lamy@sobeco.com (Jean-Francois Lamy) (07/25/90)
After starting slattach from sh, csh or ksh on a RC3230 or an M/120 running
4.50, one sees the following (mis)behaviour:
- "tty" run on the shell reports the tty on which the login took place, as it
should.
- ps, however, reports that no terminal is associated with the shell or any
process started under it.
- attempts to fire up programs that expect a controlling terminal (e.g. "su")
will fail ("su" will actually give a Memory Fault, what fun, when it can't
open /dev/tty -- ought to check return values of open, neh?)
Note that slattach is not being performed on the port where the login took
place, and that it does not matter whether the slattach-ed port is a Mips
CPU port or a tty port on our local communication board.
Fixed in 4.50B, Roger :-)?
Jean-Francois Lamy lamy@sobeco.com, uunet!sobeco!lamy
Groupe Sobeco, 505 ouest, bd Rene-Levesque, Montreal Canada H2Z 1Y7
p.s. anyone done Van Jacobsen compressed header slip under 4.50 before I go
ahead and do it?lawc@super.ece.jhu.edu (Chiu-Tai Law) (07/27/90)
In article <1990Jul25.132828.1819@sobeco.com> lamy@sobeco.com (Jean-Francois Lamy) writes: >After starting slattach from sh, csh or ksh on a RC3230 or an M/120 running >4.50, one sees the following (mis)behaviour: > >- "tty" run on the shell reports the tty on which the login took place, as it > should. >- ps, however, reports that no terminal is associated with the shell or any > process started under it. It sounds like it is the same problem that I experienced when I tried to compile the xterm program in X11R3 and X11R4 under bsd environment. I have discovered this a long while ago but do not time to write a program to demonstrate that there are some strange characteristics in the bsd tty driver and report to mips. It seems to me that the description in TTY(4-BSD) -- if a process which has no control terminal opens a terminal file, then that terminal file becomes the control terminal for that process -- does not hold. If I define USE_SYSV_TERMIO in the compiling xterm, the problem of missing control tty is got. Are you compiling under bsd environment? >- attempts to fire up programs that expect a controlling terminal (e.g. "su") > will fail ("su" will actually give a Memory Fault, what fun, when it can't > open /dev/tty -- ought to check return values of open, neh?) add script to the list.A By the way, our system (M/120) still runs 4.01. When will the upgrade 4.50 come? -- Chiu Tai Law lawc@super.ece.jhu.edu (301) 338-7976 lawc@alpha.ece.jhu.edu