Settable minimum RSA key sizes on the client end for legacy devices.

[Steve Sether]

Yes, I considered re-compiling, but it just seemed like an enormous pain 
to have to go through the source and figure out where the key size limit 
was.  It also occurred to me that I'm likely not the only one with this 
problem, but many just didn't understand what the underlying problem 
was, and more wouldn't know where to go to talk about addressing it.  
95% of the administrators of the world are just going to enable telnet, 
and be done with it.  You have to remember that one person reporting a 
bug or requested feature represents many, many more that never caught 
it, or didn't know how to report it, or (as above) found another workaround.

I also tried looking for other clients that might support lower sized 
keys.  putty seems to blow up because of the buggy ssh implementation of 
the server that openSSH kindly deals with. There's some other 
implementations that I didn't want to pay for just to see if they worked.

On 12/26/19 8:08 AM, Philipp Marek wrote:
>
>>> I'd rather not dredge up a big fight, but I _would_ like to express a
>>> desire for some form of overriding the minimum key size.
>>
>> This can be done by recompiling if necessary. This restriction has been
>> a pain for me at times but honestly I think it's for the best that it's
>> been done.
>
> I may be alone with that opinion, but for such things I've always hoped
> for a global _variable_ whose location is available as a dynamic symbol
> in the ELF - so instead of getting all the sources (and all their
> dependencies, resp. the headers etc.) and reconfiguring (with all
> required research to get the right options) and recompiling, a 1-minute
> session with a hex editor to patch the 2 bytes would be enough...
>
>
> Yeah, some distributions make recompilation much easier (Debian has
> "apt-get source") - but still it's much more work than switching a few
> bytes.