Re: Using nvdbimport

2000-03-20 23:04:16
Subject: Re: Using nvdbimport
From: lclark AT us.ibm DOT com
To: nv-l AT lists.tivoli DOT com
Date: Mon, 20 Mar 2000 23:04:16 -0500
I'm the one who reported that fields did not have to be pre-defined. That
information was from an internal update held when the function was first
made available. I see that the V6 man page clearly states that the fields
must be predefined, so it is probably true for V5 as well. I always do
pre-define them so I can add the Locate flag. I clearly have not checked
the man page lately!

Your comma in the Selection Name just won't work, as far as I can see,
since there appear to be no alternatives to comma-separated fields.
 Quotes around it won't be accepted, either. Personally, I avoid blanks
and punctuation in selection names because of little things like this
jumping up to bite you.

It fails with an error message if you give it too many values for the
of fields in the header, which is how yours will be interpreted. And if the
Selection Name does not exist, that node is skipped with an error message,
so either way you won't get unwanted fields set in the db.

Maybe if you clarify what you are trying to achieve, someone will have
a suggestion. Have you considered setting the field right on the router?
(I assume those are not router names with the commas in them.)

Something like, make a Collection of things with 'Atlanta' in the
field, and set the circuitid for everything in that Collection, maybe?


Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking

Hi, Linda --

The nvdbimport command requires that the fields already be defined.
Instead of editing ovw_fields, though, you should create a separate file
under /usr/OV/fields/C, so that your modifications are clearly identifiable
and to make future migration or updates easier.

I noticed that you left a space after the comma in " CircuitID" after the
comma.  The nvdbimport command will try to update the field " CircuitID"
instead of "CircuitID".

The commas in your Selection names would be interpreted as field
delimiters.  I'm not sure if there's some sort of a workaround for that.
Someone else may be able to answer that.

Tivoli Education

In a message dated Mon, 20 Mar 2000  4:46:48 PM Eastern Standard Time,
"Ferreira, Linda" <FerreiraL AT nabisco DOT com> writes:

> Hi everyone,
> In order to use the nvdbimport, do you first need to create the fields
> editing ovw_fields and running ovw-fields)?  and then use ovdbimport to
> populate the fields?  Or can the nvdbimport create and populate at the
> time.  In the manual it says that the fields need to exist first before
> populate, but in reading some entries from the NL-LIST I decided to
> some of you experts who have already done this?
> Also, I have a question as far as populating the fields.
> If, for example, I already have a Selection Name and I want to enter
> "CircuitId" to correspond w/ that selection name do I do the
> following???????
> Selection Name, CircuitId
> I know it must be comma delimited, the problem is that my "Selection
> have comma's in it, will it ignore the commas in the "Selection Name" or
> will it mess up my database by changing my Circuit Id's to "GA" and "IL".
> would appreciate any feedback from anyone who has used nvdbimport before.
> Is the format above correct for populating the fields????  Thanks.
> Linda Ferreira
> Network Management Center
> 973-682-6065
> ferreiral AT nabisco DOT com
> _________________________________________________________________________
> NV-L List information (unsubscribing, policies, posting, digest version,
> searchable archives): http://www.tkg.com/nv-l


NV-L List information (unsubscribing, policies, posting, digest version,
searchable archives): http://www.tkg.com/nv-l

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