Following this message, I'll mail a compressed tar file that contains the software you'll need, plus a directory structure that various shell scripts and binaries expect. In fact, this tar file contains more directory structure than you'll need -- it has directories for 4 oasis cans: m1a, m1b, m2, and the drifter. You're only concerned with the drifter, but I've included the rest so that you can poke around and get a feel for what's going on. To use, save to a file, then: uudecode uncompress oasis.tar.Z tar xvf oasis.tar On our system, the "oasis root" is /usr/local/oasis, but you can make it anything you want. You may need to edit various scripts and config files if you change the root name. GETTING AROUND I suggest reading ./doc/oasis.qd, the quick-and-dirty guide to OASIS. It's as close as I've come to documenting OASIS. Keep in mind it was written 1.5 years ago, so it may be wrong in some places. But it documents most of the parts of the OASIS user interface you'll need. I call your attention to the fact that there are two commands to download data: "getdata" and "getlog". The "getdata" command is embedded into the "getoasis" C code that downloads onto the host system. "Getdata" was intentionally left out of the "help" screen -- it was intended to be invisible, and used only by the download code. Note that there are bugs that may cause you to lose data (see BUGS). In that case, you'll need to manually log on (through a terminal emulator) and retrieve those records. In that case, use the "getlog" command, documented in oasis.qd. Note that, while "getdata" always sends uuencoded data, "getlog" defaults to ASCII form, which is twice as large and isn't understood by the "extract" and "decode" commands. However, check the "parm" command on OASIS. There is a parameter called "ASCII" which tells "getlog" whether to send ASCII or uuencoded. The proper procedure, then, is to enter a "parm ascii 0" command, and then your "getlog". I apologize for the complication. After reading oasis.qd, you should look at the scripts in the bin directory. "oasis.cron" is the cron script I use for our system, but it includes several commands for moving data around for our scientists' use. For your use, you'll probably have a cron script that just calls "getdrift" at the appropriate interval, and leave out the other stuff. Look at "getdrift", though; you may need to change the values of variables and macros. It may be a little difficult to follow what's going on at first, but here's a quick rundown. The following assumes you're calling "getdrift" periodically from a cron script. "Getdrift" first looks for an appropriately named output file, based on year and julian day. The output file will be created in the "raw" (for raw data) directory. Next, "getoasis" is called with the TNC address of the can as its command-line parameter. In our case, this parameter is "mbari-5 v mbari-2". This tells getoasis to reach the can by issuing a "c mbari-5 v mbari-2" command, which for the TNC is shorthand for "connect mbari-5 via mbari-2"; in other words, connect to the TNC named mbari-5 using mbari-2 (our repeater on Mt. Toro) as a repeater. Since you won't be using a repeater, this should be changed to just "mbari-5". (Aside - I'll check with Gary Thurmond to see if we should change the TNC name. I'll get back to you on that). When the download is done, the script calls the "extract" command, with command-line parameters that identify the config file and the data file. By default, the config files are stored in directory ./cfg. If no command-line parameter is passed, "extract" defaults to file /usr/local/oasis/cfg/m1a.cfg if the data file name begins with m1a, /usr/local/oasis/cfg/m1b.cfg if the data file name begins with m1b, and similarly for m2 and d1 ("m" denoting mooring, "d" denoting drifter). The contents of the config file should be self-explanatory, but send email if you need help. In particular, it calls out the names of other calibration files for specific instruments, and it tells "extract" where to find the header files that "extract" will put at the beginning of the resulting data files. That's it for the downloading. Also of interested is the "decode" command. It's vary similar to "extract", and uses the same config and calibration files. Whereas "extract" creates output files in various subdirectories, "decode" just prints its output in human-readable form to stdout. Useful for quick checks on your data. BUGS The download program, "getoasis", sometimes hangs on our hp850. Haven't been able to track it down, but occasionally the TNC is also dead (don't know cause and effect). There are two ways to fix this: the traditional way is documented in the file ./doc/hangup.doc. I've also included a shell script that Pat Allen wrote for this purpose. It's in ./bin/kill_oasis. It would need to be edited for your environment. Also, it's new, and hasn't been tested in practice. As configured for our moorings, the poor old 87C196 in the OASIS controller is overworked. As a consequence, it occasionally gets out of sync with the XON/XOFF protocol from the TNC, and consequently a number of records get lost. The symptom of this is when "extract" or "decode" complains of a bad record header in . I've added a command-line parameter, -n, to "decode" to aid in finding out where. "decode -n" adds the record number to each record it prints out. By looking at record numbers, you'll be able to determine which records need to be downloaded again. Then, you can use your terminal emulator to log on directly to OASIS, and then use the "getlog" command with the record numbers you determined. OTHER I've mentioned logging onto OASIS with a terminal emulator several times. Although you can use "Procomm" or something similar on your PC, you'll then need to upload the resulting uuencoded data files to your workstation. I've found it more useful to use log on directly from the workstation. I have a public domain program called "pcomm" that I can send you if you wish. It's basically a "procomm" look-alike for unix workstations. Let me know if you want it. Take a little time to try out this stuff. Then contact me and we'll get a download schedule set up. The drifter can is available in the lab right now for you to try it on. Francisco will be using it for a drifter experiment in early September, and I'll need to take it down before then for new EPROMs. Call me with any problems.