There are no requirements for me to support the IF-MIB now, but it is mentioned
in the document which I have been looking through. Anyway, if it turns out I
need to support this MIB in the future and I make changes to the NetSNMP code,
then I would be happy to submit these to the NetSNMP team as a patch.
Also, I am not really trying to reduce the size of the SNMP package anymore,
since I believe we have plenty of space to store the files during the build
process (on our build machine). I had space concerns for our target hardware
box since at first I was assuming I would need to copy "all" the entire
"/usr/local" directory tree on to our target box and that might be a problem
due to the size. I believe you mentioned I would only need (at a minimum) the
following:
- snmpd.conf
- library files
- snmpd executable
I am assuming I do not need anything under the "/usr/local/include" directory
as well.... correct? This has a bunch of ".h" files and I do not plan on
copying this stuff to the our target box. Not really sure why all these ".h"
files exist here anyway since the executables and libraries are already created.
Dave Shield <[EMAIL PROTECTED]> wrote: On 17/07/07, Need Help wrote:
> When you describe "--disable-mibs" and "--disable-mib-loading" options you
> mention "MIB files". Does this represent all code located in the
> "../agent/mibgroup" directory which is related to the MIBs
No - those are the "MIB modules".
> or are you
> referring to the MIB.txt text files which are placed in the
> "/usr/local/share/snmp/mibs" directory?
Yes.
> On more thing. I believe I remember that the IF-MIB comes with NetSNMP,
> however, not all field values are supported at this time.
Then
a) raise a bug for the particular fields that are missing, and/or
b) submit a patch to provide the missing functionality.
> If I
> wanted to
> support the IF-MIB then I know I could update the master agent software to
> do this
That would be the best approach, yes - particularly if you then submitted
that update back to us for inclusion in the main source code. That way,
you could actually repay some of the support you've been receiving
recently.
> but would it be possible to simply create a subagent to handle the
> IF-MIB instead. This way, updating future releases of the master agent
> would be a lot easier since I would not have to merge in any changed code,
> just simply UnTar the new version and compile and install it. Or would this
> be unwise since the current foundation of the IF-MIB already exists in the
> NetSNMP framework.
a) It would be unwise
b) It would be selfish
If you've got fixes for the agent, let us have copies so we can include them
in the main code.
However, as I think I've said before - you do seem to be trying to race ahead
and do too much too quickly. I would *STRONGLY* recommend that you
concentrate on the immediate task of implementing the particular MIB
that you've been given first. Improving the overall behaviour of the agent
can wait until that's done.
Believe me, it's very very easy to get sucked into the detailed internals
of this project, and lose sight of the original task.
Dave
---------------------------------
Get the Yahoo! toolbar and be alerted to new email wherever you're surfing. -------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
