There is something wrong either with your summary or with some emailer - I
get partial lines like "- Corrected registering of" and "- Applying patch
from Magnus Fromreide to fix the".
Larry Dickson
Cutting Edge Networked Storage
On 2/4/09, Robert Story wrote:
>
> This
s is related to your
problem but it could be.
Larry Dickson
Cutting Edge Networked Storage
On 7/28/08, Sinha, Birendra (Birendra) <[EMAIL PROTECTED]> wrote:
>
>
> Hi,
>
> Please see the mail thread below, Mike suggetions didn't solve my
> problem.
>
> I am
Hi Latha,
We had a similar problem going from 5.2.1 to 5.4.1 that was discussed on
this list around April 18. The key fix was to make sure the NetSNMP::OID
call comes AFTER the NetSNMP::agent call. The ordering did not use to matter
but now it does.
Larry Dickson
Cutting Edge Networked Storage
/18/08, Larry Dickson <[EMAIL PROTECTED]> wrote:
>
> PROGRESS! ...
>
> There is an error in your "pass.pm" file, line 56. Instead of
>
> @inputs = ();
>
> it should have
>
> @inputs = @ARGV;
>
> because the way it is it never sees anything. After f
ion - even with your snmpd.conf file.
Larry
On 4/18/08, Larry Dickson <[EMAIL PROTECTED]> wrote:
>
> I'm assuming that being root means you have permissions for everything.
> But it does not seem to work.
>
> I ran your snmpd.conf with the following changes: (1) replaced y
ulik <[EMAIL PROTECTED]> wrote:
>
>
> On Apr 18, 2008, at 6:05 PM, Larry Dickson wrote:
>
> Thank you, Dw. I will see if it works. I notice that embedded perl and
> > subagent are treated as mutually exclusive, so I will strip out the embedded
> > perl line from the s
says
/var/agentx/master was the default; I could never get a socket in 5.4.1
until I added an explicit agentXSocket line to snmpd.conf.
Larry
On 4/18/08, Dirk-Willem van Gulik <[EMAIL PROTECTED]> wrote:
>
>
> On Apr 18, 2008, at 4:25 PM, Larry Dickson wrote:
>
> Thanks, Thom
n 4/17/08, Thomas Anders <[EMAIL PROTECTED]> wrote:
>
> Larry Dickson schrieb:
> > Going on, I discovered man/snmpd.examples.5 suggested putting single
> > quotes (?) around do *.pl in the "perl do" line, and doing that made
> > Bareword go away, but still does n
ent). If anyone knows of a
subagent that works in 5.4.1 it might help.
Larry
On 4/17/08, Larry Dickson <[EMAIL PROTECTED]> wrote:
>
> Actually it was adding the agentXSocket directive that got me that far.
> Before, it just shot itself in the head. Now at least it stays alive along
&
On 4/17/08, Magnus Fromreide <[EMAIL PROTECTED]> wrote:
>
>
> On ons, 2008-04-16 at 10:35 -0700, Larry Dickson wrote:
>
> > ::
> > snmpd.conf-old
> > ::
> > #version 1
> > rocommunity public
> > rwcommunity private
> &
ere in the codebase surely it must trigger a
communication, presumably across the socket "master", to tell the agent it
exists?
Larry
On 4/16/08, Dave Shield <[EMAIL PROTECTED]> wrote:
>
> On 16/04/2008, Larry Dickson <[EMAIL PROTECTED]> wrote:
> > the n
uot; if ($DEBUG);
$agent->register( 'myname', $regoid, \&my_snmp_handler );
print STDERR "AFTER registration at $regoid\n" if ($DEBUG);
if ($subagent) {
my $firstrun = 1;
print STDERR "Yes I am a subagent approaching running loop.\n" if
($DEBUG);
# We ne
subtree
when I do a "getnext". But I expect I can turn on DEBUG in the subagent and
track it down. If not, I'll get back to you...
Thanks,
Larry
On 4/16/08, Dave Shield <[EMAIL PROTECTED]> wrote:
>
> On 15/04/2008, Larry Dickson <[EMAIL PROTECTED]> wrote:
Hi all,
Our path from net-snmp 5.2.1 to 5.4.1 has been blocked for months because
the latter refuses to talk to our subagent. It worked fine in 5.2.1; we
could even do a full walk through our MIB. Has anyone run across a problem
like this? Following are some collected notes.
TIA,
Larry
The new /
TECTED]> wrote:
>
> Hi! Larry Dickson
>
> Thank you very much for kind reply.
>
> I don't understand what you're trying to say and
> what do you mean by select is failing?.
>
>
Have you checked whether a select is failing somewhere? It can generate a
tight loop with such full CPU utilization. I remember this from a non-SNMP
project, but select is used everywhere.
Larry Dickson
On 3/25/08, Siva Prakash Reddy G <[EMAIL PROTECTED]> wrote:
>
> Hi!
>
>
On 2/12/08, Dave Shield <[EMAIL PROTECTED]> wrote:
>
> On 11/02/2008, Larry Dickson <[EMAIL PROTECTED]> wrote:
> > I inserted ### before the "perl do" line in each copy of snmpd.conf that
> had
> > it.
>
> And restarted the agent, I presume?
Yes,
Connection from
before it jumps to our private subtree; this does not appear except when I
jump out of netSnmp subtree; but I do not believe that behavior has changed
either.
Thanks,
Larry Dickson
On 2/11/08, Dave Shield <[EMAIL PROTECTED]> wrote:
>
> On 11/02/2008, Larry Dickson <
307
engineBoots 211
oldEngineID 0x80001f88803fa6fa6b0f778f47
=== Contents of (recently created) /etc/snmp/snmpd.conf
#version 1
trapsink 192.168.1.122
rocommunity public
rwcommunity private
syscontact Larry Dickson
syslocation Cutting Edge
master agentx
perl do /etc/edgeware-subagent.pl
Interleaved below...
On 2/11/08, Dave Shield <[EMAIL PROTECTED]> wrote:
>
> On 11/02/2008, Larry Dickson <[EMAIL PROTECTED]> wrote:
> > Response to getnext at netSnmp.1.3 is timeout.
>
> OK - what about
>GETNEXT netSnmp.1.5
>GETNEXT netSnmp.1.7
>
.
But my question is also more algorithmic: How is it (the agent/subagent
combination) supposed to do a getnext that jumps out of the subagent's
subtree?
Larry Dickson
On 2/11/08, Dave Shield <[EMAIL PROTECTED]> wrote:
>
> On 23/01/2008, Larry Dickson <[EMAIL PROTECTED]> wro
Hi all,
I'm new at this and using a variant of version 5.2.1 (ca 2005-05-26). I'm
using iReasoning mib browser and/or snmpwalk and friends. Attempts to do
getnext at end of a subtree get inconsistent results. In the MIB
files, ucdavis = .1.3.6.1.4.1.2021 and netSnmp=.1.3.6.1.4.1.8072 - and given
t
22 matches
Mail list logo
