rlogin ---> no controlling tty ? (bug ?)

Thread View: comp.sys.next.sysadmin

2 messages

[T] Back to Threads [V] Tree View [G] Group Articles

2 total messages Started by Huisjes MJ Wed, 30 Sep 1992 16:15

#1644

Author: Huisjes MJ
Date: Wed, 30 Sep 1992 16:15

48 lines
1627 bytes

Hi,

For some reason when you do a rlogin, the rlogind on the other side does
not set your controlling tty. This can be repeated by any user with /bin/sh as
login shell :

	$ rlogin <some machine>
	..
	$ tty
	/dev/ttyp0
	$ echo hello > /dev/tty
	/dev/tty: cannot create
	$ ls -l /dev/tty
	crw-rw-rw-  1 root              2,   0  Jul  30 22:50 /dev/tty
	$

The actual error (errno) is not "cannot create" but :
	
	$ csh
	machine-name> echo hello > /dev/tty
	/dev/tty: No such device or address
	machine-name> exit
	$ ps -l $$
        F  UID   PID  PPID CP PRI BASE VSIZE RSIZE WCHAN STAT TT  TIME COMMAND
        1   32  1150  1148  0  10   10 8.64M  144K     0  S   ?   0:00 -sh (sh)
							     ^^^

	$

e.g. no controlling tty. You can acquire one because the first terminal
you open will become you controlling tty.
	
	$ echo hello > `tty`
	hello
	$ echo hello > /dev/tty
	hello
	$

For /bin/sh this is no big problem just add "echo -n > `tty`" to
your .profile . But tcsh will refuse to acquire a new controlling tty
(which is ofcourse correct).

This happens in both NS 2.x and NS 3.0.

This is a bug isn't it ?
-- 
Maarten Huisjes.	mjh@cs.vu.nl (192.31.231.42)
			(..!uunet.uu.net!cs.vu.nl!mjh)

Re: rlogin ---> no controlling tty ? (bug ?)

#1733

Author: Eric P. Scott
Date: Sun, 04 Oct 1992 05:01

27 lines
1642 bytes

In article <BvEFu2.8xI@cs.vu.nl> mjh@cs.vu.nl (Huisjes MJ) writes:
>For some reason when you do a rlogin, the rlogind on the other side does
>not set your controlling tty.

>This happens in both NS 2.x and NS 3.0.
>
>This is a bug isn't it ?

Yes, it's a bug.  And it's a bug that's been known for a long time.
The fix for 2.x is available by anonymous ftp from sutro.sfsu.edu
[130.212.15.230] as pub/2.0-fixed-rlogind.tar.Z

I haven't investigated the behavior under 3.0, but I'm not the
least bit surprised to hear of yet-another-known-but-unfixed-bug
in the networking software.  3.0 seems to have quite a lot of
them, and NeXT neither seems interested in rectifying the
situation nor releasing the relevant sources to allow others to
do so.  Fortunately, I'm in no hurry to "upgrade," and in our
heterogeneous/multivendor environment, I can easily find a non-
NeXT machine where the bundled software works the way it's
supposed to (and the /usr/include files haven't been bulldozed,
in case I really do need to recompile something from the
publicly-available sources).

"Really looking forward to 3.1"  (yeah, right)

					-=EPS=-

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads