On 24/09/2019 10:02, Dave Thorn wrote: > On Sun, Sep 01, 2019 at 08:51:47AM +0000, Mark Thomas wrote: >> On August 31, 2019 7:31:12 AM UTC, Dave Thorn <da...@fysh.org.INVALID> wrote: >> >>> There is also a haproxy between client and server. You wouldn't think >>> it goes fiddling around in the middle of a response, thought. I'm >>> running some tests today that go straight to the server, skipping the >>> proxy, but again with the intermittency... >> >> Indeed. Chunking is sufficiently fundamental that I'm surprised that >> any component has a bug in this area. > > (Following up for the record, even though this doesn't (seem to) relate > to tomcat.) > > I have spent the last nearly-a-month running tests on this, and it > looks like the problem lies in haproxy.
Thanks for the update. I'd agree that the evidence seems to be pointing towards haproxy at the moment. If that changes then do let us know. One small positive to come of this was that I reviewed Tomcat's chunking code and made a small optimisation. The code is now a little cleaner, a little easier to maintain and might even be marginally faster (although I doubt any improvement is measurable). Thanks again for keeping us updated. Mark > > When using haproxy version 1.7.something I haven't been able to > trigger the error. When using version 2.0.something I have triggered > it a handful of times. > > > This issue on github at least suggests it's not just me having some > kind of intermittent issue with chunking in haproxy: > > https://github.com/haproxy/haproxy/issues/171 > > --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
