# Help with proxy\_protocol\_networks and HAPROXY

**URL:** <https://forums.percona.com/t/help-with-proxy-protocol-networks-and-haproxy/6691>\
**Category:** Percona XtraDB Cluster 5.x\
**Created:** [November 19, 2018, 11:58am UTC](https://forums.percona.com/t/help-with-proxy-protocol-networks-and-haproxy/6691 "2018-11-19T11:58:33Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![marcio](https://avatars.discourse-cdn.com/v4/letter/m/f19dbf/32.png) [@marcio](https://forums.percona.com/u/marcio)\
**Post date:** [November 19, 2018, 11:58am UTC](https://forums.percona.com/t/help-with-proxy-protocol-networks-and-haproxy/6691/1 "2018-11-19T11:58:33Z")

</div>

Hi !!

I have 3 node XtraDB Cluster (5.7.23) setup working, with HAPROXY acting as proxy with the configuration bellow:

db1: fd00:faca:0:a0::91  
db2: fd00:faca:0:a0::92  
db3: fd00:faca:0:a0::93  
haproxy: fd00:faca:0:a0::45

This is the current wsrep.cnf file that is working in all nodes:

proxy\_protocol\_networks = fd00:faca:0:a0::/118 # haproxy ip address  
bind\_address = fd00:faca:0:a0::9X # host ip address (X=1, 2 or 3 depending of host)

When I try change the ip address for proxy\_protocol\_networks, the HAPROXY can’t connect to host and show it’s not available;  
On both tests I can connect locally using sock or tcp, only HAPROXY still not connecting to host;  
I’ve tried the following settings without success:

1st test:  
proxy\_protocol\_networks = fd00:faca:0:a0::45, ::1 # haproxy ip address  
bind\_address = fd00:faca:0:a0::92 # host ip address (X=1, 2 or 3 depending of host)

2nd test:  
proxy\_protocol\_networks = ::1, fd00:faca:0:a0::45 # haproxy ip address  
bind\_address = fd00:faca:0:a0::92 # host ip address (X=1, 2 or 3 depending of host)

3rd test:  
proxy\_protocol\_networks = fd00:faca:0:a0::45 # haproxy ip address  
bind\_address = fd00:faca:0:a0::92 # host ip address (X=1, 2 or 3 depending of host)

I have followed this guide: [URL=“[https://www.percona.com/blog/2015/10/15/proxy-protocol-percona-xtradb-cluster-quick-guide/](https://www.percona.com/blog/2015/10/15/proxy-protocol-percona-xtradb-cluster-quick-guide/)”][https://www.percona.com/blog/2015/10...r-quick-guide/[/URL]](https://www.percona.com/blog/2015/10...r-quick-guide/%5B/URL%5D)

Here my haproxy.cfg:  
listen mysql\_cluster  
mode tcp  
bind :::3306  
maxconn 4000  
option tcplog  
timeout client 30m  
timeout server 30m  
balance leastconn  
option mysql-check user haproxy post-41  
default-server on-error mark-down inter 3s downinter 3s rise 3 fall 3 slowstart 30s maxconn 600 maxqueue 50  
server db1 fd00:faca:0:a0::91:3306 check send-proxy-v2 weight 100  
server db2 fd00:faca:0:a0::92:3306 check send-proxy-v2 weight 80  
server db3 fd00:faca:0:a0::93:3306 check send-proxy-v2 weight 100

Any idea why HAPROXY can’t connect to host ?  
Thank you.

---

<div class="post-metadata">

**Author:** ![marcio](https://avatars.discourse-cdn.com/v4/letter/m/f19dbf/32.png) [@marcio](https://forums.percona.com/u/marcio)\
**Post date:** [November 20, 2018, 5:53am UTC](https://forums.percona.com/t/help-with-proxy-protocol-networks-and-haproxy/6691/2 "2018-11-20T05:53:05Z")

</div>

Solved !!

The sysadmin give me the wrong HAPROXY ip address. After add the correct ip address, I can connect using socket and tcp/ip, and too use the extra port.  
Thank you.

---

<div class="post-metadata">

**Author:** ![lorraine.pocklington](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/lorraine.pocklington/32/37_2.png) [@lorraine.pocklington](https://forums.percona.com/u/lorraine.pocklington)\
**Post date:** [November 20, 2018, 6:29am UTC](https://forums.percona.com/t/help-with-proxy-protocol-networks-and-haproxy/6691/3 "2018-11-20T06:29:25Z")

</div>

Ah [marcio](https://www.percona.com/forums/member/41721-marcio) thanks for the update, appreciated - also it gives other users with the same problem something extra to check “just in case” so that’s great!