FLATLAND96: SOFTWARE Messages: 20 Entries..

The ASTER startup script in /etc/rc3.d
now has the SLIP connection to the ADAM (using FreeWave Radio)
commented out until the radio/adam is available and needed.


The file now looks like:.....

#
# Start ASTER processes
#

user=aster

case "$1" in
'start')
     
#     echo " Starting slip on cua/b at 34800 baud"
#     echo "/etc/slattach cua/b 128.117.87.29 128.117.87.30 38400"
#     /etc/slattach cua/b 128.117.87.29 128.117.87.30 38400

#     echo "Adding route to goldenrod"
#     route add host 128.117.87.2 128.117.87.30 1

     echo " Logging in as $user to start ASTER processes..." >/dev/console
     su - $user -c "Aster_Start" >/dev/console 2>&1
     echo " Aster processes started" >/dev/console
    ;;
'stop')
     ;;
esac

The 4 available serial ports will be assigned to:

Aster:	/dev/cua/a	U.S. Robotics DP/PPP to Net66 Internet access
	/dev/cua/b	Freewave Radio SLIP connection to ADAM
			[Not installed as of 6/x/96]

Cocklebur:
	/dev/cua/a	Freewave Radio "eve_rf" to PAMs, 19200-bps
	/dev/cua/b	GPS Time Code Receiver, 4800-bps


EVE_RF:
can be run on cocklebur using the system startup script:
	/etc/rc3.d/S80pam start
or can be terminated using 'stop' instead
It simply runs the polling program with the correct pass parameters,
and writes the log file to /pam/log/eve_rf.log.


GPS:
The GPS program is run every hour on Cocklebur using root cron:

    # Run time_code to get time from GPS receiver. Use -s option to set clock
    #
    0 * * * *       /home/aster/bin/sparcsunos5/time_code -L -s -d /dev/cua/b >> /home/aster/log/time_code.log 2>&1

And on ASTER root cron, 5 minutes later:
    #
    5 * * * *       rdate cocklebur > /dev/null 2>&1

The Startup script for the Internet Connection PAM GOES reader
is in:	/etc/rc3.d/S80pam	for ASTER startup

and a duplicate copy in:	/pam/bin/...

It creates log files in: /pam/log

The program can be initiated using: /etc/rc3.d/S80pam start
and terminated using 'stop'

PPF Data files are put in /data/incoming/ppf/FLATLAND/tcp_ppf

The Startup script for the EVE_RF Polling Program
is in:	/etc/rc3.d/S80pam	for COCKLEBUR startup

It creates log files in: /pam/log/eve_rf.log

The EVE_RF program can be initiated using: /etc/rc3.d/S80pam start
and terminated using 'stop'

PPF Data files are put in /data/incoming/ppf/FLATLAND/eve_rf
These will be processed into netCDF by 'run_pam_file_translators'
and put into /data/ppf/FLATLAND96/eve_rf

I loaded the 'June 12th' version of Burghart's
pam2netCDF translator program.
The old version was renamed to ".old"
The new version allows fields to be generated for even if there is no data.

I finally got around to copying down Matt's new version of TKLOG which
is designed to work with his Paradox file import stuff on the PC for
grabbing station visit reports generated on the Compaq.

old version saved in: /home/aster/apps/tklog/tklog.save.working

Removed Message Types PAM1,PAM2,PAM3 because we're now using the site number for
that.  Matt's stuff will recognize that convention.

Gordon has revised the method for inputing time to the dailyplot
routines, enabling an arbitrary starting time and length.  The new
method for passing time to the routines is via the function dpar(),
which is like par():

Prior to making the first dailyplot within an S session you must 
execute dpar, e.g.

dpar(year=1996, mon=6, day=15, hour=12, lenhr=24)

Then fun.xxxx.plot will produce a 24 hour plot beginning at 12:00 gmt on
June 15.  Subsequently dpar(day=20) will produce the same plot for June
20 or dpar(lenhr=96) will produce a plot of 4 days of data.

dpar.next() and dpar.prev() will increment dpar()$hour by + or -
dpar()$lenhr.  

dpar.now() gives a plot ending at the present time and is also a handy
way to initialize dpar at the beginning of a session.  dpar.now() is
currently executed by $ASTER/.Data/.First()

Try dplot() to get a graphical interface to the dailyplots in the
daily_S_plots script.  Provide comments and feedback to Gordon.

Moved the crontab entries to run the netcdf file translators from
militzer to aster acount.

  I added some code to allow for direction
changes in sonic booms as they are moved up the
mast. This was done because station 2 sonic was
moved yesterday.

Installed new software on the Matrix card. Writes to the ingest output
socket in the sync task are now non-blocking, so the adam does not crash if
the network connection is jambed.

Samples can be lost when the network is jambed, however.  To see the total
count of lost samples, do

	sshow cosmos 0

You'll see output like so:


c:000 cosmos mem:131572 nent:153 nlost:0 nlate:0 pS:0 pD:0 dT:5410 msec, blocked:0, lost:0, writerr:0

The last three parameters (blocked, lost, writeerr) result from network jambs.
They are totals since the last adam reboot.

If the network is jambed for more than two minutes, then ingest will
timeout and shutdown, in which case the adam will need to be rebooted.

  I modified daily_S_plots to included the temperature
and relative humidity data, from cosmos, in the temperature
and humidity daily plots. This confirmed EVE and cosmos have
the same coefficients for the sensor.

When Tom requested an upload of 'eve_rf' data to cover the GOES outage
a few days ago, I put together on steam:
	/rdss/pam/bin/rcp_flatland_data

It is being run under my crontab at 1:15 every day to automatically copy
'yesterday's ppf file.   That file is subsequently converted to netCDF by
Matt's crontab initiation of the 'run_pam_file_translators'
It can be run for any day by using:
	rcp_flatland_data 1-jul-96
(or)	rcp_flatland_data 960701
etc.
It first checks to see if the file is there, and thus that the net is up also.
If the net's down, it waits 5-minutes and trys again.  You can change the
default limit (12) on the number of these retry attempts by using:
	rcp_flatland_data -r 3

There are several plots which combine several variables (weather, energy 
balance) which die with "length not a multiple..." errors and exit dplot.
This happens when "Plotting Recent" while the netcdf files are being updated.
We could spend a lot of time modifying the plot codes, but in the meantime
there are two workarounds:
1. Don't make these plots at these times (easy to know because the RASS also
goes off at the ppf->netcdf update times!)
2. Use "Plot" rather than "Plot Recent".
Running the plots manually (not window-driven) avoids having to clean up
the dead windows when this occurs.

A new version of tklog was installed.

Old version is in:      /home/aster/apps/tklog/tklog.save.16jun96

Modifications relate to HTML generation:
        - site # grouping added.
        - both ascending and descending message ordering available.
        - all chronological messages are now contained in the
          'contents' file, not just the message titles.
        - index button added to return to top of a file, where
          the header contents for that file are located so that
          it's a bit faster than scrolling (provided the browser
          is accessing the files locally).

I found that the temperature coefficient for the CS615 needed to
be multiplied by 100 in fun.qsoil.plot.  I have made this correction
and changed this routine to plot only the corrected data if CS615=T.
This is now the default in fun.soil.pam.

The temperature correction now appears to eliminate the diurnal cycle
in the soil moisture at the lower soil moisture values found at sites 1
and 2 (10-20% by volume).  It is insufficient at the 30% soil moisture
being measured at site 3.

I've added better support for multiple variables in fun.var.plot, so it
should be relatively easy to build new (somewhat crude) dailyplots in S.
See my "SPO's picks" in the dplot "Weather" menu for how this can be
used (abused!).  See comments in the source 
($ASTER/apps/sfun/newdailies/var.plot.q) for how to use it.

It has become apparent that we want a command to look at recent data from the
stations.  Steve S. told me about the "last" command, which works on stout
but not here (even when using the ppf= option).  As a temporary cludge, I
have created the aliases in the aster account: "last1", "last2", and "last3" 
which run the "last" command on stout for each pam station.  This serves our
need for the moment.

Version 2.9 of TKLOG is now installed.

Graphics image files are now included in the HTML file generation.

GIF is the preferred format.  TKLOG will attempt to convert non-gif
images into gif for the html stuff to save space, load time, compatibility,
etc.   Multiple images per message are possible in Matt's graphics handlers,
and also for the html.

I've connected the S+ legend locator function to my fun.array.minmax routine,
which I think will speed up multiple day plotting.

I've changed the atik calibration routine to have the default ambient pressure 
entered as a parameter in prep.config.  I've set this to 990mb for FLATLAND.

I just "make"d prep, covar, xcockpit, and xstrip and restarted covar on
cosmos.

Gordon: I tried another way to do this at first, but the variable
"p_baro_share" was being written to by some routine which I couldn't
find, so I just overwrote the value in sonic_atik.  You may want to
look at this further.