On 2/1/2012 9:42 AM, Guy Harrison wrote:
Hi Ryan,
On Wednesday 01 February 2012 13:43:32 Ryan Johnson wrote:
On 01/02/2012 5:46 AM, Guy Harrison wrote:
Hi Folks,
Can anyone help interpret this? I am fairly certain the problem lies
with IBM but I am no crypto expert. Is (for instance) the server
rejecting the connection because (say) it does not understand ECDSA?
Unfortunately I do not have an older instance of cygwin ssh to try that
theory out. The failure is recent. I upgraded my cygwin instances over
xmas.
My primary concern is that the latter (linux) connection (after ~~~)
may fail after a future upgrade.
I would definitely check with your local network security folks. When I
was last at IBM I had trouble connecting from a certain machine -- just
that one -- and nobody could figure out why. Finally, it turned out that
I had a lot of locales installed and the long list of supported
languages announced by my ssh client triggered some firewall rule.
Unfortunately I forgot to mention the problem occurs both from my home
network and via my work network (which I could easily have believed was at
fault - they've messed with it a lot recently). The ~~~ linux box above
connects via my home network but I have an aix box at work that also
connects successfully whereas work cygwin (that's on XP) fails in the same
fashion as my original post.
So you're defining a successful connection as one where any key file is
ignored/invalidated and you're left to login with your password? That's
what you're showing with the Linux machine. If that's the benchmark, have
you tried eliminating your keys on your Cygwin machine to see if you get
to the same point as Linux?
--
Larry
_____________________________________________________________________
A: Yes.
> Q: Are you sure?
>> A: Because it reverses the logical flow of conversation.
>>> Q: Why is top posting annoying in email?
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
