Re: Problem with net-snmp 5.2.1.2

2005-11-15 Thread md salauddin
  Hello, When i say cat /proc/net/dev iam able to get all interface name and other information but if i query using SNMP manager or snmpwalk command iam able to get only three values from interface mib. They are as follows 1. interfaces.IfNumber.0   6 2. interface.IfTable.ifentry.ifIndex.1 :1

udpTable on solaris

2005-11-15 Thread Robert Story
It seems that the udpTable is not getting populated on the sf compile farm nightly builds. I did some debugging, and have some fixes. However, the fixes are completely contrary to the current code comments, so someone needs to look into it further to see if the fixes will break other systems. Teste

Re: bad ifSpeed value - bug ?

2005-11-15 Thread Robert Story
On Tue, 15 Nov 2005 11:52:15 +0100 xavier wrote: XD> i use Net-Snmp 5.2.1 on FreeBSD 4.11. XD> Net-SNMP always return the maximum interface speed (which may be XD> return with ifHighSpeed) and not the nominal bandwith. This is true for all releases and all platforms. In general, the OS does not tr

Re: snmpd memory grows on invalid udp requests --- security????

2005-11-15 Thread Robert Story
On Tue, 15 Nov 2005 10:15:20 -0800 Fong wrote: FT> RS> I'm guessing there is another leak in 5.0.9 that has been fixed in FT> 5.0.10.x FT> RS> that you need to back-port. FT> FT> Do you mean line: 5327 SNMP_FREE(isp->packet) in snmp_api.c (v5.0.10.2)? FT> That part seems correct since when NETSNMP

RE: snmpd memory grows on invalid udp requests --- security????

2005-11-15 Thread Fong Tsui
-Original Message- From: Robert Story [mailto:[EMAIL PROTECTED] Sent: Monday, November 14, 2005 1:42 PM RS> I'm guessing there is another leak in 5.0.9 that has been fixed in 5.0.10.x RS> that you need to back-port. Do you mean line: 5327 SNMP_FREE(isp->packet) in snmp_api.c (v5.0.10.2

bad ifSpeed value - bug ?

2005-11-15 Thread xavier delannoy
Hi, i use Net-Snmp 5.2.1 on FreeBSD 4.11. Net-SNMP always return the maximum interface speed (which may be return with ifHighSpeed) and not the nominal bandwith. ifSpeed OBJECT-TYPE SYNTAX Gauge32 MAX-ACCESS read-only STATUS current DESCRIPTION "An estimate

Re: running

2005-11-15 Thread Thomas Anders
Robert Story wrote: The latest change int 5.3.pre3 was to *restore* backwards compatibility by - changing the get-next loop condition to "while(netsnmp_running)" - declaring netsnmp_running in the agent library - updating snmpd/snmptrapd to use the library globa netsnmp_running instead of

Re: running

2005-11-15 Thread Dave Shield
On Tue, 2005-11-15 at 10:22 -0500, Robert Story wrote: > The latest change int 5.3.pre3 was to *restore* backwards compatibility by > > - changing the get-next loop condition to "while(netsnmp_running)" > - declaring netsnmp_running in the agent library > - updating snmpd/snmptrapd to use the

Re: running

2005-11-15 Thread Robert Story
On Tue, 15 Nov 2005 10:14:15 + Dave wrote: DS> I've also re-read this thread, and am somewhat confused as to DS> exactly: DS> DS> a) what the original problem was DS> (esp. this get-next loop that went bad) It seems one of the examples (don't recall which) wasn't indicating a

Re: snmpd memory grows on invalid udp requests --- security????

2005-11-15 Thread Dave Shield
On Tue, 2005-11-15 at 10:09 -0500, Robert Story wrote: > DS> (And probably update 5.1.x first, or at least simultaneously). > > 5.1.3.1 doesn't have the same potential leak, so it's > really a matter of whether or not the license issue > warrants an immediate release. A release - yes. An immediat

Re: snmpd memory grows on invalid udp requests --- security????

2005-11-15 Thread Robert Story
On Tue, 15 Nov 2005 09:12:57 + Dave wrote: DS> Are you sitting down, Robert? DS> I'm about to agree with you - I think we probably DS> should plan to cut another 5.0.x release. DS> DS> But that's only partly because of this problem - I'm DS> not convinced it warrants opening up a closed line D

Re: snmpd.conf update

2005-11-15 Thread Dave Shield
On Tue, 2005-11-15 at 01:33 +0100, Thomas Anders wrote: > Dave Shield wrote: > > But I'll get the main man page documentation finished first. > > I like what you did. Not "did" - "doing". I haven't finished yet - there's a limit to how much documentation I can rewrite when I'm out at concerts, an

Re: running

2005-11-15 Thread Dave Shield
On Mon, 2005-11-14 at 13:22 -0500, Robert Story wrote: > I've filed a bug report as well, though I'm not sure how > we'll ever fix the problem if this is too large a change. Just to be clear, I was suggesting that this was perhaps too large a change at this point in the release cycle.

Re: DisMan Event MIB selection

2005-11-15 Thread Dave Shield
On Mon, 2005-11-14 at 09:16 -0800, Wes Hardaker wrote: > > On Wed, 09 Nov 2005 09:58:03 +, Dave Shield <[EMAIL PROTECTED]> > > said: > > >> But does the behaviour do what the MIB says it should? > > Dave> In most cases - yes. (E.g. not generating a mteTrigger* > Dave> notification u

Re: Good example of GetNext?

2005-11-15 Thread Dave Shield
[ First - *please* don't mail me privately, without copying any responses to the mailing list. I don't have the time or inclination to offer private, unpaid, SNMP consultancy. Keep discussions to the list, where others can both learn and offer advice. Thanks. ] On Mon,

Re: snmpd memory grows on invalid udp requests --- security????

2005-11-15 Thread Dave Shield
On Mon, 2005-11-14 at 13:29 -0500, Robert Story wrote: > So, then the question to coders and the admins is, does > this qualify as a DOS, worth a 5.0.10.3 release? I'd lean > towards yes. Are you sitting down, Robert? I'm about to agree with you - I think we probably should plan to cut another 5.0