Scott Lawrence wrote:
> On Thu, 2009-09-24 at 16:20 -0400, Dale Worley wrote:
>> On Thu, 2009-09-24 at 16:07 -0400, Raymond Dans wrote:
>>> In the main stream, however, I'd like some thoughts on a better/more
>>> generic way of speculating on the size of the response that's going to
>>> be built so that we can reduce/eliminate the UtlString resizing.  Please
>>> note that this would still not be the ideal solution.  The underlying
>>> problem of using XML-RPC in these scenarios will still exist and a "new"
>>> approach will eventually be required.
>> All of what you say is good.  But we may be able to attack the problem
>> by observing that sipXconfig isn't going to display all 10,000
>> registrations at once.  So maybe it's possible to define an XML-RPC that
>> selects a subset of the registrations (e.g., "starting with seq. no. N,
>> continuing for M entries").  That would ipso facto have a small return
>> value.
> 
> Similarly, the problem with registry replication is just with the
> initial sync - there's no reason when that could not be a series of
> calls that each have some reasonable maximum number of entries.  Some
> careful thought needs to be done about not having failures delay startup
> too long, but the current strategy of doing the initial resync as just
> one call clearly won't scale... we knew that when we built it, but ...
> 

The problem is that sipXconfig currently reprocesses registrations and
removes duplicates. Unless it's somehow done on a server it needs to
retrieve all registrations. Granted it could do it in batches. Improving
XML/RPC IMDB methods performance would not hurt of course.

REST actually lends itself quite well for queries that would be needed to
retrieve or update a subset of IMDB records. Additional benefit would be
language independent access to IMDB. Changing the protocol of course does
not automatically solve the registrations issue.
D.

_______________________________________________
sipx-dev mailing list [email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev
sipXecs IP PBX -- http://www.sipfoundry.org/

Reply via email to