stever@videovax.Tek.COM (Steven E. Rice, P.E.) (12/23/87)
Recently, Yuan Chang and Keith Doyle reported problems with Amiga Kermit. In article <1320@uhccux.UUCP>, Yuan Chang (cs313s02@uhccux.UUCP) writes: > Hi. Does anyone use Amiga's version of C-Kermit? I recently got > a version (dated Mar 19, 1986) off our local DEC-20, and it seems to run > correctly. It'll transfer files with my AT in even parity mode, but not > in no parity mode. But that wasn't the big problem. The problem is that > it somehow loose characters very consistently. For example, if I download > (or upload, if you will) a 1000 byte file from my AT, it always ends > up as 998 bytes (I haven't tried to see if different files of the same > length gives the same result). And I chose type 3 for block-check. > Any ideas?!? > > Systems: AT clone running MS-Kermit 2.29 (never had a problem) > Amiga 512K running C-Kermit 4C(058) > Link: Null modem. All baudrates yielded same results. > Cable: Short (about 8 feet) First, the problem may not be Kermit, at all! The IBM PC uses CR/LF (the sequence, carriage return followed by line feed) as its record terminator for text files, while the Amiga follows the UNIX convention and uses NL (newline = LF) only. This can cause a difference in byte counts on text files. One way to check if this is the culprit is to transfer the file from the PC to the Amiga and then transfer it back to the PC. Compare the two files on the PC to see if they are the same. A second possibility for trouble is that you are using Kermit 4D(058). I have had version 4D(060) on my Amiga for quite a while (I would have to look at the dates on the disks to find out just how long, and they are at home. . .). I have had VERY little trouble with this version -- in fact, so little that while I built version 4D(061) for our VAX (and use it on the VAX), I have not bothered to compile version 4D(061) for the Amiga. In article <1928@cadovax.UUCP>, Keith Doyle (keithd@cadovax.UUCP) responded to Yuan's posting: > I had similar problems when using Amiga C-Kermit with a vax. It seemed that > odd byte-length files ended up with the last byte dropped. My solution: > I gave up on C-Kermit. Other Amiga based kermit programs had no problem, > (though they weren't full featured). I'd like to go back to using C-Kermit, > but only if I could depend on it to work. I have not had problems of this kind with version 4D(060). There is one minor bug in 4D(060) that can be annoying -- the trailing carriage return of some (not all, not even most; maybe 2%) of the files is lost. This bug has lurked in C Kermit since its inception, and has finally been tracked down and squashed in the new (still experimental and not yet released) version 4E(067). In article <2966@cbmvax.UUCP>, Carolyn Scheppner (carolyn@cbmvax.UUCP) replied to Yuan: > Have you tried the Kermit in VT100 (available on Fish disks). > I've recently started plugging my VAX line into my A500 and use VT100's > kermit to download and upload files and directories. I don't know > much about kermit, but I was shown how to use it in server image mode > (kermit -x -i on the VAX), and use VT100's Kermit Get and Kermit Send > selections to do the transfers. Works great. (VAX line is 9600 baud > 8 bits no parity). The problem with VT100's Kermit (at least, the last I knew -- version 2.4) is that it does not accept the "take" command. With a properly- constructed "take" file and a short Amiga script file, you can transfer not just a file or files, but an entire directory, complete with its internal structure. This is very useful when dealing with a collection of files whose relationships are expressed in hierarchically-arranged directories. It also allows me to go somewhere else and do something else for an hour or more while the transfer grinds along (2400 baud), because I don't have to keep starting file transfers. I have a set of C-shell scripts that I use to facilitate file transfers between my Amiga and our VAX. These construct a "take" file and a "<xxx>.check" directory parallel to the original directory ("<xxx>" denotes an arbitrary directory name). Then, when I initiate a transfer, it copies the files to my Amiga and back to the VAX. On the VAX, I can then do a "diff -cr <xxx> <xxx>.check | more" to see if there are any differences. I have offered these scripts before, and had a few people ask for them. If you would like them, let me know. (They are small, and I am quite willing to email them.) I am not willing to email Amiga Kermit (it is large!), but I would be happy to make a copy of the disks (source and executable) for anyone who wants them. If you are interested, send email to me. Our VAX will be off the net from December 24th to January 3rd because of power system and internal network maintenance. We have been promised that all mail will be held for us by our gateway site, tektronix. If you send email and don't get a reply by January 10th, follow up on your request (that will give me a week to get caught up after the holidays). Steve Rice ----------------------------------------------------------------------------- * JESUS Is The REASON For The SEASON! (Merry Christmas and Happy New Year) * new: stever@videovax.tv.Tek.com old: {decvax | hplabs | ihnp4 | uw-beaver}!tektronix!videovax!stever
A4422DAE@AWIUNI11.BITNET (Konrad Neuwirth) (10/03/88)
As I am in the unhappy state of owning just one drive and
512 K, and needing Kermit very badly, could please someone
send me the LAST version uuencoded. I know there is a version
on a fishdisk, but it doesn't fulfill all my needs.
I will get a harddisk Real Soon Now!
SIGNED, AS ALWAYS
I^ /I +----
I ^/ I +--
I I +----
"SORRY FOR LIVING, I WILL NEVER DO IT AGAIN"
KONRAD NEUWIRTH (A4422DAE AT AWIUNI11) (KONRAD ON RELAY)