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

20030425: UFRJ as a primary IDD redistribution point for Brazil (cont.)

>From:  David Garrana Coelho <address@hidden>
>Organization:  UFRJ
>Keywords:  200304251624.h3PGOl7U007475 IDD LDM

Hi David,

>       I'm writing to inform you that LDM didn't reacted well to a local
>router restart at April the 23th. Only noticed it yesterday, and I had
>to restart ldmadmin to solve the problem. We had the router restarted
>before, and LDM always kept working nicely. Is that an issue I will need
>to be aware in the future?

I noticed fairly soon after brisa stopped reporting real time
statistics and tried to logon to find out what was going.  Since I was
unsuccessful in that login attempt (using ssh, of course), I figured
that something was amiss with the network to brisa or that brisa itself
was not up.

We just logged onto to brisa and one of its upsteam feeder,
sapodilla.rsmas.miami.edu, and traced through log files to see what
might have happened.  We saw the connection to the upstream sites fail
and then the LDM on brisa began trying to reconnect.  All reconnection
attempts were unsuccessful, but they stopped, or, at least, the logging
of the reconnect attempts stopped on Apr 23 at 12:07:24.  At no time
during the reconnect attempts by brisa did the log file for the
upstream site indicate that reconnection was being attempted.  The
mystery to us at the moment is why the logging stopped at 12:07:24.

We see that the LDM was restarted at 21:23:00 on April 24.  Again, I
noticed brisa's stats being reported and so figured that whatever
problem that was causing the outage had been corrected.

The thing that is a mystery to us at the moment is why brisa's LDM
stopped logging reconnection attempts.  Did you notice anything unusual
about the machine before stopping and restarting the LDM?  If not, we
will try to recreate the failure here with one of our RedHat Linux
machines to see if there is a problme with the LDM-6.

>       I'd like to ask too about the status of satellite imagery you
>mentioned a while ago. I know the GOES series just were replaced, and
>maybe it would be a good time to add a (south) South America sector in the
>products, since probably all full disk cuts had to be redefined...just a
>kind reminder...=)

Thanks for reminding me.  I had put off the installation of McIDAS on
brisa until it had been upgraded to our new LDM distribution.  The
development of the LDM-6 took a lot longer than I had anticipated back
last November, but we feel that the results were worth the wait.

I will try to get to the McIDAS installation and creation of cron
scripts that go out and create GOES-12 sectors for Brazil and Argentina
next week.  In the mean time, I would be happy to hear what imager
sectors you would like to see.  What I have in mind are things like
temporal and spatial resolution, location, and band (wavelength
channel: visible, thermal IR, water vapor, near IR, far IR, etc.).  The
other possibility will be channels from the sounder instrument.

>Best regards,

I will let you know when I have the image sector creation working on
brisa.

Cheers,

Tom
--
+-----------------------------------------------------------------------------+
* Tom Yoksas                                             UCAR Unidata Program *
* (303) 497-8642 (last resort)                                  P.O. Box 3000 *
* address@hidden                                   Boulder, CO 80307 *
* Unidata WWW Service                             http://www.unidata.ucar.edu/*
+-----------------------------------------------------------------------------+