Hallo,
The AgentX subagent tries to automatically re-register its MIB range as soon as the
master agent is up again. It should not be necessary to manually do all that stuff. Is
there a specific reason why you use the IPC heartbeat messaging?
Regards,
Bernhard
> -Original Message-
> F
Hi Benhard,
I've a similiar requirement, in which subagent is part of a large
application. Since application can be started/stopped at run-time same is
applicable for subagent also.
any help on this?
thganks & regards,
Nitin
-Original Message-
From: Penz, Bernhard
> I need to implement the following scenario:
[snip]
> Does agentxPingInterval help here?
Yes.
As Bernhard has indicated, this is *exactly* the scenario that
the 'agentxPing' request was intended to address.
Dave
---
SF.Net email i
> Now we have partly implemented DISMAN
> We hope our work can be shared with others,we hope our work can be
> part of netsnmp.But I don't know how to submit my codes to netsnmp,
Go to
http://www.net-snmp.org/patches/
and upload the code to there.
Dave
Steve Friedl wrote:
Michael J. Slifcak wrote:
Steve, the patch looks good from five hundred feet elevation %^)
To complete it, you'll need to add an #undef patch to acconfig.h
so the #undef will be propagated to net-snmp-config.h.in .
I've been bit several times by that one before.
Ok, I've submit
I know there has been a thread going on man pages, but with the CVS code
downloaded last night I get the following on make install;
make[1]: Entering directory `/home/luser/src/net-snmp-linux/man'
make[1]: *** No rule to make target `netsnmp_todo.3', needed by
`maninstall'. Stop.
make[1]: Leavi
> I know there has been a thread going on man pages, but with the CVS code
> downloaded last night I get the following on make install;
[snip]
> Maybe that has been fixed.
It was fixed on the 5.1.x line.
I've just applied the same fix to the main development line as well.
Simply delete t
But I think my implementation is not only a patch,
I have implemented rfc 2925,and I have independently implemented some MIBs which
haven't been implemented in the current netsnmp yet,and it can improve the functions
of netsnmp on remote operations.And I hope
my implementation can be added into t
> > Go to
> > http://www.net-snmp.org/patches/
> > and upload the code to there.
> But I think my implementation is not only a patch,
So?
That's still the simplest way to submit code to the project.
It allows you to provide a description of what the code does,
upload any files that might be
Dave, Bernhard,
Thanks a lot for your quick response.
I just tried agentxPingInterval, it does work!
The interal IPC I mentioned is mainly
for other things. One use is to control when
subagent should register with the master agent
(agentxPingInterval is automatic).
I saw an API agentx_reopen_sessio
Title: pass and pass_persist on win32
Hi
Is the pass and pass_persist features not available for win32?
I have struggled a long time now, the HAVE_EXEC flag gives me headache
Best regards
the Happy SNMP rookie..
Not available, but only because some bright Win32 API programmer
has yet to offer a patch that uses CreateProcess,
and provide a separate worker thread,
and finally the cleanup of that worker thread when the agent shuts down.
Clues for worker thread can be found in snmplib/winservice.c,
but not f
everything can be a patch, even you re-write all the code.:)
>-Original Message-
>From: [EMAIL PROTECTED]
>[mailto:[EMAIL PROTECTED] On Behalf
>Of [EMAIL PROTECTED]
>Sent: 2004年8月11日 22:02
>To: Dave Shield
>Cc: [EMAIL PROTECTED]
>Subject: Re：: Re: how to submit my codes to netsnmp?
>
>B
But how can my implementation be added into the net-snmp package?
What are the detail steps?
Ling, Xiaofeng <[EMAIL PROTECTED]> 写 :
> everything can be a patch, even you re-write all the code.:)
>
> >-Original Message-
> >From: [EMAIL PROTECTED]
>
> >[mailto:[EMAIL PROTECTED]
> On B
Hi, Greeting,
Is the followiing command correct? I am on Linux with v5.0.9.
snmpwalk -c public tcp6:[fe80::2e0:81ff:fe25:8097]:161 ucdavis
I lanuch snmpd on udp6:161,tcp6:161 and netstat shows it is lisening on tcp. However,
I seem have a problem to run above command. It gives me "snmpwalk: U
1.1. Include your copyright information in COPYRIGHT file
in your source directory.
1. Prepare a gzip'd tar file containing your sources.
no .o .a .so files; no binaries
2. Go to http://www.net-snmp.org/patches/
3. Click on "submit new" link.
4. In the Summary field, provide a short
But my implementations can be added into the next version of netsnmp?
> 1.1. Include your copyright information in COPYRIGHT file
> in your source directory.
>
> 1. Prepare a gzip'd tar file containing your sources.
> no .o .a .so files; no binaries
> 2. Go to http://www.net-snmp.org/patch
I believe there are patches and a bug report in the net-snmp repository
that might provide ideas that can be applied to your situation.
In other words, there may be a need to parse the '[' and the ']'
more intelligently ?
patch [ 942155 ] IPv6 addr_string fix
patch [ 550830 ] Fixes for libwrap
Title: Re: Get data using Tcp6
I have checked patch [ 942155 ] IPv6 addr_string fix,
patch [ 550830 ] Fixes for libwrap. Both fixes seem
manipulating data after netsnmp_tcp6(udp6)_fmtaddr() forms the address.
Why we need to reformat the ipv6 address(add [..]:port) at
the first place ?
What follows is Slifcak's understanding, and does not
represent the official position of the Net-SNMP project.
--
I think your question is partially answered through additional reading.
Download 5.1.2 source from http://www.net-snmp.org/download/,
and read these files (in no particular order):
> From: "Fong Tsui" <[EMAIL PROTECTED]>
>
> I have checked patch [ 942155 ] IPv6 addr_string fix,
> patch [ 550830 ] Fixes for libwrap.
> Both fixes seem manipulating data after netsnmp_tcp6(udp6)_fmtaddr()
> forms the address. Why we need to reformat the
> ipv6 address(add [..]:port) at the fi
21 matches
Mail list logo
