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

[IDD #OJU-555308]: Connection Request...

Hi Otis,

re: all machines in the cicsnc.org domain ALLOWed for all unrestricted
feeds on idd.cise-nsf.gov

> Thanks…Scott will be configuring these feeds tomorrow and Friday.  We'll let
> you know if we have any hiccups :)

OK, sounds good.

> So, I take it that we should make the NSF feed primary and the Unidata
> source the backup?  What is your preference?

We will configure our cluster to ALLOW your machine(s) access to all
unrestricted feeds after Scott has gotten the feeds working from
idd.cise-nsf.gov (we setup our cluster to ALLOW all sites in the community
so it can act as an ultimate backup feed site).  We will then identify
a university site (likely PSU) who will act as a secondary source for the
data you want.


- have you discussed getting feeds from NCSU?

  The first contact there would be Gary Lackmann <address@hidden>.
  Gary can forward your feed requests to the appropriate LDM

FYI: the configuration of primary/secondary feeders was done away with
quite some time ago.  The LDM will now promote to "primary" mode that
site that is able to provide the feed-specific data the fastest.  The
other site(s) that that one requests a feed from automatically get demoted
to "secondary" feed status.  Because of this behavior, we recommend that
all sites have at least 2 upstreams for each feed type that they want.
The only requirement that one must follow is to setup the feed REQUESTs
exactly the same for each feed.  For instance:

<in ~ldm/etc/ldmd.conf>

REQUEST IDS|DDPLUS ".*" idd.cise-nsf.gov
REQUEST IDS|DDPLUS ".*" idd.unidata.ucar.edu


If the regular expressions in the redundant feed requests do not
match, undesired consequences _will_ result -- the site
that can feed the data the fastest wins and the contents of that
feed may be different than the contents of the redundant feed.

One can force each redundant feed REQUEST to be "primary" by
making the regular expression different while still being the
same.  For example:

REQUEST IDS|DDPLUS ".*" idd.cise-nsf.gov
REQUEST IDS|DDPLUS "(.*)" idd.unidata.ucar.edu 

The parentheses added to the second REQUEST insure that the
request patterns do not match ( .* != (.*) ), while the
REQUEST actually asks for the same set of data.


If this is unclear, please submit a separate inquiry to
address@hidden, and we will work through a number
of examples with you.  Also, please let us know when the feeds
are working from idd.cise-nsf.gov.  We will then provide you with
the name of a second upstream feed site.

> Thanks again for the quick help...

No worries.

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: OJU-555308
Department: Support IDD
Priority: Normal
Status: Closed