[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[LDM #IXG-702275]: LDM setup up problem questions

Hi Yoori,

> First of all, I really appreciate about your continuous effort for my help.

No worries.  Like I said before, we are glad to help when we can.

> The reason why I'm trying to setup LDM is : My advisor recently got funded
> project for his research work. To continue the project, we need streamflow
> data from other research institution/university. To request and get the data
> (streamflow data) from them, we needed to setup LDM in Linux machine.

Very good.  Out of curiousity what is the other research institution?

> And about upgrading issue you recommended:
> Of course, I want to upgrade into FC6 (http://fedora.redhat.com/). But isn't
> it hurt to current setting (LDM setting, etc.) what our machine currently
> has like losing files/LDM setup, etc.?

Yes, but the installing/upgrading the LDM on a machine typically only takes
about 10-15 minutes.  The reason it took a lot more effort on your machine
was because of the lack of a full development enviornment.  Our experience
is that setting up a machine for LDM use after the operating system has been
loaded AND a full development environment has been installed (which should
be done as part of the operating system installation) is very quick and
simple.  I am most concerned with the security of your machine since RedHat
9 has had no security patches for a number of years.
 
> The only use of LINUX machine is only
> for LDM setup to receive the data from other sources. We will not use for
> any other reason.

OK.  If you turn off all unneeded services (like the printer daemon, etc.)
and really lock down the machine via your firewall setup, it will be OK.

> It's why I'm kind of hesitating to touch machines, but I
> definitely understand your saying.

OK.  I just wanted to make a case for security :-)

> And as a purpose of clarification for my next step and what to do next:
> From
> http://www.unidata.ucar.edu/software/ldm/ldm-current/tutorial/index.html:
> My left-over is the part after "4. Installing the LDM" what I think. Before
> I go to next step, I should first contact to people who I need to request
> and receive the data. Is it correct?

This is correct.  The site(s) who yo want to feed from have to configure
their LDMs to allow your LDM to request data.

> And even though the definition is
> stated in website, it wasn't clear with my language. Could you let me know
> difference between upstream and downstream means?

When you setup your LDM to request data from one or more machines, your
LDM will be the downstream node, and the feeders will be the upstream
node.  The analogy is that like water flowing down hill, data flows
from an upstream LDM to one or more downstream LDMs.

> From the manual, there are
> many indications of these words like 'upstream data-feeds', 'upstream LDM',
> 'downstream LDM', etc.? And if you can explain what I have to do next
> (assume I finished reading manual in website) to ingest streamflow data from
> other university?

You now need to:

- contact the institution that you want to receive streamflow data from and
  provide them with the fully qualified name of the machine your LDM is
  installed on, nopp.cae.drexel.edu.  When doing this, it is always a good
  idea to also provide the site with the IP address of your machine

- the upstream site(s) that you contact will add an 'allow' line in their
  ~ldm/etc/ldmd.conf file that tells their LDM to allow your LDM to request
  data.  The LDM administrator on the upstream site(s) will inform you of
  the data feed name and possibly give you an example of the product meta
  data (ID) that will be included in the feed.

- you will need to configure your LDM to request the data from the upstream
  site(s) you contact.  You will do this after learning about the data feed
  that the data will be available on and the product metadata (ID).  The
  product metadata (ID) can be used to request a portion of the data feed
  if you don't need all of what is available.  This subselection of data is
  done by devising a regular expression that matches that portion of the
  data feed that you want.

  The following simple example should illustrate the difference between
  requesting all of a data feed and part of it.

  Compare the following request lines:

  request   NEXRAD2   ".*"    thelma.ucar.edu
  request   NEXRAD2   "KFTG"  thelma.ucar.edu

  The first line requests all of the data available in the NEXRAD2 data feed
  from the upstream host thelma.ucar.edu.

  The second line requests that fraction of the NEXRAD2 data feed whose product
  metadata (ID) matches the regular expression KFTG.  Please note that KFTG
  really is a regular expression.  It is equivalent to, but more efficent than
  *KFTG*.

So, the first order of business is to contact the LDM administrator of the
the upstream site(s) that you will want to request data from.

> Thanks a lot.

No worries.

Cheers,

Tom
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: IXG-702275
Department: Support LDM
Priority: Normal
Status: Closed