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

20021119: LDM/IDD setup at UBA (cont.)

>From: "Moira E. Doyle" <address@hidden>
>Organization: University of Buenos Aires
>Keywords: 200207311510.g6VFAE903432 LDM IDD MeteoForum

Hi Moira,

re: Marisa's email address
>Marisa´s mail is address@hidden

Thanks!  It turns out that she sent me an email pretty soon after you
did, and I have already responded.

>I tried connecting to Brazil and it worked but the answer was slower than to
>your server.

Interesting.  I will need to run some traceroutes from here to Brazil
and here to Buenos Aires so I can better understand the networking
picture.  We will insure that the server you are accessing data from is
the one that has provides you with the lowest latencies.

>What is the difference between ldmping and ldmping -i 2?.

ldmping cycles through a connection every 30 seconds or so.  The '-i'
flag to ldmping tells it how often to query, so that an 'ldmping -i 2'
will query the named host every two seconds.  Can you tell me what sort
of numbers you get back with your ldmping to atm.geo.nsf.gov and
brisa.meteoro.ufrj.br?  This will be helpful in determining which site
will work better as your upstream feed host.

>When I tried the
>second trying to connect to Brazil it answered faster, though the 3rd time
>gave error, the 4th, time out, and the 5th it connected again and continued

The first time that ldmping connects, you can expect a large number.
Afterwards, the connection times shown should stabilize assuming that
the connection is not intermittant.

By the way, this is the sort of testing that is always done when
determining IDD topology.  I would have thought that your connection to
Rio would be better, but networking routing is somewhat mysterious.