On Mon, 12 Dec 2016 09:50:05 -0700, Jack J. Woehr wrote:
>Paul Gilmartin wrote:
>> No "walls", just pitfalls. Should I submit an SR about the "script" glory
>> hole?
>
>Possibly. Venkat cannot be the only person to run up against the limitation of
>TN3270 sessions presenting no ptty. I've
>experienced it.
>
I belive that the 3270 OMVS command does create a PTY, but the function ssh
uses (tcsetattr()?) fails to mask passwords on a 3270, Ssh is aware of this,
so it refuses to prompt for a password when it recognizes that the PTY is from a
3270. Log from 3270 OMVS:
user@OS/390.25.00: tty
/dev/ttyp0000
user@OS/390.25.00: script
Script command is started. The file is typescript.
...
you have mail ...
user@OS/390.25.00: tty
/dev/ttyp0001
... but ssh doesn't recognize the script PTY as a 3270, so prompts for a
password.
It appears that FTP uses a GETPASS interface, long deprecated by POSIX
to prompt for a password. But if I call FTP under script it prompts for a
password, failing to mask it.
IBM really ought to fix tcsetattr() and remove all these problems. I suspect
that if I open an SR they'll modify script to present a 3270 terminal to
clients, leaving tcsetattr() broken as they did for an SR I submitted
years ago.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
