nv-l

Re: Problem Loading Mibs

2000-04-19 09:18:32
Subject: Re: Problem Loading Mibs
From: James_Shanks AT tivoli DOT com
To: nv-l AT lists.tivoli DOT com
Date: Wed, 19 Apr 2000 09:18:32 -0400

Chris -

I took a look at his, and you are correct that the V1 mibloader in UNIX and NT
does not recognize comments within comments.  But that is not likely to be fixed
at this late date.  More likely we would just edit the MIB that is shipped.   I
have looked at adding new flexibility to the parser algorithm in the V1 loader,
and the only way to do it would be to re-write the whole thing, since every
other part of the loader expects to get back from the algorithm just what it
gets now.

But even so, I am very confused about why you would want to load rcf1212.smi in
the first place. It does nothing.  It is simply part of the reference document
which defines OBJECT-TYPE for use in SNMP.  OK, but the code to recognize and
parse OBJECT-TYPE is built into the NetView compiler already, so this document
is just that, a document and not a MIB.  Granted, some MIB compilers would make
use of this definition, but NetView's does not.  It already has it built-in.

And since this "MIB" gives you nothing to browse with the MIB browser, or
collect on with snmpCollect, why would you want to load it?    I suspect that is
the answer to your question about why this has never been fixed.  Nobody in
development thought to do what you are doing.  So is there a purpose here we
missed?   Or did you just write a script to load everything in the directory?

James Shanks
Tivoli (NetView for UNIX and NT) L3 Support



Chris Cowan <chris.cowan AT 2nd-wave DOT com> on 04/14/2000 11:19:56 AM

Please respond to IBM NetView Discussion <nv-l AT tkg DOT com>

To:   IBM NetView Discussion <nv-l AT tkg DOT com>
cc:    (bcc: James Shanks/Tivoli Systems)
Subject:  Re: [NV-L] Problem Loading Mibs




James_Shanks AT tivoli DOT com wrote:
>
> Your copy of the MIB is just needs a little fixing up, syntax wise.  We will
be
> shipping a corrected version of it in NetView 5.1.4 and 6.01.  By the way,
your
> 5.0 is out of support.  You need to consider migrating to either 5.1 or 6.0.
>

Speaking of MIBs, ASN.1, and loader ...

Are there any plans to have the Netview compiler/loader ignore the
     -- Some comment text -- More comment text
scenario?

I complained about this once before, because some of the 6.0 MIBs (e.g
RFC 1212) are shipped with this syntax.  Basically, someone attempted to
block comment out part of the file.   Technically though, a second set
of -- characters is a comment termination token.

I have a perl script to clean this up, but it's a pain to deal with over
and over again.   There are several MIBs in the standard set that have
this problem.

Attachment: chris.cowan.vcf
Description: Binary data

<Prev in Thread] Current Thread [Next in Thread>