Score:0

HAProxy closes connection on exact client timeout, even though there was data flowing in the middle of that period

hn flag

I'd like to have long lived connections(if possible infinite), from app containers to RabbitMQ behind HAProxy 2.8.1.

The problem is that the HAProxy severs the connection at exact client timeout, even though there was communication in the middle of that period.

If the timeout is set to 10 minutes and the RabbitMQ communication occurs at minute 5 of that period, the timeout isn't extended to minute 15, instead it's severed at minute 10.

global
  log stdout format raw local0

defaults
  log global
  mode tcp
  option tcplog  
  timeout connect  5s
  timeout client  10m
  timeout server  10m

resolvers docker
  nameserver dns1 127.0.0.11:53  
  hold valid  10s  
  resolve_retries  30
  timeout retry  1s
  
frontend stats
  mode http
  bind *:8404
  stats enable
  stats uri /
  stats refresh 10s

listen rabbitmq-cluster
  bind *:5672
  mode tcp
  balance roundrobin
  option tcp-check
  option tcpka  
  server rabbit1 rmq-ubuntu-a:5672 send-proxy-v2 check inter 5s rise 2 fall 3 resolvers docker
  server rabbit2 rmq-ubuntu-b:5672 send-proxy-v2 check inter 5s rise 2 fall 3 resolvers docker
  server rabbit3 rmq-ubuntu-c:5672 send-proxy-v2 check inter 5s rise 2 fall 3 resolvers docker
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.