pysnmp-users — PySNMP usage issues

[pysnmp-users] snmpbulk.py blocks on select

[Snezhkov, D. <Dim...@pe...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

Guys,

I just downloaded 
pysnmp-2.0.8 distribution and tried to do the following :


# ./snmpbulk.py  rtken CNSroCN 1.3.6.1.2.1.1.4    


<<<<<< Hangs .....  


Traceback (most recent call last):
  File "./snmpbulk.py", line 151, in ?
    client.dispatch()
  File "/opt/python/lib/python2.1/site-packages/pysnmp/bulkrole.py", line
219, in dispatch
    (response, src) = self.transport.receive()
  File "/opt/python/lib/python2.1/site-packages/pysnmp/role.py", line 155,
in receive
    r, w, x = select.select(r, w, x, self.timeout)
KeyboardInterrupt


However, 

root@hp19rm4 [/opt/pysnmp/pysnmp-2.0.8/examples] 
# ./snmpbulkwalk.py  rtken community 1.3.6.1.2.1.1.4
.1.3.6.1.2.1.1.4.0 ---> Name 


What's going on? 
Also, I read that snmpbulk.py script uses SNMP v.1/v.2c GET request to
retrieve one or more SNMP objects from multiple SNMP agents at once. How 
do I specify multiple agents ?  should I create a thread for  each request ?
Thanks,
dimitry.

Re: [pysnmp-users] snmpwalk.py

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

Mark,

I've modified the snmpwalk.py code to get it acting closer to Net-SNMP's
one. The experimental version is here:

ftp://ftp.glas.net/users/ilya/tools/pysnmp/unreleased/pysnmp-2.0.8.tar.gz

Could you please try it out and let me know if it works alright to you?

Thanks,
ilya

On 5 Dec 2002, Mark Kilmartin wrote:

> Thank you for the fast response on this.
> 
> In most of the cases I will be using it starting at N+1 is not a
> problem.
> 
> I will look at the snmptable code and see if it helps me much.
> I am unlikely to look at it till next week due to upcoming exams.
> 
> I also look forward to testing the new snmpwalk.py code.
> 
> Again thanks for the quick help.
> 
> MArk
> 
> 
> On Thu, 2002-12-05 at 08:33, Ilya Etingof wrote:
> > Mark,
> > 
> > You are absolutely correct in your analysis. I must have overlooked the
> > behavior of Net-SNMP's snmpwalk tool when implementing its Python
> > version.
> > 
> > I'm going to fix snmpwalk.py to make it acting more similar to
> > snmpwalk. This might take a few days. Will let you know then.
> > 
> > Meanwhile, I suggest you to try using the examples/snmptable.py tool
> > which, at least, stops as it walks out of the initial sub-tree:
> > 
> > [ilya@cray examples]$ ./snmptable.py ts29.moscow.st.net public
> > .1.3.6.1.2.1.1.1
> > .1.3.6.1.2.1.1.1.0  --->  'Cisco Internetwork Operating System Software
> > \015\012IOS (tm) 5400 Software (C5400-JS-M), Version 12.2(11.8b),
> > MAINTENANCE INTERIM SOFTWARE\015\012Copyright (c) 1986-2002 by cisco
> > Systems, Inc.\015\012Compiled Tue 30-Jul-02 19:02 by pwade'
> > [ilya@cray examples]$
> > 
> > Please, note, that snmpwalk.py/snmptable.py starts from the N+1 variable
> > in the MIB so you would have to give it the N-1 variable to fetch N-th.
> > Since snmpwalk seems to start from N, I would probably also have to fix
> > it in snmpwalk.py tool.
> > 
> > -ilya
> > 
> > > I have been playing with snmpwalk.py.
> > > 
> > > But it does not work as I would expect.
> > > It may be a fact that my expectations are wrong.
> > > 
> > > When I run the standard snmpwalk utility as follows I get the following
> > > output.
> > > snmpwalk 10.0.0.253 public .1.3.6.1.2.1.17.4.3.1.3
> > > 17.4.3.1.3.0.2.179.136.108.99 = 3
> > > 17.4.3.1.3.0.2.179.136.114.68 = 3
> > > 17.4.3.1.3.0.6.91.60.22.153 = 3
> > > 17.4.3.1.3.0.6.91.60.22.172 = 3
> > > 17.4.3.1.3.0.6.91.138.79.13 = 3
> > > 17.4.3.1.3.0.6.91.138.79.81 = 3
> > > 17.4.3.1.3.0.6.91.138.79.114 = 3
> > > 17.4.3.1.3.0.6.91.138.79.125 = 3
> > > 17.4.3.1.3.0.6.91.138.79.128 = 3
> > > 17.4.3.1.3.0.6.91.138.79.135 = 3
> > > 17.4.3.1.3.0.6.91.140.49.95 = 3
> > > 17.4.3.1.3.0.6.91.140.50.59 = 3
> > > 17.4.3.1.3.0.8.116.144.127.36 = 3
> > > 17.4.3.1.3.0.176.208.170.249.105 = 3
> > > 17.4.3.1.3.0.176.208.243.101.68 = 3
> > > 
> > > This is exactly what I expect.
> > > 
> > > But when I run the following instance of snmpwalk.py I get thousands of
> > > lines of output the first 15 of which are as above.
> > > I would expect snmpwalk.py to stop at this point but it stays walking
> > > through the rest of the SNMP system.
> > > 
> > > 
> > > Another example is if I run.
> > > snmpwalk 10.0.0.253 public .1.3.6.1.2.1.1.1.0
> > > I get the system description as expected.
> > > 
> > > On the other hand if I run
> > > ./snmpwalk.py 10.0.0.253 public 1.3.6.1.2.1.1.1.0
> > > I get lots of output none of which is the system description.
> > > 
> > > If anybody can help I would apreciate it.
> > > 
> > > MArk
> > > 
> > > 
> > > 
> > > 
> > > 
> > > -------------------------------------------------------
> > > This SF.net email is sponsored by: Microsoft Visual Studio.NET 
> > > comprehensive development tool, built to increase your 
> > > productivity. Try a free online hosted session at:
> > > http://ads.sourceforge.net/cgi-bin/redirect.pl?micr0003en
> > > _______________________________________________
> > > pysnmp-users mailing list
> > > pys...@li...
> > > https://lists.sourceforge.net/lists/listinfo/pysnmp-users
> > > 
> -- 
> Mark Kilmartin <mr...@eu...>
> 

Re: [pysnmp-users] snmpwalk.py

[Mark K. <mr...@eu...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

Thank you for the fast response on this.

In most of the cases I will be using it starting at N+1 is not a
problem.

I will look at the snmptable code and see if it helps me much.
I am unlikely to look at it till next week due to upcoming exams.

I also look forward to testing the new snmpwalk.py code.

Again thanks for the quick help.

MArk


On Thu, 2002-12-05 at 08:33, Ilya Etingof wrote:
> Mark,
> 
> You are absolutely correct in your analysis. I must have overlooked the
> behavior of Net-SNMP's snmpwalk tool when implementing its Python
> version.
> 
> I'm going to fix snmpwalk.py to make it acting more similar to
> snmpwalk. This might take a few days. Will let you know then.
> 
> Meanwhile, I suggest you to try using the examples/snmptable.py tool
> which, at least, stops as it walks out of the initial sub-tree:
> 
> [ilya@cray examples]$ ./snmptable.py ts29.moscow.st.net public
> .1.3.6.1.2.1.1.1
> .1.3.6.1.2.1.1.1.0  --->  'Cisco Internetwork Operating System Software
> \015\012IOS (tm) 5400 Software (C5400-JS-M), Version 12.2(11.8b),
> MAINTENANCE INTERIM SOFTWARE\015\012Copyright (c) 1986-2002 by cisco
> Systems, Inc.\015\012Compiled Tue 30-Jul-02 19:02 by pwade'
> [ilya@cray examples]$
> 
> Please, note, that snmpwalk.py/snmptable.py starts from the N+1 variable
> in the MIB so you would have to give it the N-1 variable to fetch N-th.
> Since snmpwalk seems to start from N, I would probably also have to fix
> it in snmpwalk.py tool.
> 
> -ilya
> 
> > I have been playing with snmpwalk.py.
> > 
> > But it does not work as I would expect.
> > It may be a fact that my expectations are wrong.
> > 
> > When I run the standard snmpwalk utility as follows I get the following
> > output.
> > snmpwalk 10.0.0.253 public .1.3.6.1.2.1.17.4.3.1.3
> > 17.4.3.1.3.0.2.179.136.108.99 = 3
> > 17.4.3.1.3.0.2.179.136.114.68 = 3
> > 17.4.3.1.3.0.6.91.60.22.153 = 3
> > 17.4.3.1.3.0.6.91.60.22.172 = 3
> > 17.4.3.1.3.0.6.91.138.79.13 = 3
> > 17.4.3.1.3.0.6.91.138.79.81 = 3
> > 17.4.3.1.3.0.6.91.138.79.114 = 3
> > 17.4.3.1.3.0.6.91.138.79.125 = 3
> > 17.4.3.1.3.0.6.91.138.79.128 = 3
> > 17.4.3.1.3.0.6.91.138.79.135 = 3
> > 17.4.3.1.3.0.6.91.140.49.95 = 3
> > 17.4.3.1.3.0.6.91.140.50.59 = 3
> > 17.4.3.1.3.0.8.116.144.127.36 = 3
> > 17.4.3.1.3.0.176.208.170.249.105 = 3
> > 17.4.3.1.3.0.176.208.243.101.68 = 3
> > 
> > This is exactly what I expect.
> > 
> > But when I run the following instance of snmpwalk.py I get thousands of
> > lines of output the first 15 of which are as above.
> > I would expect snmpwalk.py to stop at this point but it stays walking
> > through the rest of the SNMP system.
> > 
> > 
> > Another example is if I run.
> > snmpwalk 10.0.0.253 public .1.3.6.1.2.1.1.1.0
> > I get the system description as expected.
> > 
> > On the other hand if I run
> > ./snmpwalk.py 10.0.0.253 public 1.3.6.1.2.1.1.1.0
> > I get lots of output none of which is the system description.
> > 
> > If anybody can help I would apreciate it.
> > 
> > MArk
> > 
> > 
> > 
> > 
> > 
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Microsoft Visual Studio.NET 
> > comprehensive development tool, built to increase your 
> > productivity. Try a free online hosted session at:
> > http://ads.sourceforge.net/cgi-bin/redirect.pl?micr0003en
> > _______________________________________________
> > pysnmp-users mailing list
> > pys...@li...
> > https://lists.sourceforge.net/lists/listinfo/pysnmp-users
> > 
-- 
Mark Kilmartin <mr...@eu...>

Re: [pysnmp-users] snmpwalk.py

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

Mark,

You are absolutely correct in your analysis. I must have overlooked the
behavior of Net-SNMP's snmpwalk tool when implementing its Python
version.

I'm going to fix snmpwalk.py to make it acting more similar to
snmpwalk. This might take a few days. Will let you know then.

Meanwhile, I suggest you to try using the examples/snmptable.py tool
which, at least, stops as it walks out of the initial sub-tree:

[ilya@cray examples]$ ./snmptable.py ts29.moscow.st.net public
.1.3.6.1.2.1.1.1
.1.3.6.1.2.1.1.1.0  --->  'Cisco Internetwork Operating System Software
\015\012IOS (tm) 5400 Software (C5400-JS-M), Version 12.2(11.8b),
MAINTENANCE INTERIM SOFTWARE\015\012Copyright (c) 1986-2002 by cisco
Systems, Inc.\015\012Compiled Tue 30-Jul-02 19:02 by pwade'
[ilya@cray examples]$

Please, note, that snmpwalk.py/snmptable.py starts from the N+1 variable
in the MIB so you would have to give it the N-1 variable to fetch N-th.
Since snmpwalk seems to start from N, I would probably also have to fix
it in snmpwalk.py tool.

-ilya

> I have been playing with snmpwalk.py.
> 
> But it does not work as I would expect.
> It may be a fact that my expectations are wrong.
> 
> When I run the standard snmpwalk utility as follows I get the following
> output.
> snmpwalk 10.0.0.253 public .1.3.6.1.2.1.17.4.3.1.3
> 17.4.3.1.3.0.2.179.136.108.99 = 3
> 17.4.3.1.3.0.2.179.136.114.68 = 3
> 17.4.3.1.3.0.6.91.60.22.153 = 3
> 17.4.3.1.3.0.6.91.60.22.172 = 3
> 17.4.3.1.3.0.6.91.138.79.13 = 3
> 17.4.3.1.3.0.6.91.138.79.81 = 3
> 17.4.3.1.3.0.6.91.138.79.114 = 3
> 17.4.3.1.3.0.6.91.138.79.125 = 3
> 17.4.3.1.3.0.6.91.138.79.128 = 3
> 17.4.3.1.3.0.6.91.138.79.135 = 3
> 17.4.3.1.3.0.6.91.140.49.95 = 3
> 17.4.3.1.3.0.6.91.140.50.59 = 3
> 17.4.3.1.3.0.8.116.144.127.36 = 3
> 17.4.3.1.3.0.176.208.170.249.105 = 3
> 17.4.3.1.3.0.176.208.243.101.68 = 3
> 
> This is exactly what I expect.
> 
> But when I run the following instance of snmpwalk.py I get thousands of
> lines of output the first 15 of which are as above.
> I would expect snmpwalk.py to stop at this point but it stays walking
> through the rest of the SNMP system.
> 
> 
> Another example is if I run.
> snmpwalk 10.0.0.253 public .1.3.6.1.2.1.1.1.0
> I get the system description as expected.
> 
> On the other hand if I run
> ./snmpwalk.py 10.0.0.253 public 1.3.6.1.2.1.1.1.0
> I get lots of output none of which is the system description.
> 
> If anybody can help I would apreciate it.
> 
> MArk
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Microsoft Visual Studio.NET 
> comprehensive development tool, built to increase your 
> productivity. Try a free online hosted session at:
> http://ads.sourceforge.net/cgi-bin/redirect.pl?micr0003en
> _______________________________________________
> pysnmp-users mailing list
> pys...@li...
> https://lists.sourceforge.net/lists/listinfo/pysnmp-users
> 

[pysnmp-users] snmpwalk.py

[Mark K. <mr...@eu...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

I have been playing with snmpwalk.py.

But it does not work as I would expect.
It may be a fact that my expectations are wrong.

When I run the standard snmpwalk utility as follows I get the following
output.
snmpwalk 10.0.0.253 public .1.3.6.1.2.1.17.4.3.1.3
17.4.3.1.3.0.2.179.136.108.99 = 3
17.4.3.1.3.0.2.179.136.114.68 = 3
17.4.3.1.3.0.6.91.60.22.153 = 3
17.4.3.1.3.0.6.91.60.22.172 = 3
17.4.3.1.3.0.6.91.138.79.13 = 3
17.4.3.1.3.0.6.91.138.79.81 = 3
17.4.3.1.3.0.6.91.138.79.114 = 3
17.4.3.1.3.0.6.91.138.79.125 = 3
17.4.3.1.3.0.6.91.138.79.128 = 3
17.4.3.1.3.0.6.91.138.79.135 = 3
17.4.3.1.3.0.6.91.140.49.95 = 3
17.4.3.1.3.0.6.91.140.50.59 = 3
17.4.3.1.3.0.8.116.144.127.36 = 3
17.4.3.1.3.0.176.208.170.249.105 = 3
17.4.3.1.3.0.176.208.243.101.68 = 3

This is exactly what I expect.

But when I run the following instance of snmpwalk.py I get thousands of
lines of output the first 15 of which are as above.
I would expect snmpwalk.py to stop at this point but it stays walking
through the rest of the SNMP system.


Another example is if I run.
snmpwalk 10.0.0.253 public .1.3.6.1.2.1.1.1.0
I get the system description as expected.

On the other hand if I run
./snmpwalk.py 10.0.0.253 public 1.3.6.1.2.1.1.1.0
I get lots of output none of which is the system description.

If anybody can help I would apreciate it.

MArk

[pysnmp-users] ANN: pysnmp-3.2.2 has been released

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

This is mostly a bugfix release against 3.2.1. Most significant changes
include:

- Bugfix to default value for 'dst' address parameter in
  mapping/udp/role.py, mapping/udp/asynrole.py and
  mapping/udp/bulkrole.py.
- asyncode compatibility bugfix to asynrole.manager.handle_error()
  and asynrole.agent.handle_error() parameters list. Affected the
  mapping/udp/asynrole.py and examples/asyncmgr.py scripts.
- Bugfix to transport-level exceptions handling code in
  examples/snmpbulk.py and mapping/udp/bulkrole.py.
- Bugfix to reportTypeFlag handling at examples/snmpbulk.py.

Thanks,
ilya

Re: [pysnmp-users] bulkrole.py problem

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

I'm going to release the 3.2.2 right away.

Please, let me know if your further experimentation with it
unveals any more glitches.

Anyone's contribution to this project is always welcome!

Thanks,
ilya

On 14 Nov 2002, Mark Kilmartin wrote:

> With a quick test using snmpbulk.py in the examples seems to work.
> 
> I will not be able to test my own code with it until tomorrow.
> 
> But everything looks good.
> 
> Hopefully in the future my python will be good enough to allow me to
> make some contributions to this excellent project.
> 
> MArk
> 
> 
> On Thu, 2002-11-14 at 12:25, Ilya Etingof wrote:
> > 
> > Mark,
> > 
> > I've fixed a couple of bugs in the relevant code. Please, try the
> > release 3.2.2 (attached) and let me know if it works alright to you so I
> > would release it to public.
> > 
> > Thanks,
> > ilya
> > 
> > On 14 Nov 2002, Mark Kilmartin wrote:
> > 
> > > I now have a different error.
> > > 
> > > 
> > > 
> > > Traceback (most recent call last):
> > >   File "./snmpbulk.py", line 157, in ?
> > >     client.dispatch()
> > >   File
> > > "/usr/lib/python2.2/site-packages/pysnmp/mapping/udp/bulkrole.py", line
> > > 184, in dispatch
> > >     transports[idx].send(question, dst)
> > >   File "/usr/lib/python2.2/site-packages/pysnmp/mapping/udp/role.py",
> > > line 79, in send
> > >     self.open()
> > >   File "/usr/lib/python2.2/site-packages/pysnmp/mapping/udp/role.py",
> > > line 59, in open
> > >     if self.agent[0] is not None and self.agent[1] != 0:
> > > TypeError: unsubscriptable object
> > > 
> > > 
> > > MArk
> > > 
> > > On Thu, 2002-11-14 at 11:13, Ilya Etingof wrote:
> > > > A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
> > > > List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
> > > > 	<mailto:pys...@li...?subject=unsubscribe>
> > > > List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>
> > > > Date: Thu, 14 Nov 2002 14:11:51 +0300 (MSK)
> > > > 
> > > > 
> > > > Oh, that's a bug. Please, try the following patch and let me know if it
> > > > helps. I'm going to commit it to the next (3.2.2) revision of pysnmp.
> > > > 
> > > > Thanks,
> > > > ilya
> > > > 
> > > > 8X-----------------------------------------
> > > > 
> > > > *** bulkrole.py.orig	Tue Oct  1 12:16:07 2002
> > > > --- bulkrole.py	Thu Nov 14 14:04:37 2002
> > > > ***************
> > > > *** 183,189 ****
> > > >                   try:
> > > >                       transports[idx].send(question, dst)
> > > >                       
> > > > !                 except role.Error, why:
> > > >                       # Return exception
> > > >                       self._responses[idx] = (self._responses[idx][0], why,\
> > > >                                               self._responses[idx][2])
> > > > --- 183,189 ----
> > > >                   try:
> > > >                       transports[idx].send(question, dst)
> > > >                       
> > > > !                 except error.SnmpOverUdpError, why:
> > > >                       # Return exception
> > > >                       self._responses[idx] = (self._responses[idx][0], why,\
> > > >                                               self._responses[idx][2])
> > > > ***************
> > > > *** 213,219 ****
> > > >                               try:
> > > >                                   (answer, src) = transports[idx].receive()
> > > >   
> > > > !                             except role.Error, why:
> > > >                                   # Return exception
> > > >                                   self._responses[idx] = (self._responses[idx][0],\
> > > >                                                           why,\
> > > > --- 213,219 ----
> > > >                               try:
> > > >                                   (answer, src) = transports[idx].receive()
> > > >   
> > > > !                             except error.SnmpOverUdpError, why:
> > > >                                   # Return exception
> > > >                                   self._responses[idx] = (self._responses[idx][0],\
> > > >                                                           why,\
> > > > 
> > > > On Thu, 14 Nov 2002, Mark Kilmartin wrote:
> > > > 
> > > > > A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
> > > > > List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
> > > > > 	<mailto:pys...@li...?subject=unsubscribe>
> > > > > List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>
> > > > > Date: 14 Nov 2002 10:16:30 +0000
> > > > > 
> > > > > I have only been using pysnmp for a short while now but I have uncovered
> > > > > an error in bulkrole.py
> > > > > 
> > > > > When I try to run the snmpbulk.py example or my own program written to
> > > > > use bulkrole I get the following error.
> > > > > 
> > > > > Traceback (most recent call last):
> > > > >   File "./snmpbulk.py", line 157, in ?
> > > > >     client.dispatch()
> > > > >   File
> > > > > "/usr/lib/python2.2/site-packages/pysnmp/mapping/udp/bulkrole.py", line
> > > > > 186, in dispatch
> > > > >     except role.Error, why:
> > > > > AttributeError: 'module' object has no attribute 'Error'
> > > > > 
> > > > > 
> > > > > Unfortunately my python knowledge is currently not good enough to track
> > > > > down this problem.
> > > > > 
> > > > > MArk
> > > > > 
> > > > > 
> > > > > 
> > > > > -------------------------------------------------------
> > > > > This sf.net email is sponsored by: To learn the basics of securing 
> > > > > your web site with SSL, click here to get a FREE TRIAL of a Thawte 
> > > > > Server Certificate: http://www.gothawte.com/rd524.html
> > > > > _______________________________________________
> > > > > pysnmp-users mailing list
> > > > > pys...@li...
> > > > > https://lists.sourceforge.net/lists/listinfo/pysnmp-users
> > > > > 
> > > > 
> > > > 
> > > > 
> > > > -------------------------------------------------------
> > > > This sf.net email is sponsored by: To learn the basics of securing 
> > > > your web site with SSL, click here to get a FREE TRIAL of a Thawte 
> > > > Server Certificate: http://www.gothawte.com/rd524.html
> > > > _______________________________________________
> > > > pysnmp-users mailing list
> > > > pys...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/pysnmp-users
> > > 
> > > 
> 
> 

Re: [pysnmp-users] bulkrole.py problem

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

Oh, that's a bug. Please, try the following patch and let me know if it
helps. I'm going to commit it to the next (3.2.2) revision of pysnmp.

Thanks,
ilya

8X-----------------------------------------

*** bulkrole.py.orig	Tue Oct  1 12:16:07 2002
--- bulkrole.py	Thu Nov 14 14:04:37 2002
***************
*** 183,189 ****
                  try:
                      transports[idx].send(question, dst)
                      
!                 except role.Error, why:
                      # Return exception
                      self._responses[idx] = (self._responses[idx][0], why,\
                                              self._responses[idx][2])
--- 183,189 ----
                  try:
                      transports[idx].send(question, dst)
                      
!                 except error.SnmpOverUdpError, why:
                      # Return exception
                      self._responses[idx] = (self._responses[idx][0], why,\
                                              self._responses[idx][2])
***************
*** 213,219 ****
                              try:
                                  (answer, src) = transports[idx].receive()
  
!                             except role.Error, why:
                                  # Return exception
                                  self._responses[idx] = (self._responses[idx][0],\
                                                          why,\
--- 213,219 ----
                              try:
                                  (answer, src) = transports[idx].receive()
  
!                             except error.SnmpOverUdpError, why:
                                  # Return exception
                                  self._responses[idx] = (self._responses[idx][0],\
                                                          why,\

On Thu, 14 Nov 2002, Mark Kilmartin wrote:

> A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
> List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
> 	<mailto:pys...@li...?subject=unsubscribe>
> List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>
> Date: 14 Nov 2002 10:16:30 +0000
> 
> I have only been using pysnmp for a short while now but I have uncovered
> an error in bulkrole.py
> 
> When I try to run the snmpbulk.py example or my own program written to
> use bulkrole I get the following error.
> 
> Traceback (most recent call last):
>   File "./snmpbulk.py", line 157, in ?
>     client.dispatch()
>   File
> "/usr/lib/python2.2/site-packages/pysnmp/mapping/udp/bulkrole.py", line
> 186, in dispatch
>     except role.Error, why:
> AttributeError: 'module' object has no attribute 'Error'
> 
> 
> Unfortunately my python knowledge is currently not good enough to track
> down this problem.
> 
> MArk
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: To learn the basics of securing 
> your web site with SSL, click here to get a FREE TRIAL of a Thawte 
> Server Certificate: http://www.gothawte.com/rd524.html
> _______________________________________________
> pysnmp-users mailing list
> pys...@li...
> https://lists.sourceforge.net/lists/listinfo/pysnmp-users
> 

[pysnmp-users] bulkrole.py problem

[Mark K. <mr...@eu...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=pysnmp-users>

I have only been using pysnmp for a short while now but I have uncovered
an error in bulkrole.py

When I try to run the snmpbulk.py example or my own program written to
use bulkrole I get the following error.

Traceback (most recent call last):
  File "./snmpbulk.py", line 157, in ?
    client.dispatch()
  File
"/usr/lib/python2.2/site-packages/pysnmp/mapping/udp/bulkrole.py", line
186, in dispatch
    except role.Error, why:
AttributeError: 'module' object has no attribute 'Error'


Unfortunately my python knowledge is currently not good enough to track
down this problem.

MArk

[pysnmp-users] ANN: pysnmp-3.1.0 is out

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchives/forum.php?forum=pysnmp-users>

The most significant changes against 3.0.0 revision include:

* A limited snmpy compatibility facility. Now you can run [some]
  SNMPY-based applications over PySNMP code. For more information
  on SNMPY package, see http://www.sf.net/projects/snmpy/
* __hash__() methods implelented for Asn1Object-based objects.
* A few bugs fixed.

Here is the source code:

http://osdn.dl.sourceforge.net/sourceforge/pysnmp/pysnmp-3.1.0.tar.gz

Thanks,
ilya

[pysnmp-users] ANN: the first beta of PySNMP 3.x has been released

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchives/forum.php?forum=pysnmp-users>

This is the first beta release of 3.x branch of PySNMP software, a SNMP
framework aimed at developing a highly extensible, modular and standards
compliant SNMP implementation purely in Python programming language.

Features of the 3.0.0 release:

* The ASN.1 code has been re-written to become a generic framework
  for building and handling ASN.1 data types.
* The BER code has been split off the ASN.1 code and re-written as
  a set of classes that can be mixed-into ASN.1 ones to provide
  BER functionality to the resultant object. The idea behind this
  design aims at allowing many concurrent ASN.1 objects serialization
  methods.
* A highly flexible, low-level interface to SNMP functionality has been
  introduced.
* A compatibility APIs to previous PySNMP versions have been added.
* Package documentation re-worked and extended.

The source code of pysnmp-3.x can be downloaded from SourceForge:

http://sourceforge.net/project/showfiles.php?group_id=14735&release_id=113926

Bug fixes and design improvements are very welcome! ;-)

Thanks,
ilya

[pysnmp-users] RE: decoding MAC values in snmptable.py

[Ilya E. <il...@gl...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchives/forum.php?forum=pysnmp-users>

I think that the exact data representation format for arbitrary OID is
normally provided by MIB. Since there is no MIB compiler bundled with
pysnmp at the moment, there seems to be no universal way to handle that.

Once you know the complete set of OIDs you are going to use in your
particular application, a workaround would be to hardcode values
representation rules based on OID, that is what John suggested.

To figure out the representation format for particular data type --
consult appropriate MIB.

-ilya

>From: John David <john@mn...>
>Subject: decoding MAC values in snmptable.py  
>
> Here's how I did it:
> 
> #
> import binascii

[pysnmp-users] decoding MAC values in snmptable.py

[John D. <joh...@mn...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchives/forum.php?forum=pysnmp-users>

Here's how I did it (from snmptable.py):

#
import binascii

# ... later, in the same code ...

    (answer, src) = client.send_and_receive(\
                    req.encode(community=args[1], encoded_oids=encoded_oids))
    # Attempt to decode SNMP response
    rsp.decode(answer)
    # Make sure response matches request (request IDs, communities, etc)
    if req != rsp:
        raise 'Unmatched response: %s vs %s' % (str(req), str(rsp))
    # Decode BER encoded Object IDs.
    oids = map(lambda x: x[0], map(asn1.OBJECTID().decode, \
                                   rsp['encoded_oids']))
    # Decode BER encoded values associated with Object IDs.
    vals = map(lambda x: x[0](), map(asn1.decode, rsp['encoded_vals']))

# ... various error checking and cruft skimming skipped over here ...

    # Print out results
    for (oid, val) in map(None, oids, vals):
        print oid + ' ----> ', binascii.b2a_hex(val)


.1.3.6.1.2.1.17.4.3.1.1.0.2.185.184.176.0 ---->  0002b9b8b000


snmptable.py gets the 'vals' list like this:
vals = map(lambda x: x[0](), map(asn1.decode, rsp['encoded_vals']))

The vals list ends up looking like this, for example:
vals:  ['\x00\x02\xb9\xb8\xb0\x00']


Each element in that list (in this case a list with one element), is a
non-packed sequence of 'raw' bytes.  Notice that the above looks like an
ethernet address, but when you print it, it comes out as various
non-printables.  By using bin2ascii.b2a_hex, you can print the hexadecimal
representation of the byte sequence.


Info on using snmp to get ethernet address table info can be found at:

http://www.cisco.com/warp/public/477/SNMP/cam_snmp.shtml

From the above url, the bad news is that you must query the 'CAM' or mac
address table for each lan on a cisco switch via snmp to get an accurate mac
address table, similar to what the ios commands 'sho cam dynamic', or 'sho
mac-address-table dynamic'would give you.


here is the oid to get vlan assignments on a cisco switch:
.1.3.6.1.4.1.9.9.68.1.2.2.1.2


To get that table, for example,

$snmpwalk my-switch.domain.name public .1.3.6.1.4.1.9.9.68.1.2.2.1.2
enterprises.9.9.68.1.2.2.1.2.1 = 9
enterprises.9.9.68.1.2.2.1.2.2 = 9
enterprises.9.9.68.1.2.2.1.2.3 = 9
enterprises.9.9.68.1.2.2.1.2.4 = 9
enterprises.9.9.68.1.2.2.1.2.5 = 9
enterprises.9.9.68.1.2.2.1.2.6 = 9
enterprises.9.9.68.1.2.2.1.2.7 = 9
enterprises.9.9.68.1.2.2.1.2.8 = 9
enterprises.9.9.68.1.2.2.1.2.9 = 9
enterprises.9.9.68.1.2.2.1.2.10 = 9
enterprises.9.9.68.1.2.2.1.2.11 = 9
enterprises.9.9.68.1.2.2.1.2.12 = 9
enterprises.9.9.68.1.2.2.1.2.13 = 9
enterprises.9.9.68.1.2.2.1.2.14 = 9
enterprises.9.9.68.1.2.2.1.2.15 = 9
enterprises.9.9.68.1.2.2.1.2.16 = 9
enterprises.9.9.68.1.2.2.1.2.17 = 9
enterprises.9.9.68.1.2.2.1.2.18 = 9
enterprises.9.9.68.1.2.2.1.2.19 = 9
enterprises.9.9.68.1.2.2.1.2.20 = 9
enterprises.9.9.68.1.2.2.1.2.21 = 9
enterprises.9.9.68.1.2.2.1.2.22 = 9
enterprises.9.9.68.1.2.2.1.2.23 = 9
enterprises.9.9.68.1.2.2.1.2.24 = 9
enterprises.9.9.68.1.2.2.1.2.25 = 9
enterprises.9.9.68.1.2.2.1.2.26 = 9
enterprises.9.9.68.1.2.2.1.2.27 = 9
enterprises.9.9.68.1.2.2.1.2.28 = 9
enterprises.9.9.68.1.2.2.1.2.29 = 9
enterprises.9.9.68.1.2.2.1.2.30 = 9
enterprises.9.9.68.1.2.2.1.2.31 = 9
enterprises.9.9.68.1.2.2.1.2.32 = 9
enterprises.9.9.68.1.2.2.1.2.33 = 9
enterprises.9.9.68.1.2.2.1.2.34 = 9
enterprises.9.9.68.1.2.2.1.2.35 = 9
enterprises.9.9.68.1.2.2.1.2.36 = 9
enterprises.9.9.68.1.2.2.1.2.37 = 9
enterprises.9.9.68.1.2.2.1.2.38 = 9
enterprises.9.9.68.1.2.2.1.2.39 = 9
enterprises.9.9.68.1.2.2.1.2.40 = 9
enterprises.9.9.68.1.2.2.1.2.41 = 9
enterprises.9.9.68.1.2.2.1.2.42 = 9
enterprises.9.9.68.1.2.2.1.2.43 = 39
enterprises.9.9.68.1.2.2.1.2.44 = 9
enterprises.9.9.68.1.2.2.1.2.45 = 40
enterprises.9.9.68.1.2.2.1.2.46 = 9
enterprises.9.9.68.1.2.2.1.2.47 = 9
enterprises.9.9.68.1.2.2.1.2.48 = 9
enterprises.9.9.68.1.2.2.1.2.50 = 2

snmptable.py also get the above info must fine, except the 'encoded_vals' part
of the rsp object isn't a raw sequence of bytes representing an ethernet
address - its just the vlan number.

The above was executed against a catalyst 2950G 48 port switch.  I'm putting
this info in this message, because finding the oid was a lot of work, so this
should save somebody out there a lot of time.

Hope this helps,

joh...@mn...

[pysnmp-users] decoding MAC values in snmptable.py

[John D. <joh...@mn...>]

A list for users of pure-Python SNMP framework <pysnmp-users.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/pysnmp-users>,
	<mailto:pys...@li...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchives/forum.php?forum=pysnmp-users>

Here's how I did it:

#
import binascii


    (answer, src) = client.send_and_receive(\
                    req.encode(community=args[1], encoded_oids=encoded_oids))
    # Attempt to decode SNMP response
    rsp.decode(answer)
    # Make sure response matches request (request IDs, communities, etc)
    if req != rsp:
        raise 'Unmatched response: %s vs %s' % (str(req), str(rsp))
    # Decode BER encoded Object IDs.
    oids = map(lambda x: x[0], map(asn1.OBJECTID().decode, \
                                   rsp['encoded_oids']))
    # Decode BER encoded values associated with Object IDs.
    vals = map(lambda x: x[0](), map(asn1.decode, rsp['encoded_vals']))

# ... various error checking and cruft skimming not included here ...

    # Print out results
    for (oid, val) in map(None, oids, vals):
        print oid + ' ----> ', binascii.b2a_hex(val)


.1.3.6.1.2.1.17.4.3.1.1.0.2.185.184.176.0 ---->  0002b9b8b000




the rsp object look like this:

{'bindings': {'value': {'encoded_oids':
['\x06\x13+\x06\x01\x02\x01\x11\x04\x03\x01\x01\x0
0\x02\x819\x818\x810\x00'],
                        'encoded_vals': ['\x04\x06\x00\x02\xb9\xb8\xb0\x00']}},
 'msg': {'value': {'community': 'imysts',
                   'pdu':
'\xa2*\x02\x01\x00\x02\x01\x00\x02\x01\x000\x1f0\x1d\x06\x13+\x0
6\x01\x02\x01\x11\x04\x03\x01\x01\x00\x02\x819\x818\x810\x00\x04\x06\x00\x02\xb9\xb8\xb0\x
00',
                   'version': 0}},
 'pdu': {'value': {'bindings':
'0\x1f0\x1d\x06\x13+\x06\x01\x02\x01\x11\x04\x03\x01\x01\x0
0\x02\x819\x818\x810\x00\x04\x06\x00\x02\xb9\xb8\xb0\x00',
                   'error_index': 0,
                   'error_status': 0,
                   'request_id': 0,
                   'tag': 'GETRESPONSE'}}}



snmptable.py gets the 'vals' list like this:
vals = map(lambda x: x[0](), map(asn1.decode, rsp['encoded_vals']))

The vals list ends up looking like this:
vals:  ['\x00\x02\xb9\xb8\xb0\x00']



Each element in that list (in this case a list with one element), is a
non-packed sequence of 'raw' bytes.  Notice that the above looks like an
ethernet address, but when you print it, it comes out as various
non-printables.  By using bin2ascii.b2a_hex, you can print the hexadecimal
representation of the byte sequence.

The bad news is that you must query the 'CAM' or mac address table for each
vlan on a cisco switch via snmp to get an accurate mac address table, similar
to what the ios commands 'sho cam dynamic', or 'sho mac-address-table dynamic'
would give you.

here is the oid to get vlan assignments on a cisco switch:
.1.3.6.1.4.1.9.9.68.1.2.2.1.2

To get that table, for example,

$snmpwalk myswitch public .1.3.6.1.4.1.9.9.68.1.2.2.1.2
enterprises.9.9.68.1.2.2.1.2.1 = 9
enterprises.9.9.68.1.2.2.1.2.2 = 9
enterprises.9.9.68.1.2.2.1.2.3 = 9
enterprises.9.9.68.1.2.2.1.2.4 = 9
enterprises.9.9.68.1.2.2.1.2.5 = 9
enterprises.9.9.68.1.2.2.1.2.6 = 9
enterprises.9.9.68.1.2.2.1.2.7 = 9
enterprises.9.9.68.1.2.2.1.2.8 = 9
enterprises.9.9.68.1.2.2.1.2.9 = 9
enterprises.9.9.68.1.2.2.1.2.10 = 9
enterprises.9.9.68.1.2.2.1.2.11 = 9
enterprises.9.9.68.1.2.2.1.2.12 = 9
enterprises.9.9.68.1.2.2.1.2.13 = 9
enterprises.9.9.68.1.2.2.1.2.14 = 9
enterprises.9.9.68.1.2.2.1.2.15 = 9
enterprises.9.9.68.1.2.2.1.2.16 = 9
enterprises.9.9.68.1.2.2.1.2.17 = 9
enterprises.9.9.68.1.2.2.1.2.18 = 9
enterprises.9.9.68.1.2.2.1.2.19 = 9
enterprises.9.9.68.1.2.2.1.2.20 = 9
enterprises.9.9.68.1.2.2.1.2.21 = 9
enterprises.9.9.68.1.2.2.1.2.22 = 9
enterprises.9.9.68.1.2.2.1.2.23 = 9
enterprises.9.9.68.1.2.2.1.2.24 = 9
enterprises.9.9.68.1.2.2.1.2.25 = 9
enterprises.9.9.68.1.2.2.1.2.26 = 9
enterprises.9.9.68.1.2.2.1.2.27 = 9
enterprises.9.9.68.1.2.2.1.2.28 = 9
enterprises.9.9.68.1.2.2.1.2.29 = 9
enterprises.9.9.68.1.2.2.1.2.30 = 9
enterprises.9.9.68.1.2.2.1.2.31 = 9
enterprises.9.9.68.1.2.2.1.2.32 = 9
enterprises.9.9.68.1.2.2.1.2.33 = 9
enterprises.9.9.68.1.2.2.1.2.34 = 9
enterprises.9.9.68.1.2.2.1.2.35 = 9
enterprises.9.9.68.1.2.2.1.2.36 = 9
enterprises.9.9.68.1.2.2.1.2.37 = 9
enterprises.9.9.68.1.2.2.1.2.38 = 9
enterprises.9.9.68.1.2.2.1.2.39 = 9
enterprises.9.9.68.1.2.2.1.2.40 = 9
enterprises.9.9.68.1.2.2.1.2.41 = 9
enterprises.9.9.68.1.2.2.1.2.42 = 9
enterprises.9.9.68.1.2.2.1.2.43 = 39
enterprises.9.9.68.1.2.2.1.2.44 = 9
enterprises.9.9.68.1.2.2.1.2.45 = 40
enterprises.9.9.68.1.2.2.1.2.46 = 9
enterprises.9.9.68.1.2.2.1.2.47 = 9
enterprises.9.9.68.1.2.2.1.2.48 = 9
enterprises.9.9.68.1.2.2.1.2.50 = 2

snmptable.py also get the above info must fine, except the 'encoded_vals' part
of the rsp object isn't a raw sequence of bytes representing an ethernet
address - its just the vlan number.


The above was executed against a catalyst 2950G 48 port switch.  I'm putting
this info in this message, because finding the oid was a lot of work, so this
should save somebody out there a lot of time.

Hope this helps,

joh...@mn...

[pysnmp-users] addendum to last email

[seth v. <sk...@ph...>]

Actually - I looked at a little deeper - the data I want is being handed
back in Macaddress format - which MIGHT be an octetstring  but I think
it might be its own format.

If I can sort out how Macaddress is packed I should be able to unpack
it, right?

thanks
-sv

[pysnmp-users] decoding a value

[seth v. <sk...@ph...>]

Hi,
 I just started you pysnmp 2.0.5 on a cisco 4006 switch.

I'm trying to read the cam table - I've got the tables I need to get the
data but when I read one of the tables I just get garbage - when I view
with it net-snmp snmptable I see the hex addresses of machines on my
network - I've tried using just snmpget to get a particular entry in the
table and it says the type of 'val' is a string - but I don't seem to
know how its packed so I can decode it.=20

anyone have any ideas about that?


-sv

[pysnmp-users] Re: problem with the example

[Ilya E. <il...@gl...>]

> Ok, sorry didn't read your mail right, I have tried snmpwalk.py und 
> snmpget.py.
> Were can I find the snmptable.py example ? It is not on the website.

It's in pysnmp distribution. I've not put it up on the web as the code is
very similar to snmpwalk.py.

-ilya

[pysnmp-users] Re: problem with the example

[Ilya E. <il...@gl...>]

Hi Felix,

> > * If you need to retrieve a SNMP "table", try using snmptable.py
> Doesn't do what I want, I use snmpwalk because I thought there is a solution 
> in it.
> snmptable is just a smaller version without the wile and so on, isn't it ?

It still looks like snmptable.py is what you need. The tool is exactly the
same code as snmpwalk.py BUT it stops when next OID is lexicographically
larger than previous. That is:

[ilya@cray examples]$ ./snmptable.py cisco.st.net public .1.3.6.1.2.1.2.2.1.19
.1.3.6.1.2.1.2.2.1.19.1 ---> 1204
.1.3.6.1.2.1.2.2.1.19.2 ---> 160
.1.3.6.1.2.1.2.2.1.19.3 ---> 0
.1.3.6.1.2.1.2.2.1.19.4 ---> 85
...
.1.3.6.1.2.1.2.2.1.19.1473 ---> 0
.1.3.6.1.2.1.2.2.1.19.1474 ---> 0
.1.3.6.1.2.1.2.2.1.19.1475 ---> 0
[ilya@cray examples]$

What exactly does not work for you with snmptable.py?

-ilya