OK, I think I have stumbled across an ARM issue... When I connect w/ the
'-nt' flag, it just hangs there, with lshd taking up 98% CPU forever (20
minutes so far). Looks like :
Client :
[jason@netwinder jason]$ /usr/local/bin/lsh -v -nt -p 4711 204.123.163.98
Client version: SSH-2.0-lsh_0.0 lsh - a free ssh
Server version: SSH-2.0-lshd_0.0 lsh - a free ssh
handle_dh_reply()
Received debug: Key exchange successful!
Garbage collecting while busy...
Objects alive: 92, garbage collected: 26
Password for jason:
User authentication successful.
Server :
server_initiate()
Client version: SSH-2.0-lsh_0.0 lsh - a free ssh
handle_dh_init()
Received debug: Key exchange successful!
I know there were some issues with gmp on ARM, so maybe its hanging in the
library code, but I don't know.
Any ideas?
Jason.
ps - Sorry to keep pestering everyone, but ARM Linux is a pretty cool
platform ($25 or so processors w/ good performance = a good thing) and it
would be wonderful to have secure access to these little boxes!
----------
>From: [EMAIL PROTECTED] (Niels M�ller)
>To: "Jason Titus" <[EMAIL PROTECTED]>
>Subject: Re: lsh 1999-03-07
>Date: Mon, Mar 8, 1999, 5:20 AM
>
> "Jason Titus" <[EMAIL PROTECTED]> writes:
>
>> Well, the 1999-03-07 snapshot still does this on ARM Linux 2.2.1 when
>> recieving a connection (from the same host through TCP-IP) -
>>
>> server_initiate()
>> Client version: SSH-2.0-lsh_0.0 lsh - a free ssh
>> handle_dh_init()
>> Client requesting a tty...
>> granted.
>> tty_setctty: unlockpt() failed,
>> (errno = 14): Bad address
>> lshd: Can't set controlling tty for child!
>> Child 2467 died with exit code 1.
>
> Strange. The code in the latest snapshots should have an #if 0 around
> that call to unlockpt().
>
>> I'll keep hacking around with it....
>
> Please do. Note that you can tell the client not to request a pty, by
> running lsh -nt.
>
> /Niels
>
