toxicity parameter ignored

[tboloo]

My setup:
toxicity 2.1.4 in docker, redis 5.0.7 also in docker.

I have created proxy to redis, and later added toxicity with probablity 0.7 to it. Inspect shows that proxy was configured correctly,

/go/bin # ./toxiproxy-cli  inspect redis
Name: redis     Listen: [::]:26379      Upstream: redis:6379
======================================================================
Upstream toxics:
Proxy has no Upstream toxics enabled.

Downstream toxics:
timeout:        type=timeout    stream=downstream       toxicity=0.70   attributes=[    timeout=5000    ]

With such proxy I would expect that 7 out of 10 times the connection would timeout, however in reality it does not timeout at all, or at most once:

$ docker run -it --rm --net toxic redis:5.0.7 redis-cli -h proxy -p 26379
proxy:26379> get test
"100"
proxy:26379> get test2
"abcd"
proxy:26379> get adasd
(nil)
proxy:26379> set asdasd sadasd
OK
proxy:26379> get efas
(nil)
proxy:26379> get t
(nil)
proxy:26379> get t
(nil)
proxy:26379> get t
(nil)
proxy:26379> get t
(nil)
proxy:26379> get t
(nil)
proxy:26379> get t
(nil)
proxy:26379> get t
(nil)
proxy:26379>

[tboloo]

My bad, redis-cli keeps connection open, so obviously it will affect at most one operation...

[zhengziyi0117]

I also encountered the same problem. Is there a way to simulate timeout in an established connection?