What to do about HPSS protocol backwards compatibility for fasterpush?

Matthew D. Fuller fullermd at over-yonder.net
Thu Oct 18 02:01:07 BST 2007


On Thu, Oct 18, 2007 at 10:32:53AM +1000 I heard the voice of
Andrew Bennetts, and lo! it spake thus:
> > Aaron Bentley wrote:
> > 
> > > This could either be inferred from the version, or be explicit
> > > like smtp.
> 
> Well, I don't really want to have any handshake.  But if I did, I'd
> be in favour of having defined protocol versions like that.  More
> fine-grained capability reporting seems like a YAGNI, and a
> dangerous one for those sorts of reasons.

I tend to like a combination of the two, where we have
"occasionally"-rolled protocol versions (probably major.minor, so we
can use major for flag days that remove caps), and list additional
capabilities beyond the set in that version.  That way we get easy
fallback and enumeration of caps, without ending up with an
eternally-growing list of individually-enumerated bits.


-- 
Matthew Fuller     (MF4839)   |  fullermd at over-yonder.net
Systems/Network Administrator |  http://www.over-yonder.net/~fullermd/
           On the Internet, nobody can hear you scream.



More information about the bazaar mailing list