Raphaël Valyi wrote:
Hi,
rev #5665 seems to result in Rails printing HTTP headers instead of
the body of the pages in some cases. I tested with rev #5664 and it
was OK.
Unfortunately this regression doesn't seems to show up on simple
scaffold Rails app, but I've it on my real life Rails app.
May be I find more time to narrow the range a bit further. but there
are lots of changes in 5665 so may be that's not so easy. I
Anyway it's worth a check and don't be surprise if you get other kind
of trouble after rev #5665. Regards,
This was my fault, I wasn't flipping a buffer before trying to flush it.
It's fixed in 5670 and I added a test for a small socket write followed
by a large buffer-flushing write.
- Charlie
---------------------------------------------------------------------
To unsubscribe from this list please visit:
http://xircles.codehaus.org/manage_email
