[5422] in Kerberos
Re: Solaris2 kerberos problem: no access to tty; thus no job control . .
daemon@ATHENA.MIT.EDU (Christopher Hylands)
Mon Jun 26 15:41:36 1995
To: kerberos@MIT.EDU
Date: 26 Jun 1995 19:27:30 GMT
From: cxh@markov.berkeley.edu (Christopher Hylands)
I wrote:
I've run up against the following bug under Solaris2.4:
When I rlogin to a Solaris2 host running klogind, I get the message:
Warning: no access to tty; thus no job control in this shell...
This message comes from /bin/csh. Running a trace shows that just
before the message is output, a TIOCGPGRP ioctol is failing:
ioctl(18, TCGETA, 0xEFFFFC5C) = 0
ioctl(18, TIOCGPGRP, 0x00047A88) Err#25 ENOTTY
ioctl(17, TIOCLGET, 0xEFFFFB5C) = 0
write(17, " W a r n i n g : n o ".., 64) = 64
The termio man page says:
ENOTTY The file associated with fildes is not a terminal.
I get a similar message with tcsh-6.05, where tcgetpgrp() is failing.
I get this message with Cygnus' CNS code, and with a local KerberosIV
installation. My guess is that this is some sort of nasty Solaris
botch.
Interestingly, the CNS telnetd works fine, I have job control and
everything. Both telnetd and klogind are using the same login.krb
binary.
-----
This was a real bear to figure out. It turns out that /bin/csh was
unable to get any pagegroup information because there was no tty
associated with the log in. Here's how I found out:
If I set my shell to /usr/local/bin/csh.tst, which contains
#!/bin/sh
truss -o /tmp/tr csh
and add /usr/local/bin/csh.tst to <CODE>/etc/shells
Then /tmp/tr contains
fcntl(2, F_DUPFD, 0x00000012) = 18
getpid() = 14302 [14301]
fcntl(18, F_SETFD, 0x00000001) = 0
[. . .]
ioctl(18, TCGETA, 0xEFFFFC5C) = 0
ioctl(18, TIOCGPGRP, 0x00047A88) Err#25 ENOTTY
ioctl(17, TIOCLGET, 0xEFFFFB5C) = 0
write(17, " W a r n i n g : n o ".., 64) = 64
The termio man page says:
ENOTTY The file associated with fildes is not a terminal.
I got something similar with tcsh. After looking around at the Cygnus
CNS source code, I realized that the fix was to edit login_tty() in
our local kerberosIV/libexec/telnetd/sys_term.c and close and reopen
the tty. I'm not sure why this works, but it does. (The changes
below are relative to a fairly hacked up version of kerberosIV, so
they might not work as is, but you get the idea.)
rcsdiff -c sys_term.c
===================================================================
RCS file: RCS/sys_term.c,v
retrieving revision 1.8
diff -c -r1.8 sys_term.c
*** /tmp/T0a001jT Sun Jun 25 17:35:04 1995
--- sys_term.c Sun Jun 25 16:40:28 1995
***************
*** 1358,1363 ****
--- 1358,1365 ----
# endif /* __hpux || SOLARIS */
# ifndef SOLARIS
close(open(ptyline, O_RDWR));
+ # else
+ close(open(line, O_RDWR));
# endif /* !SOLARIS */
# endif
if (t != 0)
------------------------------------------------------------------------
I also fixed a problem where Stty showed an output speed of 0 on a DEC
mips machine
stty
new tty, input speed 38400 baud output speed 0 baud ; -tabs crtbs crterase crtkill ctlecho
decctlq
The fix is to change kerberosIV/usr.bin/kerberos/rlogin.c:
rcsdiff -c rlogin.c
===================================================================
RCS file: RCS/rlogin.c,v
retrieving revision 1.9
diff -c -r1.9 rlogin.c
*** /tmp/T0a001kU Sun Jun 25 17:45:54 1995
--- rlogin.c Sun Jun 25 16:39:39 1995
***************
*** 468,474 ****
--- 468,478 ----
#ifdef __hpux
return (speeds[tt.c_cflag & CBAUD]);
#else /* !__hpux */
+ #if defined(__sun) && (defined(__SVR4) || defined(__svr4__))
+ return (speeds[(int)cfgetospeed(&tt)]);
+ #else
return (speeds[(int)cfgetispeed(&tt)]);
+ #endif
#endif /* __hpux */
#endif /* MIPSEL */
}
---------------------------------------------------------------------------
Also for our hacked up version of kerberosIV, we were getting
/etc/motd and 'you have mail' echoed twice. The solution was to edit
kerberosIV/usr.bin/login/login.c and #ifdef out the lines that call
motd() and that check for mail. Under So-lose-us, these are taken
care of in /etc/.login (I think).
My current problem is that connecting from Macs results in some sort
of control character problem, where pine behaves oddly. I'm also
seeing CR/LF problems.
-Christopher
--
Christopher Hylands University of California
cxh@eecs.berkeley.edu 558 Cory Hall #1770
ph: (510)643-9841 fax:(510)642-2739 Berkeley, CA 94720-1770
home: (510)526-4010 (if busy -4068) (Office: 493 Cory. US Mail: 558 Cory.)
