Hi again Cyril, On Sat, Jan 23, 2016 at 11:55:27PM +0100, Cyril Bonté wrote: > Here is the sample configuration I use : > global > tune.ssl.default-dh-param 2048 > > listen test > mode http > bind :9443 ssl crt localhost.pem > server s localhost:80 > > > At commit 5506e3f8 (BUG/MINOR: stats: correctly set the request/response > analysers) > > $ ab -n1000 -c1 -k https://127.0.0.1:9443/1mb > Time taken for tests: 2.403 seconds > Requests per second: 416.08 [#/sec] (mean) > > The next commit (5be2f352 MAJOR: polling: centralize calls to I/O > callbacks) seems to break keep-alive, I can't test any commit until > e583ea58 where I have this result (still true with current 1.7 head) : > > $ ab -n1000 -c1 -k https://127.0.0.1:9443/1mb > Time taken for tests: 2.933 second > Requests per second: 340.97 [#/sec] (mean)
Tried to reproduce here, but for now I can't see any difference. All I can see is that ab introduces sleeps in https that do not happen in http, so I can't reach that level with a concurrency of 1 and for this I have to chain two haproxy (one ssl client and one ssl server). Let's declare it's late and I'm overseeing something obvious. Cheers, Willy
