CASES99: BaseSystems Messages: 10 Entries..

ops4 has been created for the post-intercomparison of the radiometers
on the FLAB roof

Stopped all systems to begin ops3.  Ops3 was needed to change the
gain on the PAM A/D.

PAM stations at 0022 GMT, Oct 20
cosmos       at 0037 GMT
daisy        at 0040 GMT
marigold     at 0041 GMT
ragwort      at 0042 GMT

In order to redo recently lost covars, edit
$ASTER/projects/CASES99/scrips/covar.redo with an appropriate start time.
Then execute covar.redo.

The covars quit around 17:25 CDT Oct 17.  Those for daisy and marigold
were restarted following the hangup with Clover, but the others were not 
restarted until 0900 CDT oct 18.

While Tom and I were installing the CO2 sensor at 20 meters,
clover suffered a crash. Dave Miller was taking credit for it!
I had to power cycle clover to get it going again. This crash
also shutdown the archiving of blazing data. After a short conversation
with Gordon, marigold and daisy were mxreset to get things going again.

Tom,

I Fedex'd 100 CD labels this morning.  The Boulder store also did
not have the identical product.  I bought some Memorex
labels that are perforated in the middle and were folded in
half in their box. Otherwise they looked the same as the others.
If you feed them in whole they should work.

If you run out of labels before they arrive, don't use a pen
to label the CD's.  You're supposed to use a special CD pen -
otherwise some inks can apparently leach through the plastic
and spoil the substrate.

If you forget what is on a CD, you can just insert it into
one of the CD drives on clover or cocklebur (except the CD-R drive).
A file manager window should popup, which will allow you
to browse the CD, and then eject it when you're done.

If a window doesn't popup, you can use shell comands to look
in the /cdrom directory.  Afterwards, cd out of /cdrom
and do:
	eject cdrom


I notice that marigold got in a state where it flooded the network
with packets.  This happened once in early Oct too.  I don't
know the reason or any fix, except rebooting.

Gordon

We write 4 CDs a day:
	stns			~360 Mbytes
	tower			~350 Mbytes
	film 00:00-12:00	~550 Mbytes
	film 12:00-24:00	~550 Mbytes

	total			~1720 Mbytes/day

The archive hard disk is /data/archive/projects/CASES99/raw_data,
an internal disk on cocklebur, which is 8 Gb.

To write a CD, find a cocklebur window, and
put a new CD into the CD-R drive on top of cocklebur

Execute one of the scripts, passing the month number (10=october, duh)
and the day of the month:

	bkcd.stns mon day
	bkcd.tower mon day

first half day of hotfilm data:
	bkcd.film mon day 0 12
last half day:
	bkcd.film mon day 12 24

This script moves files for that day to the "cdbackup" directory,
and then writes them to the CD.  

IF IT SUCCEEDS, IT DELETES FILES FROM THE HARD DISK, so don't
damage the CD. That's your only archive.

The label is then sent to the printer.
Wait for the manual feed light to appear before inserting a label 
sheet. Slide the label sheet into the manual feed slot on
the back of the printer, until it stops, label side toward you,
with an unprinted label in the lower right hand corner.	

After it prints, peel the label of and put it, sticky side up, 
in the CD Stomper. Put the CD, LABEL SIDE DOWN, DATA SIDE UP!!!! 
over the spindle. Anybody who puts the label on the wrong side of
a CD must stay an extra week!!!!

Put the CD in a sleeve, and stick the narrow label on a strip of
perforated paper and slide it into the top slot of the sleeve (use 
the black plastic thingy).

The log of the CD writing is written to $ASTER/projects/CASES99/cdlogs.

Don't write directly on the CD with any pen - reportedly the ink
can leach through and damage the substrate.








	

	

A day's hotwire data is 5.0 Gig.

The hotwire files are on
	/data/blaze/aster/projects/CASES99/raw_data/hw

This is an 18 gig disk, so we can hold 3.5 days.  To avoid
running out of space you should archive yesterday's files
each day.

To archive a day's data to disk:

Log onto clover
Insert a new tape into clover's exabyte, then enter:

	blaze2Tape 0 mon day

where mon and day are the month and day to be written to tape.

blaze2Tape takes about 2 hours or so to complete.  It will also delete the
files of (day-1) if they were previously written to tape.

The blaze2Tape log files and tape listings are on:
	$ASTER/projects/CASES99/blazelists

If you experience tape errors,
	insert a "premium" cleaning tape into the drive
	restart the day's archive by deleting the 2 log files for
	that day on $ASTER/projects/CASES99/blazelists,
	and restarting from the beginning of a tape.

I have made changes to prep, channel, archive, and
covar config files.
(1) 5 channels have been added to COSMOS to accommodate
the PYG intercomparison. COSMOS serial channels 211 thru
215 were used for these sensors. Gordon will need to come
up with a new cockpit display for COSMOS because there are
too many data coming in to show all the data. The designation
for these data are:
	channel		name (s0x: x is pam station number)
 	211		boom
	212		s01
	213		s02
	214		s03
	215		s05

(2) The 30 meter radiation boom was mounted at 50 meters.
So, the data for this sensor will go thru RAGWORT. The
designation for the sensor's data has been changed from
.30m to .50m. The data is going into channel 206.


 

toucan:
  headless, internal disk only: done 7/29
  solaris: core package with basic networking, ppp, ntp, yp
  latest patches: done 7/29

  priority paging done
  mount disks from aster: automount/NIS done


cocklebur:
  latest patches (2.7) done 7/29
  mount disks from aster: automount/NIS

clover:
  backup existing system to restore when clover returns
 
a replacement of clover for tony:


aster:
  NIS

manus: ppp system at ANL
  headless IPC with magma serial card
  to boot IPC from CDROM:  boot sd(0,6,2)
  install solaris: done 7/29
  install magma card from aster & driver

  install standalone ssh, tcpwrappers

framemaker?
Splus 5?