Re: [Freeipmi-devel] md2/md5 ...

[Anand Babu]

,----[ Albert Chu <address@hidden> ]
| Last night I decided to program md2 just for fun.  I have a bit of
| debugging to do, but for the most part it was very easy.  Took around
| an hour or so to get the core code done.  I may have grossly
| overestimated the difficulting of programming md2/md5.  Perhaps we
| should write our own implementations for freeipmi.
`----

I think it is not worth having dependency on another library just for
MD2/5 algorithm. Doesn't matter if we copy and strip down a version
into our source just for IPMI.

Having lot of dependencies for system tools is a hassle for System
Administrators, especially when they are in trouble-shooting
mode. Most of the time, only when some thing breaks, network goes
down, or in a panic situation, a system administrator's attention will
be required. If has to meet package dependencies to bring up the
system, ...

-ab


----- Original Message -----
From: Albert Chu <address@hidden>
Date: Tuesday, December 16, 2003 4:02 pm
Subject: [Freeipmi-devel] md2/md5 ...

> Hey AB,
> 
> I'd like to look at adding md2 and md5 authentication to ipmipower
> semi-soon.  What do you see as the best method for adding md2/md5
> hashing algorithms to freeipmi ... 
> 
> A) Make a freeipmi rpm module dependency to some other common crypto
> library, like openssl or nss.
> B) Package some sub-library along with freeipmi ("libfoocrypto").  
> C) write our own md2/md5 hashing algorithms for freeipmi ...
> 
> Pros
> 
> A) Easiest, put development on other people
> 
> B) Pretty darn easy, have to update as other developers update, not a
> huge deal.
> 
> C) no restrictions on licensing/anything ... 
> 
> Cons
> 
> A) I haven't the slightest idea how popular/widespread packages like
> openssl or nss are.  Yeah they are reasonably popular, but I don't
> really know.
> 
> B) Some of the good crypto libraries seem to have funny licensing 
> issues.  Like we'd have to package their entire library, not just 
> the md2/md5
> algorithms.  
> 
> C) I think it'd be a good excercise to code these algorithms out, but
> they're not exactly a walk in the park.  I can't help but see porting
> issues.  Not to mention wasted effort, since there are tons of
> implementations out there already.
> 
> Al
> 
> --
> Albert Chu
> address@hidden
> Lawrence Livermore National Laboratory
> 
> 
> 
> _______________________________________________
> Freeipmi-devel mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/freeipmi-devel
> 



_______________________________________________
Freeipmi-devel mailing list
address@hidden
http://mail.nongnu.org/mailman/listinfo/freeipmi-devel


-- 
 _.|_ 
(_||_)
Free as in Freedom <www.gnu.org>