Atsuhiko Yamanaka schrieb: > Hi, > > +-From: Paul Ebermann <paul-eberm...@gmx.de> -- > |_Date: Tue, 10 May 2011 17:17:54 +0200 _______ > | > |Yes, I understand that one normally don't want this. I think > |I described a usecase where this might be useful (not with the real host > key), > |though. > > I could not understand the advantage of host-based auth over publickey auth > in that > use-case. What is the impossible thing with publickey auth, but possible with > host-based auth.
The main difference is that we only need one key for the application/installation instead of one key for each user, i.e. it is easier to configure on the server side. I don't really want to persuade you that this must be included in JSch (I think you are right that this is normally not too useful, and public-key auth is better). More important is that applications should be able to implement this (and other custom authentication methods) without being in the com.jcraft.jsch package. See my last mail for some modifications which I think would be necessary/useful. We could put it as an example for "Extending JSch" to the wiki, then, with big warnings about the security implications. Paŭlo ------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ JSch-users mailing list JSch-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jsch-users
