Updating mibs/rfclist programmatically

For years, I've crawled over Internet-Drafts and RFCs to find things that look like MIB modules, to construct my set of pages at http://www.icir.org/fenner/mibs/ . It occurred to me, after seeing that some of the MIBs that Net-SNMP distributes are out of date, that this info could be relatively ea

Re: MSTP MIB?

On Mon, Sep 22, 2008 at 7:07 AM, Chris Nelson <[EMAIL PROTECTED]> wrote: > Is there such a thing? I'm embarking on an 802.1Q-2005 implementation > and can't find any MIB text. Surely MSTP can be monitored and > administered via SNMP. Can't you use multiple logical entities implementing dot1dStp

Re: Review Request: Don't splat /var/agentx/master when starting an extraneous snmpd

On Thu, Apr 2, 2009 at 2:24 PM, Wes Hardaker wrote: >>>>>> On Thu, 02 Apr 2009 18:49:45 -, "Bill Fenner (gmail)" >>>>>> said: > BF> --- > BF> This is an automatically generated

Implement IMPLIED in MIB hash

I have a mib2c-based project that needs to be able to know whether an INDEX is IMPLIED - this allows the decision to use ASN_PRIV_IMPLIED_OCTET_STR to be: if ( ($node->{type} == "OCTETSTR") && (@{ $node->{ranges} } == 1) ) { my $range = $node->{ranges}->[0]; if ( $range->{high} == $

Performance and correctness problems in Python bindings

Last summer, I posted a patch for the Python bindings at https://sourceforge.net/tracker/index.php?func=detail&aid=2074817&group_id=12694&atid=112694 This fixes the problem that it's impossible to use the Python bindings to fetch an object with more than one sub-ID in its INDEX, e.g., it works fo

Re: Performance and correctness problems in Python bindings

We did some further analysis of the regexp. We found: 1. Replacing \w+(?:\-*\w+)+ with \w[-\w]+ turns 5 hours into a few ms. It still varies based on the length of the string, but it maxed out at 100ms for hrSWInstalledLastUpdateTime. 2. This regexp is different from the one in the perl module

Re: problem with mib2c

On Fri, Jan 8, 2010 at 12:20 PM, Hemant Soundankar wrote: >  I am implementing RFC 1697 for postgres. > While generating the code template for rdbmsSrvParamTable via mib2c , no > code/ functions for inserting proper value for rdbmsSrvParamName is > generated. That's because it's not-accessible, s

Changing the signature of snmp_parse_args()

In order to fix bug 2941983 (snmpd enables stderr logging if trapsess is configured in snmpd.conf), I'd like to change the signature of snmp_parse_args(): --- snmp_parse_args.c (revision 18006) +++ snmp_parse_args.c (working copy) @@ -195,13 +195,15 @@ /* * This method does the real work fo

Re: Changing the signature of snmp_parse_args()

On Fri, Jan 29, 2010 at 11:06 AM, Dave Shield wrote: > On 29 January 2010 15:40, Bill Fenner wrote: >> My question is: is it sufficient to change all of the callers in the >> net-snmp source tree, or should I call the function with the new >> signature snmp_parse_args2(

Re: Changing the signature of snmp_parse_args()

On Fri, Feb 12, 2010 at 2:45 PM, Wes Hardaker wrote: >> On Fri, 29 Jan 2010 16:06:13 +, Dave Shield >> said: > > DS> I'd also suggest a more general "flags" parameter, rather than one > DS> specifically for logging.  (Just in case we need to extend the behaviour > DS> of this routine

Re: rfc: changing snmpwalk default starting point

On Tue, Mar 9, 2010 at 9:35 AM, Robert Story wrote: > There was some discussion in the IRC channel this morning about the default > behavior of snmpwalk biting newbies. They will often run snmpwalk without > specifying a starting point, and by default snmpwalk will start the walk on > the mib-2 br

broken in 5.6?

I updated net-snmp to 5.6, and all of my net-snmp consumers broke. It looks like (included from accidentally used "netsnmp_transport *" instead of "struct netsnmp_transport_s *": NETSNMP_IMPORT int netsnmp_sess_config_and_open_transport(netsnmp_session *in_session,

Re: broken in 5.6?

On Wed, Oct 13, 2010 at 12:16 PM, Wes Hardaker wrote: > > BF> I updated net-snmp to 5.6, and all of my net-snmp consumers broke.  It > BF> looks like (included from > BF> accidentally used "netsnmp_transport *" > BF> instead of "struct netsnmp_transport_s *": > > Hi Bill! > > Though I think your

Re: broken in 5.6?

On Wed, Oct 13, 2010 at 12:23 PM, Dave Shield wrote: > On 13 October 2010 17:16, Wes Hardaker wrote: >> BF> I updated net-snmp to 5.6, and all of my net-snmp consumers broke.  It >> BF> looks like (included from >> BF> accidentally used "netsnmp_transport *" >> BF> instead of "struct netsnmp_tr

Re: 5.6.1.pre1 released for testing

On Wed, Dec 1, 2010 at 1:42 AM, Wes Hardaker wrote: > We've released Net-SNMP 5.6.1.pre1 which contains bug fixes over the 5.6 > release.  Please give it a whirl and let us know how it works for you! My TCP-MIB regression test passes now, the bug described in patch 3085295 is gone. This is, of c

Prototypes for debug functions?

I would like to request that the prototypes for void debug_print_registered_tokens(void); int debug_enable_token_logs(const char *); int debug_disable_token_logs(const char *); be exposed in a net-snmp header file. I realize they were probably put there for use from gdb, but in our system we alr

Net-snmp 5.6 uses the wrong interface on a multi-homed box with asymmetric routing

First, a picture:                     10.1.1.0/24                -                 |                 | eth0 manager router1           router2                 |                 | eth1                -                     10.2.2.0/24 The manager is singl

Re: Net-snmp 5.6 uses the wrong interface on a multi-homed box with asymmetric routing

On Mon, Dec 13, 2010 at 5:21 PM, Robert Story wrote: > On Mon, 13 Dec 2010 11:16:54 -0500 Bill wrote: > BF> I'm not sure what to suggest as a solution.  For those who use > BF> net-snmp as the agent on a router, where asymmetric routing like this > BF> is very common, this black hole for the respo

Re: Net-snmp 5.6 uses the wrong interface on a multi-homed box with asymmetric routing

On Wed, Dec 15, 2010 at 11:26 AM, Robert Story wrote: > On Wed, 15 Dec 2010 10:28:05 +0100 Jan wrote: > JS> The problem is not to get this information, the real > JS> challenge is to do it effectively, i.e. cache them somewhere, refresh > JS> the cache sometimes, look up in the cache effectively,

Re: Net-snmp 5.6 uses the wrong interface on a multi-homed box with asymmetric routing

On Thu, Dec 16, 2010 at 4:26 AM, Jan Safranek wrote: > On 12/15/2010 05:26 PM, Robert Story wrote: >> >> JS>  The problem is not to get this information, the real >> JS>  challenge is to do it effectively, i.e. cache them somewhere, refresh >> JS>  the cache sometimes, look up in the cache effecti

Net-SNMP agentx master/subagent subtly broken in Net-SNMP 5.6

Hi, Our system uses Net-SNMP on both ends of the AgentX master/subagent relationship. With Net-SNMP 5.5, everything was very stable. With Net-SNMP 5.6, the subagent occasionally reconnects after a period of not being able to serve its objects. The subagent tracing looks like it stops getting re

Re: Net-SNMP agentx master/subagent subtly broken in Net-SNMP 5.6

On Mon, Dec 20, 2010 at 7:42 PM, Wes Hardaker wrote: >> On Sat, 18 Dec 2010 16:43:56 -0500, Robert Story >> said: > > BF> (I've already diffed agent/mibgroup/agentx between 5.5 and 5.6, it's > BF> almost all cleanups that can't explain this; there's something more > BF> subtle going on s

Re: Net-snmp 5.6 uses the wrong interface on a multi-homed box with asymmetric routing

On Wed, Dec 15, 2010 at 4:28 AM, Jan Safranek wrote: > On 12/13/2010 11:21 PM, Robert Story wrote: >> >> On Mon, 13 Dec 2010 11:16:54 -0500 Bill wrote: >> BF>  I'm not sure what to suggest as a solution.  For those who use >> BF>  net-snmp as the agent on a router, where asymmetric routing like th

Re: CFV: patch from Bill Fenner for agentx oids with multiple varbinds

On Wed, Dec 22, 2010 at 3:28 AM, Dave Shield wrote: > On 22 December 2010 04:42, Wes Hardaker > wrote: >>  - 5.6.1 (now in rc1) > > Hmm   this isn't really a show-stopper, so strictly speaking > shouldn't be applied. It's somehow a regression against net-snmp 5.5, so IMO it is more importan

Re: CFV: patch from Bill Fenner for agentx oids with multiple varbinds

On Dec 22, 2010, at 1:55 PM, Dave Shield wrote: > Maybe we need to review our definitions of "show stopper" at > next month's admin meeting. One proposal for an addition could be regressions from the previous major release. I upgraded our system from 5.5 to 5.6 and found at least 3 regressions (t

Re: Prototypes for debug functions?

On Dec 24, 2010, at 11:52 AM, Magnus Fromreide wrote: > This functionality is already exposed using SNMP. Not for a subagent, which is my use case. Bill -- Learn how Oracle Real Application Clusters (RAC) One Node al

Re: CFV: patch from Bill Fenner for agentx oids with multiple varbinds

On Dec 27, 2010, at 12:12 PM, Wes Hardaker wrote: >>>>>> On Sun, 26 Dec 2010 08:42:14 -0500, Bill Fenner said: > > BF> One proposal for an addition could be regressions from the previous > BF> major release. I upgraded our system from 5.5 to 5.6 and f

Re: My Subagent is not working after some long time

Prakash, There was a bug in earlier versions where if you tried to start snmpd while it was already running, it would overwrite the old agentx socket before failing to open the udp transport and exiting. This would mean that if the subagent ever got disconnected, it could never reconnect. It's k

Re: My Subagent is not working after some long time

On Mon, Jan 10, 2011 at 1:27 AM, Prakash wrote: > In which version this bus was fixed? I don't know. I know the bug is not present in 5.6. Bill -- Gaining the trust of online customers is vital for the success of any

mibgroup/disman/nslookup should be omitted, not made read-only, for NETSNMP_NO_WRITE_SUPPORT

I see disman/nslookup was updated recently to make lookupCtlTable read-only for NETSNMP_NO_WRITE_SUPPORT. This seems odd, since the only way to use this MIB is to write rows into the lookupCtlTable. If there's a provision to exclude modules entirely for NETSNMP_NO_WRITE_SUPPORT, I think that's a

Re: mib2c: varbinds: enums

On Fri, Mar 11, 2011 at 5:05 PM, Eric Smith wrote: > I'm trying to access enums for varbinds and I end up with error saying > "uninitialized value in hash element" > > Code: > @foreach $noti notifications@ >   @foreach $varb varbinds@ > @if $varb.enums@ >    ( Here's a horrible thing to

not installed via "make install"?

Hi, I tried a couple of recent svn checkouts (most recently, r20159), and found that net-snmp-features.h isn't installed via "make install": In file included from /Users/fenner/tmp/include/net-snmp/types.h:421, from /Users/fenner/tmp/include/net-snmp/definitions.h:22,

Re: not installed via "make install"?

On Mar 26, 2011, at 4:02 PM, Magnus Fromreide wrote: > On Sat, 2011-03-26 at 14:08 -0400, Bill Fenner wrote: >> Hi, >> >> I tried a couple of recent svn checkouts (most recently, r20159), and >> found that net-snmp-features.h isn't installed via "make insta

Re: 5.7.pre2 available for testing (last pre-release)

On Thu, Aug 18, 2011 at 4:15 PM, Wes Hardaker < harda...@users.sourceforge.net> wrote: > > The 5.7.1.pre2 pre-release is now available for testing: > > > https://sourceforge.net/projects/net-snmp/files/net-snmp/5.7.1-pre-releases/ > This is a pretty lame bug report, because I don't understand any

Re: 5.7.pre2 available for testing (last pre-release)

On Mon, Aug 22, 2011 at 4:20 PM, Wes Hardaker < harda...@users.sourceforge.net> wrote: > >>>>> On Mon, 22 Aug 2011 14:12:07 -0400, Bill Fenner > said: > > BF> That .so is not installed by 5.7.1pre2. > > There was one version of perl on one architectur

Re: Getting snmpd to respond to DISMAN-PING-MIB requests

On Fri, Sep 2, 2011 at 3:08 PM, Doug Manley wrote: > NB> I'm not that much for numeric OID's, but I have this script > > I took the script and added a little bit of "-x" fun so that you can > see what it's doing and where it's getting the errors. > > It can do an snmpget on sysDescr.0 just fine w

clientaddr should not be premib

Hi, In commit 6a841d56518a727196e66b59e7e5b3304f54b0a1, Wes changed s/netsnmp_ds_register_config/netsnmp_ds_register_premib/ for the "snmp", "clientaddr" token. This breaks changing the clientaddr with a HUP. I guess the advantage of having it be premib is that it can be specified in the config

Net-snmp 5.7 client apps default to 0 retries

Hi, I finally discovered the cause of some apparent infrastructure flakiness after upgrading to net-snmp 5.7: we have some PDUs that end up requiring retries to accept an SNMP set. 5.7 only sends a single request and does not retry, due to a bug in the code that allows the number of retries to be

Re: Problem with installing net-snmp package 5.6.1.1 on UBUNTU version: 11.04

Rename the "SNMP_ Related" directory to not contain a space. Bill On Fri, Nov 4, 2011 at 11:02 AM, prashant n wrote: > Hi, > > The problem is that when we trying to install net-snmp package  5.6.1.1 on > UBUNTU version: 11.04,we are getting an error message as below, > > Installation steps: Ru

Re: Problem with cross-compiling net-snmp package 5.6.1.1

Prashant, There is a whole web site dedicated to making it easier to build embedded Linux systems: http://buildroot.net/ . Their system includes net-snmp 5.7.1. Maybe you can find a way to leverage their tools to accomplish your goal. Bill (I don't know anything specific about buildroot, jus

Re: Problem with cross-compiling net-snmp package 5.6.1.1

Prashant, There is a whole web site dedicated to making it easier to build embedded Linux systems: http://buildroot.net/ . Their system includes net-snmp 5.7.1. Maybe you can find a way to leverage their tools to accomplish your goal. Bill (I don't know anything specific about buildroot, just

Re: Permission denied to run libnetsnmp.so

On Thu, Dec 15, 2011 at 4:42 AM, 梅亮 wrote: > Hi Net-SNMP coders, > > When I am compiling snmpd and run it, I encountered the following errmsg: > > [root@limei net-snmp-5.5.1]# snmpd > > snmpd: error while loading shared libraries: libnetsnmp.so.20: cannot enable > executable stack as shared object

Error handling path in agentx master and subagents

I've stumbled across some really peculiar error handling problems in the agentx implementation in net-snmp 5.7.1, and was wondering if anyone was familiar enough with this code to figure out what's going on. 1. VarBinds of an unrecognized type get *stripped* from SET requests aimed at the subagent

Re: how to make net-snmp agent only be responsive to a certain management station

On Sun, Dec 18, 2011 at 10:04 PM, 梅亮 wrote: > Hi Net-SNMP coders, > > Would you please help me out with the following question? > > If I want my net-snmp agent only be responsive to a snmp packet from a > certain management station with a fixed IP address, what is the most > efficient way to do th

Re: agentx master agent not responding after an agentx subagent SET timeout

Hi Renjith, I came across this behavior in a different context and filed a bug: http://sourceforge.net/tracker/?func=detail&aid=3456337&group_id=12694&atid=112694 ("Certain agentx failures can result in apparent agent hang"). I was unaware of the "set times out" way to get into this state - ple

Re: agentx master agent not responding after an agentx subagent SET timeout

not seen. So I think as you stated in your bug, netsnmp_processing_set is > not cleared in this path causing further request queued. > > > > [Note: NET-SNMP version:  5.6.pre3] > > > > -renjith > > > > From: Bill Fenner [mailto:fen...@gmail.com] > Sent:

Re: agentx master agent not responding after an agentx subagent SET timeout

maintenance > can comment authentically like whether the two ideas that you have suggested > will fix the issue or like net-snmp 5.7.1 have done a fix on this already. > > thank you, > -renjith > > .-Original Message- > .From: Bill Fenner [mailto:fen...@gmail.com]

Re: dynamic table Table creation

The agent libraries can be used from C++, so what we do is build the master agent using a C compiler and build our subagent with g++, and serve our MIBs using AgentX. In fact, if you build libnetsnmpagent with a C compiler, you could probably then build your modules plus the snmp main loop and

Re: agentx master agent not responding after an agentx subagent SET timeout

I haven't had any luck invoking the failure mode with my misbehaving subagent. Both the disconnect and the timeout end up going through netsnmp_remove_delegated_requests_for_session(), which then causes check_delayed_request() to call netsnmp_wrap_up_request() which resets netsnmp_processing_set.

Re: [PATCH v2] Make Net-SNMP independent of system clock adjustments

I wrote regression tests for my two reported cases: 1. sysUpTime gets strange if the clock changes after net-snmp starts; 2. subagent fails to reconnect until the clock catches back up if the clock changes after it sets its reconnect timer. I haven't yet gotten the test for #2 to fail in my previ

Re: CFV: fix bug 3532090 and high-index crashes in hrDeviceDescr

On Tue, Jun 19, 2012 at 5:23 PM, Dave Shield wrote: > On 18 June 2012 23:10, Wes Hardaker wrote: >> Patch #1 is simple and just increases the shift variable allowing >> for 16 bits of index values instead of the previous 8 > > +1 > > >> Patch #2 is a bit more complex and creates new integer based

Re: CFV: fix bug 3532090 and high-index crashes in hrDeviceDescr

On Tue, Jul 3, 2012 at 4:48 PM, Niels Baggesen wrote: > Den 02-07-2012 20:22, Bill Fenner skrev: > > >> A related question: isn't it safer to pretend to the compatibility >> functions that a given interface with ifIndex > 16 bits (or 15 bits) >> doesn't ex

Relative paths in /var/net-snmp/mib-indexes/*?

Hi, I was poking around at an unrelated problem, and came across this: ==> var/net-snmp/mib_indexes/1 <== DIR /usr/share/snmp/mibs [ obviously fine ] ==> var/net-snmp/mib_indexes/2 <== DIR . ==> var/net-snmp/mib_indexes/4 <== DIR .. I think that the code that outputs the mib_indexes stash

Re: CFV: Handle TimeTicks when pretty-printing OID (patch 3560473)

Personally I would argue that it's a bug that strings in the OID are printed as numbers. While handling of TimeTicks is new, it is the most straightforward way to fix the bug of printing strings as numbers. Bill On Aug 28, 2012, at 7:46 PM, Wes Hardaker wrote: > Niels Baggesen writes: >

Re: CFV: Handle TimeTicks when pretty-printing OID (patch 3560473)

On Aug 29, 2012, at 1:31 AM, Niels Baggesen wrote: > Den 29-08-2012 01:46, Wes Hardaker skrev: >> Well, I was just about to add a "+1" and apply it... But there is one >> problem, it's a new feature technically... > > No, it adds a case that was forgotten in the original implementation. It > i

Re: X-agent performance.

[ Removing -users ] On Dec 4, 2012, at 10:57 PM, "Zheng, Wenjie (Barclay)" wrote: > Dear All, > > By studying and testing SNMP performance these days, I found that X-agent is > not efficient as I expected: > SNMP sub-agent mode slows down the total time of getbulk, by change the SNMP > agen

Re: agentx/master: unexpected callback op 4 message

On Tue, Jan 22, 2013 at 11:16 AM, Claus Klein wrote: > Hi, > > when I stop, and start the agentx master agent again while the agentx client > is still running, I got this messages: > > NET-SNMP version 5.4.4 > agentx/master: unexpected callback op 4 > agentx/master: handle pdu (req=0x166,trans=0x0

Should agent_process_and_check() handle fd_event_manager-registered file descriptors?

I'm writing a MIB module which also sends and receives packets, so in order to receive them and not block the main loop I'm using register_readfd(). For testing, I'm using standalone AgentX subagent mode, with the default main loop of while ( 1 ) { agent_process_and_check( 1 ); } However, it

mfd: table_interface.c: user-serviceable or not?

Given the comments that mfd adds at the top of table_interface.c: * *** NOTE NOTE NOTE NOTE NOTE NOTE NOTE NOTE NOTE NOTE NOTE NOTE *** * *** THIS FILE DOES NOT CONTAIN ANY USER EDITABLE CODE. *** I expected that I wouldn't have to mess with it. However, in _table_container_row_sa

Re: snmpnetstat does not show ipv6 addresses, routes, and services

On Sun, Feb 24, 2013 at 9:13 AM, Niels Baggesen wrote: > Den 23-02-2013 20:54, Claus Klein skrev: >> is it planed to extend snmpnetstat to show ipv6 address info, ipv6 >> routing table entries, and ipv6 active sockets info too? >> >> It seems that the current version show only the ipv4 related inf

Re: snmpnetstat does not show ipv6 addresses, routes, and services

On Mar 2, 2013, at 3:44 AM, Niels Baggesen wrote: > On Thu, Feb 28, 2013 at 10:29:40AM -0500, Bill Fenner wrote: >> What do you think about updating the "old" support to use >> ipCidrRouteTable, since ipRouteTable has been obsolete since 1997, and >> dropping su

Memory leaks in MFD-generated code?

I've been trying to track down some memory leaks in my pingCtlTable implementation. After I fixed the bugs from not freeing my own data structures, valgrind points out two "definitely lost" chunks: ==7467== 17,568 bytes in 6 blocks are definitely lost in loss record 1,075 of 1,086 ==7467==at

Re: Cannot find a place in code, where traps are being dumped to stdout

On Wed, Jun 26, 2013 at 10:26 PM, Oleg Gritsak wrote: > Hello! > > First of all, thank you for such a great software made under opensource > license! > Didn't want to bother anyone, but already spent several days on this > problem without > any positive result. :( Maybe someone will make a hint, p

Re: net-snmp fails to create entries with a zero in the OID

On Thu, Jul 11, 2013 at 6:39 AM, zhuyj wrote: > Hi, Magnus && Niels > > Thanks for your help! I am testing your patch. This patch can solve the > most problems. > But Ihave a question, is this a bug? I mean that '0' in snmpAdminString > is a bug, or not? > Would you like to discuss this with me?

Re: MIB compile problem

On Thu, Jul 25, 2013 at 8:49 AM, Fulko Hew wrote: > > > On Thu, Jul 25, 2013 at 2:24 AM, Andrew Hood wrote: > >> On 19/07/2013 5:47 AM, O'hara, Joseph (Contractor) wrote: >> >> > Hi net-snmp-coders, >> > I am having trouble compiling the NET-SNMP-EXTEND-MIB which I just >> > downloaded >> > from

Re: low? octet string limit on AgentX transactions

On Mon, Oct 28, 2013 at 7:01 AM, Fulko Hew wrote: > I was bit recently by a design limitation on the length of OCTET STRINGS > as received and processed by AgentX. > > ... > Comments? > I think the dual to this problem has also been reported at https://sourceforge.net/p/net-snmp/bugs/2489/ - a S

Re: Remove IF-MIB and IP-MIB

[[ I've removed net-snmp-users, since this question is more appropriate for coders. Note that http://www.net-snmp.org/support/mailinglists.html says "Please do NOT post the same message to both the user and coders lists", and has other useful information about how to politely ask for help ]] On M

Re: Help! Convert OCTET string to READABLE string

This is a limitation in the execute / pass / pass_persist infrastructure: the truncation happens when creating the output to send to the script. setPass() sets up the command string and the arguments into passthru->command, which is a statically allocated buffer. setPassPersist() has the same prob

Re: Query related to memory leaks and read errors in snmptrap command

On Thu, Feb 6, 2014 at 7:11 AM, Chaitanya Gulhane < chaitanya.gulh...@gmail.com> wrote: > > Hi, > > I was using net-snmp 5.6.2 and have built the code and installed it. > > I was using valgrind over snmptrap command. > Following is the command I ran. > > valgrind --leak-check=full --log-file=log_s

Re: Query related to memory leaks and read errors in snmptrap command

ok at http://www.net-snmp.org/ it says "Current release: 5.7.2" at the top of every page. Bill > > Regards, > Chaitanya. > > > On Tue, Feb 11, 2014 at 1:08 AM, Bill Fenner wrote: > >> On Thu, Feb 6, 2014 at 7:11 AM, Chaitanya Gulhane < >> chaitanya.g

Re: Memory Leak bug in net-snmp-5.6.2.1, in the procedure snmpv3_parse()

It's the caller's responsibility to free that memory (and snmp_free_pdu() will). Bill On Wed, Mar 12, 2014 at 11:28 AM, John Zavgren wrote: > Nevertheless... > The memory that is allocated at lines 3715 and 3724 is leaked at line 3795. > > Allocations: > > * */* > *sec_params = data;

Re: net-snmp dumps core when snmpd.conf contains an incorrect logmatch parameter regex

Hi Mohan, Thanks for the patch. Can you please upload it at http://www.net-snmp.org/patches/ so that it doesn't get lost? Thanks, Bill On Tue, Mar 18, 2014 at 8:11 AM, Mohan Gorai wrote: > Hi, > > On Solaris, whenever there is a wrong regex as logmatch parameter, when > the corresponding s

Re: Bugfix for bulk gets in SNMP.pm

Hi Steinar, Thanks for the patch. In order to make sure that we don't lose track of it, could you please upload it to http://www.net-snmp.org/patches/ ? Thanks, Bill On Wed, Apr 9, 2014 at 9:25 AM, Steinar H. Gunderson wrote: > On Tue, Apr 01, 2014 at 01:41:12PM +0200, Steinar H. Gunderso

Re: error showing SNMPWALK on the last table of a view

This kind of behavior is usually due to using an inefficient lookup, especially combined with using the table_iterator helper. See the "Con Mitigation" heading of http://www.net-snmp.org/wiki/index.php/Table_iterator if you are using the table_iterator helper. Bill On Wed, Jul 2, 2014 at 11:0

Optimizing the large_fd_set code

When doing some performance measurement, I found that a surprising amount of time was spent in netsnmp_large_fd_set_resize(). It turns out that a newly-created large fd set often has an lfs_setsize of zero, and a desired size of FD_SETSIZE, and the current code zeroes out the bits in between lfs_s

Re: Memory leak in 5.7.2.1, _sess_process_packet?

On Tue, Jun 24, 2014 at 3:47 AM, Sean Burke wrote: > Hi All, > > I am writing to ask about this code in 5.7.2.1 snmp_api.c at line 5423. > I am specifically concerned with the constraint pdu->command == > SNMP_MSG_TRAP2, > which appears to interfere with freeing the securityStateRef for request >

Re: need help for generating ,C file from MIB file

On Wed, Jul 2, 2014 at 6:32 PM, Gaurav Kasliwal wrote: > Hi, > > > > I have created a MIB file > > It is present at path : > > > > [gkasliwal@svl-junos-d072 > /b/gkasliwal/14.2_ib_psd/src/junos/shared/mibs]$ ls | grep jnx-bgpmib > > jnx-bgpmib2.mib > > jnx-bgpmib3.mib > > jnx-bgpmib3_tc.mib > >

Re: Optimizing the large_fd_set code

On Tue, Jul 8, 2014 at 1:17 AM, Magnus Fromreide wrote: > On Mon, Jul 07, 2014 at 05:33:53PM -0400, Bill Fenner wrote: > > When doing some performance measurement, I found that a surprising amount > > of time was spent in netsnmp_large_fd_set_resize(). It turns out that a &g

Re: Memory leak in 5.7.2.1, _sess_process_packet?

ieve > that GET-NEXT and GET-BULK would show the problem also. > I am running an agent session that receives these requests. > > -SEan > > > > On Jul 8, 2014, at 11:10 AM, Bill Fenner wrote: > > On Tue, Jun 24, 2014 at 3:47 AM, Sean Burke > wrote: > >> Hi A

Re: SNMP TLS snmpget error

Did you configure the certificates properly? In particular, did you configure the server with the private key? Since you're using the fingerprints from the tutorial, but using your local server instead of test.net-snmp.org, where did you get the private key? It's not part of the published set of

Re: SNMP TLS snmpget error

again i thank you for giving response. I've been waiting for some > response. > > Thanks, > sandhya > > > > > On Thu, Jul 24, 2014 at 5:44 PM, Bill Fenner wrote: > >> Did you configure the certificates properly? In particular, did you >> configure the

Re: CVE-1999-0516

On Fri, Jul 25, 2014 at 11:32 AM, Matthew Maldonado wrote: > I'm having a hard time finding resources on the web to > resolve CVE-1999-0516. Would you happen to have any information on how to > resolve this issue for the Linux side? > Configure all of your management systems and SNMP agents to u

Re: snmptable netSnmpIETFWGTable: No entries

This is a bug in the perl script. It does a "poor-man's" lexicographical ordering, only handling objects that it knows about. This works for a walk, but does not work for the (valid) request that snmptable makes. The easiest fix would be for the agent to learn that $next_oid{netSnmpIETFWGEntry.0

Re: problem in SNMPSET on table with string data type columns

I've had trouble with multiple requests in ordered table_iterator tables. I have no idea if that's what you're using, since you didn't give much info about your use case. I just worked around it by injecting the serialize handler: int result = netsnmp_register_table_iterator( reg, iinfo );

Re: problem in SNMPSET on table with string data type columns

pset). > > Could you please advise what could be the issue here? > > Thanks and regards > S Sarath > > > > > > On Wednesday, September 3, 2014 7:38 PM, Bill Fenner > wrote: > > > I've had trouble with multiple requests in ordered table_iterator tabl

Re: MFD compiled extension agent: snmpget returns "Timeout: No Response from localhost:1161."

On Fri, Sep 5, 2014 at 1:57 AM, Kenneth Hung wrote: > netsnmp_udp_getSecName: no com2sec entries < Am I > missing something here in the conf file? > Yes - despite your subject line mentioning extension agent, you're running a full agent, meaning that you have to configure it

Re: SNMPv3 Trap Question

On Wed, Sep 3, 2014 at 11:10 AM, Gary Dunlap wrote: > Hi. I’m having trouble with some of the SNMPv3 traps sent by snmpd being > rejected, and I think it’s because of the SNMPv3 replay protection > mechanism. > > > > It looks like when some event causes a reconfigure of snmpd > Can you say more

Re: does net-snmp support AES-128 & SHA96

net-snmp supports RFC3414 (which defines HMAC-SHA-96) and RFC3826 (which defines CFB128-AES-128). Bill On Tue, Sep 23, 2014 at 6:35 AM, sarath azad wrote: > Hello, > > Could you please tell if net-snmp supports AES-1

Re: problem in SNMPSET on table with string data type columns

e > called for each of them separately, rather than a single handler function > being called passing all the varbinds in input SNMPGET/SET query? > > thanks and regards > S Sarath > > > On Thursday, September 4, 2014 5:51 PM, Bill Fenner > wrote: > > > The only adv

Re: snmptrapd not processing v3 traps

On Fri, Oct 3, 2014 at 11:06 AM, Jason Pope wrote: > Hi, > > > > Has anyone successfully set up snmptrapd to accept v3 traps from a device ? > > > > I have tried to set it up, but I see nothing in the log when the device > sends v3 traps. When configured to send v2 traps you can see them in the

Re: snmptrapd not processing v3 traps

o.com] > Sent: 06 October 2014 10:34 > To: Jason Pope; 'Bill Fenner' > Cc: net-snmp-coders@lists.sourceforge.net; > net-snmp-us...@lists.sourceforge.net > Subject: RE: snmptrapd not processing v3 traps > > Jason, > > Would you mind expanding on this. Would yo

Re: Raises the netsnmp_remove_delegated_requests_for_session method of the snmpd code dump.

Please try net-snmp 5.7.3-rc1 . The patches you linked to are all present in that release candidate. Bill On Tue, Sep 30, 2014 at 1:02 AM, lijian...@rayootech.com < lijian...@rayootech.com> wrote: > Hi Snmp group > > > http://sourceforge.net/p/net-snmp/mailman/message/28799205/ > (NET-SNMP ve

Re: bug system.c - Solaris

On Tue, Sep 16, 2014 at 11:44 AM, Maria Teresa Eibe wrote: > Hi, > > I’ve noticed that there seems to be a bug in the code of system.c, > function netsnmp_os_prematch, if net-snmp is to be run on Solaris. The > problem would be the line: > > > > if ( 0 != uname(&utsbuf)) > >return -1; > > >

Re: [PATCH] Let Perl access the security engine ID

Hi Steinar, Thanks for the patches. Could you please submit them at http://www.net-snmp.org/patches/ so that we don't lose track of them? Thanks, Bill On Sun, Oct 19, 2014 at 4:29 PM, Steinar H. Gunderson < sgunder...@bigfoot.com> wrote: > Hi, > > When using SNMP::Session->new() without a S

Re: [PATCH] Let Perl access the security engine ID

On Mon, Oct 20, 2014 at 12:06 PM, Steinar H. Gunderson < sgunder...@bigfoot.com> wrote: > On Sun, Oct 19, 2014 at 09:05:05PM -0400, Bill Fenner wrote: > > Thanks for the patches. Could you please submit them at > > http://www.net-snmp.org/patches/ so that we don

How can the community help with patches? (Was Re: [PATCH] Let Perl access the security engine ID)

On Thu, Oct 23, 2014 at 9:05 AM, Michael Bunk wrote: > > As with most open source projects, committers get more and fewer cycles > to > > look at contributions. We're trying to make an effort in the next couple > > of months to get the queue down. > > Despite understanding your reasoning and lif

Re: Changing community names on the fly

Hi Vesa, I think the reason that you aren't getting any response is that most people who integrate net-snmp into their system use the configuration file and SIGHUP / snmpd restart to load changes. With SIGHUP the authentication info gets cleared out and re-read from the config file. If you're cr

Re: agentX++ interoperability with NET-SNMP version 5.7.3.rc1

On Sat, Nov 22, 2014 at 5:34 AM, Claus Klein wrote: > Hi, > > I have tested net-snmp as master agent with current agentpp subagent. > The most important problem I found is: > > > The row destroy at subagent does not work together with net-snmp master > agent! > > > The agentX++ subagent send the

Re: Configuration and --with-out-mib-modules

On Wed, Feb 4, 2015 at 8:09 PM, David Hauck wrote: > Hi, > > The 'configure' script has the following text: > > " --with-out-mib-modules="list" Compile without these mib modules. > > Default mib modules compiled into the agent (which can be removed): > > mibII support for

Re: getting shared objects to read configuration in snmpd.conf

Tim, Since the configuration file is read and processed line-by-line, you should be able to use, e.g., netsnmp_ds_register_config(), register_config_handler(), or other config file primitives in your module_init() function, and they will take affect for lines after the dlmod configuration line.

  1   2   3   >