Greetings Phil!
If you are running a Linux kernel (>=2.6.13), you may want to check
out inotify. Inotify is a filesystem even notification system that is
built into the Linux kernel. From what I understand, inotify will
broadcast messages when certain events on the filesystem occur (i.g.
file/folder access, modification, open, close, create). You can trigger
your plotting routines if your write a special script to intercept the
messages.
I use pyinotify (python interface to inotify,
http://pyinotify.sourceforge.net/) to watch directories for incoming
data from one of our remote field sites. Once my 'watcher' script
receives the message that a new file has been created and closed, I have
python generate several threads to run analysis and plotting routines.
My routines consist of more python scripts, but they could easily be any
executable (c-shell scripts pulling the strings of GEMPAK, for example).
I found that inotify takes care of the issues that arise when the data
arrival time does not coincide with the cron setup. Another nice
feature of inotify is that it uses zero system resources while waiting
for kernel messages, unlike a more traditional 'sleep-check-sleep' script.
Sean Arms
Graduate Research Assistant
School of Meteorology
University of Oklahoma
Phil Birnie wrote:
All-
I'm searching for a ideas to display model data more timely. On our
current setup, we simply wait until all data has arrived and then have
a script run in the cron that processes and plots all of our model
data. However, I would like to post these plots in a more timely
manner, preferably as they arrive. I've come up with a few ideas, but
none of them seem very efficient. I'm assuming GEMPAK does not have a
program that can output grids that are currently available. Does
anyone have any ideas (perhaps a listening script)? Any help would be
great,
Thank you,
Phil Birnie
Webmaster
Department of Geography
The Ohio State University
(614)519-6176
www.birnies.com
_______________________________________________
gembud mailing list
gembud@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit: http://www.unidata.ucar.edu/mailing_lists/
