On Wed, Apr 18, 2001 at 10:47:04AM +0800, Julius C. Duque wrote:
> On Tue, 17 Apr 2001, Thomas Seifert wrote:
> > > debug1: Allocating pty.
> > > stat(/dev/ttyp7b¾À¨`e) failed: No such file or directory
> > > debug1: Calling cleanup 0x805387c(0x80dab80)
> > > debug1: pty_cleanup_proc: /dev/ttyp7b¾À¨`e
> > > chown /dev/ttyp7b¾À¨`e 0 0 failed: No such file or directory
> > > chmod /dev/ttyp7b¾À¨`e 0666 failed: No such file or directory
> > > debug1: Calling cleanup 0x8060334(0x0)
> Getting "connection closed" could be caused by:
> 1. Your remote machine is not allowed to access your server
> 2. The ssh daemon (sshd) is not running.
While those are good guesses in the general case, I think the error
messages above point to a more serious problem.
The pseudo-terminal devices on a Red Hat system where I have sshd
(non-commercial ssh2) running are of the form "/dev/ttyp3". So it
appears that Thomas's ptys have some garbage appended to them --
in this case, "/dev/ttyp7" is probably correct, but the "b¾À¨`e"
after it looks like leftover data.
Looks like a bug in sshd. My first guess would be that something wrote
"/dev/ttyp7" to the first 10 bytes of a string which contained garbage,
and then forgot to write the '\0' in the 11th byte to terminate the
string....
