[915] in Kerberos_V5_Development
Re: pty handling in krlogind
daemon@ATHENA.MIT.EDU (Sam Hartman)
Thu Jul 20 15:02:33 1995
To: "Barry Jaspan" <bjaspan@cam.ov.com>
Cc: Ezra Peisach <epeisach@MIT.EDU>, Sam Hartman <hartmans@MIT.EDU>,
probe@MIT.EDU, eichin@MIT.EDU, krbdev@MIT.EDU
In-Reply-To: Your message of "Thu, 20 Jul 1995 11:00:59 EDT."
<199507201500.LAA23094@suan-la-chow-show.cam.ov.com>
Date: Thu, 20 Jul 1995 15:02:09 EDT
From: Sam Hartman <hartmans@MIT.EDU>
> Having hacked the rlogind code recently myself, I'd advise ripping out
> any code whose purpose you cannot identify or verify. If it is really
> needed for some platform, it will get added back (hopefully more
> portably and with comments explaining it better) soon enough. I
> suspect there is a lot of completely outdated code in there that NEVER
> needs to get added back...
>
> Barry
I talked to Richard, and basically, what happens is that if
you do a vhangup on Ultrix beforehand you don't get job control in the
rlogind-controlled pty. I suspect this has to do with the way ttys
are made controlling terminals on Ultrix; you could probably
dissociate yourself from the tty again afterwards, then reassociate
and it would work.
Removing the code altogether is probably a bad idea, because
it would work but lack any semblance of security; in this case,
leaving the code in would be more likely to get rapid attension if I
make the wrong decision. ( I do plan on testing so I end up giving
the rapid attension)
--Sam
