On Wed, Jul 16, 2008 at 8:58 AM, Damien Katz <[EMAIL PROTECTED]> wrote: > > There are two potential solutions here, the first is to send a browser ping > to keep the connection alive. Easy do do with HTTP 1.1 I think, just send an > empty HTTP chunk. The second is to make it impossible for the broken HTTP > request to kill the replication request. They aren't mutually exclusive, but > the more I think about it, the more I dislike the second solution. >
What is wrong with the second approach? The idea that the client needs to hang around for the (potentially hours-long) replication task seems flawed. I can't see why it makes sense to keep the HTTP request open during all of replication, under any circumstances. I'd really like to be able to navigate away from the page, or trigger replication from an unreliable connection, or from a remote laptop that I then shut and take on the bus with me... It seems superior to have the replication-triggering POST request return immediately with a "replication started" message, and maybe a link to a url where replication status can be found... This could be a GET on /_replication, which would list all active replication tasks, or some such. What do you think? -- Chris Anderson http://jchris.mfdz.com
