FYI,
I've committed Adam Allen's patch
https://sourceforge.net/tracker/index.php?func=detail&aid=1022628&group_id=33947&atid=409712
to CVS. This patch allows control over whether to allow the acceptance
of new host keys or not. By default, new host keys are accepted. Thanks
Adam!
I've also added a check for "Host key verification failed" in the
initial banner exchange. Since this could be an indication of a
man-in-the-middle attack, it raises a HostVerificationFailure exception.
Unless captured, this will terminate any connection without providing
any user over-ride capability. Instead, the warning banner is produced
along with the exception forcing the proper corrective action which is
to verify/correct the keys. This is consistent with, but a little more
gracefull than, the current behavior which requires a user response
before bombing out (at least with strict host checking on).
Note that these changes are in the default prompt_callback function, so
this behavior can be over-ridden by a caller specified prompt_callback.
If anyone is using custom prompt_callback functions, I'd appreciate an
education as to why it was necessary for you to use one.
I do not plan a new file release for these changes. If you want these
features, please pull the latest from anonymous cvs. My target for the
next file release is an expect() function comparable to telnetlib's. I'm
still feeling my way around the details of the code, but my goal is for
PySSH to be a drop-in replacement (e.g. API compatible) with telnetlib.
Comments, suggestions, and cautions are appreciated.
mwa
--
Mark W. Alexander
sl...@do...
The contents of this message authored by Mark W. Alexander are released under
the Creative Commons Attribution-NonCommercial license. Copyright of quoted
materials, if any, are retained by the original author(s).
http://creativecommons.org/licenses/by-nc/2.0/
|
