>> > > > > ><BR>> > >
> >
>
r u e t u r n e r<BR>> > > > > > ยท t ยท h
ยท i ยท n ยท l ยท a ยท y ยท e ยท r ยท<BR>> > > > > ><BR>>
> > > > > -- index, n.: Alphabetical list of words of no
possible<BR>> > > > > interest where<BR>> > > >
> > an alphabetical list of subjects with references ought to
be.<BR>> > > > > ><BR>> > > > > ><BR>>
> > > > >
-------------------------------------------------------<BR>> > > >
> > This SF.NET email is sponsored by: The Best Geek <BR>> >
Holiday Gifts!<BR>> > > > > > Time is running out!
Thinkgeek.com has the coolest gifts for<BR>> > > > > > your
favorite geek. Let your fingers do the typing.<BR>> > >
Visit Now.<BR>> > > > > > T H I N K G E E K . C O
M <A
href="http://www.thinkgeek.com/sf/">http://www.thinkgeek.com/sf/</A><BR>>
> > > > >
_______________________________________________<BR>> > > > >
> Nagios-users mailing list<BR>> > > > > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> > > > ><BR>> > > > > ><BR>> > >
> > ><BR>> > > > > ><BR>> > > > >
> -------------------------------------------------------<BR>> > >
> > > This sf.net email is sponsored by:ThinkGeek<BR>> > >
> > > Welcome to geek heaven.<BR>> > > > > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
> > > _______________________________________________<BR>> >
> > > > Nagios-users mailing list<BR>> > > > > >
<A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> > > > ><BR>> > > > > ><BR>> > >
> ><BR>> > > > ><BR>> > > > ><BR>> >
> > ><BR>> > > > >
-------------------------------------------------------<BR>> > > >
> This sf.net email is sponsored by:ThinkGeek<BR>> > > > >
Welcome to geek heaven.<BR>> > > > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
> > _______________________________________________<BR>> > >
> > Nagios-users mailing list<BR>> > > > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> > > ><BR>> > > ><BR>> > > ><BR>> >
> > -------------------------------------------------------<BR>> >
> > This sf.net email is sponsored by:ThinkGeek<BR>> > > >
Welcome to geek heaven.<BR>> > > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
> _______________________________________________<BR>> > > >
Nagios-users mailing list<BR>> > > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> > ><BR>> > ><BR>> > ><BR>> > ><BR>>
> > Ethan Galstad,<BR>> > > Nagios Developer<BR>> > >
---<BR>> > > Email: <A
href="mailto:nagios@nagios.org">nagios@nagios.org</A><BR>> > >
Website: <A href="http://www.nagios.org">http://www.nagios.org</A><BR>>
> ><BR>> > ><BR>> > ><BR>> > >
-------------------------------------------------------<BR>> > > This
sf.net email is sponsored by:ThinkGeek<BR>> > > Welcome to geek
heaven.<BR>> > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
_______________________________________________<BR>> > > Nagios-users
mailing list<BR>> > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> ><BR>> > ><BR>> > ><BR>> > ><BR>> >
> -------------------------------------------------------<BR>> > >
This sf.net email is sponsored by:ThinkGeek<BR>> > > Welcome to geek
heaven.<BR>> > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
_______________________________________________<BR>> > > Nagios-users
mailing list<BR>> > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> ><BR>> > ><BR>> > ><BR>> > ><BR>> >
><BR>> > ><BR>> > >
-------------------------------------------------------<BR>> > > This
sf.net email is sponsored by:ThinkGeek<BR>> > > Welcome to geek
heaven.<BR>> > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
_______________________________________________<BR>> > > Nagios-users
mailing list<BR>> > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> ><BR>> > ><BR>> > ><BR>> > ><BR>> >
> -------------------------------------------------------<BR>> > >
This sf.net email is sponsored by:ThinkGeek<BR>> > > Welcome to geek
heaven.<BR>> > > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> > >
_______________________________________________<BR>> > > Nagios-users
mailing list<BR>> > > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> > <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> ><BR>> > <BR>> > <BR>> >
-------------------------------------------------------<BR>> > This
sf.net email is sponsored by:ThinkGeek<BR>> > Welcome to geek
heaven.<BR>> > <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>> >
_______________________________________________<BR>> > Nagios-users
mailing list<BR>> > <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
> <A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
> <BR>> > <BR>> <BR>> <BR>>
-------------------------------------------------------<BR>> This sf.net
email is sponsored by:ThinkGeek<BR>> Welcome to geek heaven.<BR>> <A
href="http://thinkgeek.com/sf">http://thinkgeek.com/sf</A><BR>>
_______________________________________________<BR>> Nagios-users mailing
list<BR>> <A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR>>
<A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A><BR>>
<BR><BR><BR><BR>Ethan Galstad,<BR>Nagios Developer<BR>---<BR>Email: <A
href="mailto:nagios@nagios.org">nagios@nagios.org</A><BR>Website: <A
href="http://www.nagios.org">http://www.nagios.org</A><BR><BR><BR><BR>-------------------------------------------------------<BR>This
SF.NET email is sponsored by:<BR>SourceForge Enterprise Edition + IBM +
LinuxWorld <A
href="http://www.vasoftware.com">http://www.vasoftware.com</A><BR>_______________________________________________<BR>Nagios-users
mailing list<BR><A
href="mailto:Nagios-users@lists.sourceforge.net">Nagios-users@lists.sourceforge.net</A><BR><A
href="https://lists.sourceforge.net/lists/listinfo/nagios-users">https://lists.sourceforge.net/lists/listinfo/nagios-users</A></BLOCKQUOTE></BODY></HTML>