couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brian Candler <B.Cand...@pobox.com>
Subject Re: svn commit: r804427 - in /couchdb/trunk: etc/couchdb/default.ini.tpl.in share/www/script/test/delayed_commits.js src/couchdb/couch_db.erl src/couchdb/couch_httpd_db.erl
Date Thu, 20 Aug 2009 21:21:04 GMT
On Thu, Aug 20, 2009 at 12:52:27PM -0700, Chris Anderson wrote:
> > e.g. a PUT request could specify max_wait=2000 (if not specified, use a
> > default value from the ini file). Subsequent requests could specify their
> > own max_wait params, and a full commit would occur when the earliest of
> > these times occurs. max_wait=0 would then replace the x-couch-full-commit:
> > header, which seems like a bit of a frig to me anyway.
> 
> I like the approach you've outlined here. I think the default could
> eventually be to allow CouchDB to be adaptive for high throughput, but
> we'd have to write that code first.
> 
> Once we have all this, batch=ok could still be meaningful, when the
> client wants to be able to send a bunch of serial writes to CouchDB
> and is fine with the reduced guarantees that 202 Accepted gives.

Yes, I agree that batch=ok makes sense if it means "just give me a 202
response straight away", and the data will probably make it to disk some
time in the future.

Mime
View raw message