That was it. During the cluster install process we appear to have missed a step 
where we configure the interfaces on PF2 and PF3. Once this was done, the 
HAProxy stopped complaining.

Basically, we did exactly what was in this email to resolve the issue. But I 
didn’t have this email as our spam filter figured it was spam.

Thanks for the help! It’s very appreciated!

Cheers,
Chris

From: Ludovic Zammit <[email protected]>
Sent: April 5, 2021 9:31 AM
To: [email protected]
Cc: Chris Crawford <[email protected]>
Subject: Re: [PacketFence-users] HAProxy-Portal Errors

✉ External message: Use caution.
Hello Chris,

It means that your interfaces does not communicate properly.

Make sure to check the connectivity between your server interface like this:

From PF1:

ping -I ens160 PF2-MGNT
ping -I ens160 PF3-MGNT

ping -I ens192 PF2-REG
ping -I ens192 PF3-REG

ping -I ens224 PF2-REG
ping -I ens224 PF3-REG

From PF2:

ping -I ens160 PF1-MGNT
ping -I ens160 PF3-MGNT

ping -I ens192 PF1-REG
ping -I ens192 PF3-REG

ping -I ens224 PF1-REG
ping -I ens224 PF3-REG

From PF3:

ping -I ens160 PF2-MGNT
ping -I ens160 PF1-MGNT

ping -I ens192 PF2-REG
ping -I ens192 PF1-REG

ping -I ens224 PF2-REG
ping -I ens224 PF1-REG

You will find one or two that does not work, fix it and you will have no more 
that error message.

My guess: You have deployed that in VMware and the interfaces that you think 
are Registration or Isolation on the VM does not match the order of the VM 
system.

Check the MAC address with “ip a” command and check if all the MAC address are 
the one assigned to the correct VM network. “arp -a” help as well.

Been there, done that.

Have a nice day,

Ludovic Zammit

[email protected]<mailto:[email protected]> ::  +1.514.447.4918 (x145) ::  
www.inverse.ca<https://www.inverse.ca/>

Inverse inc. :: Leaders behind SOGo (http://www.sogo.nu<http://www.sogo.nu/>) 
and PacketFence (http://packetfence.org<http://packetfence.org/>)







On Mar 31, 2021, at 11:03 AM, Chris Crawford via PacketFence-users 
<[email protected]<mailto:[email protected]>>
 wrote:

Good morning,

I have a new cluster installed and everything appears to be operating correctly 
except for an error on login to the Admin Interface which states:
PacketFence1.fqdn haproxy haproxy-portal - health
Haproxy_haproxy-portal.health_sdown.haproxy_backend_server_status

When I look at the Status>Cluster>Services area, I do see that HAProxy-Portal 
is checked as enabled and managed on all server, but is not running on 
PacketFence2.fqdn and PacketFence3.fqdn.

***
Here is my cluster.conf:
# Copyright (C) Inverse inc.
# Cluster configuration file for active/active
# This file will have it deactivated by default
# To activate the active/active mode, set a management IP in the cluster section
# Before doing any changes to this file, read the documentation
[CLUSTER]
management_ip=10.10.10.30

[CLUSTER interface ens160]
ip=10.10.10.30

[CLUSTER interface ens192]
ip=10.45.1.1

[CLUSTER interface ens224]
ip=10.45.2.1

[packetfence1.fqdn]
management_ip=10.10.10.31

[packetfence1.fqdn interface ens160]
ip=10.10.10.31

[packetfence1.fqdn interface ens192]
ip=10.45.1.2

[packetfence1.fqdn interface ens224]
ip=10.45.2.2

[packetfence2.fqdn]
management_ip=10.10.10.32

[packetfence2.fqdn interface ens160]
ip=10.10.10.32

[packetfence2.fqdn interface ens192]
ip=10.45.1.3

[packetfence2.fqdn interface ens224]
ip=10.45.2.3

[packetfence3.fqdn]
management_ip=10.10.10.33

[packetfence3.fqdn interface ens160]
ip=10.10.10.33

[packetfence3.fqdn interface ens192]
ip=10.45.1.4

[packetfence3.fqdn interface ens224]
ip=10.45.2.4

***
Here is my HAProxy-Portal.conf
# Copyright (C) Inverse inc.
global
 external-check
 user haproxy
       group haproxy
       daemon
       pidfile %%var_dir%%/run/haproxy-portal.pid
       log /dev/log local0
       stats socket %%var_dir%%/run/haproxy-portal.stats level admin process 1
       maxconn 4000
       #Followup of https://github.com/inverse-inc/packetfence/pull/893
       #haproxy 1.6.11 | intermediate profile | OpenSSL 1.0.1e | SRC: 
https://mozilla.github.io/server-side-tls/ssl-config-generator/?server=haproxy-1.6.11&openssl=1.0.1e&hsts=yes&profile=intermediate
       #Oldest compatible clients: Firefox 1, Chrome 1, IE 7, Opera 5, Safari 
1, Windows XP IE8, Android 2.3, Java 7
       tune.ssl.default-dh-param 2048
       ssl-default-bind-ciphers <cert>
       ssl-default-bind-options no-sslv3 no-tls-tickets
       ssl-default-server-ciphers <cert>
       ssl-default-server-options no-sslv3 no-tls-tickets
       #OLD SSL CONFIGURATION. IF RC4 is required or if you must support 
clients older then the precendent list, comment all the block between this 
comment and the precedent and uncomment the following line
       #ssl-default-bind-ciphers <cert>
       lua-load %%conf_dir%%/passthrough.lua

listen stats
 bind  %%management_ip%%:1025
 mode http
 timeout connect 10s
 timeout client 1m
 timeout server 1m
 stats enable
 stats uri /stats
 stats realm HAProxy\ Statistics
 stats auth admin:packetfence


defaults
       log     global
       mode    http
       option  httplog
       option  dontlognull
       timeout connect 5000
       timeout client 50000
       timeout server 50000
       errorfile 403 %%captiveportal_templates_path%%/rate-limiting.http

backend proxy
   option httpclose
   option http_proxy
   option forwardfor
   # Need to have a proxy listening on localhost port 8888
   acl paramsquery query -m found
   http-request set-uri 
http://127.0.0.1:8888%[path]?%[query<http://127.0.0.1:8888%25[path]?%25[query>] 
if paramsquery
   http-request set-uri 
http://127.0.0.1:8888%[path<http://127.0.0.1:8888%25[path>] unless paramsquery

backend static
   option httpclose
   option http_proxy
   option forwardfor
   http-request set-uri 
http://127.0.0.1:8889%[path]?%[query<http://127.0.0.1:8889%25[path]?%25[query>]

%%http%%

***
Here is the error message in the haproxy-portal.log
Mar 31 11:11:21 packetfence1 haproxy[6238]: Proxy portal-http-10.25.1.1 started.
Mar 31 11:11:21 packetfence1 haproxy[6238]: Proxy portal-https-10.25.1.1 
started.
Mar 31 11:11:21 packetfence1 haproxy[6238]: Proxy portal-http-10.25.2.1 started.
Mar 31 11:11:21 packetfence1 haproxy[6238]: Proxy portal-https-10.25.2.1 
started.
Mar 31 11:11:21 packetfence1 haproxy[6238]: Proxy portal-http-66.70.255.147 
started.
Mar 31 11:11:21 packetfence1 haproxy[6238]: Proxy portal-https-66.70.255.147 
started.
Mar 31 11:11:31 packetfence1 haproxy[6242]: backend 10.25.1.1-backend has no 
server available!
Mar 31 11:11:46 packetfence1 haproxy[6242]: backend 10.25.2.1-backend has no 
server available!

I get this error every time I login, and every so often when it does a check. I 
have not tried the captive portal yet from a device yet. But, it does work 
through the admin interface using the Portal_Preview.

I suspect that I'm missing a section from the Load Balancers IP in the 
Configuration>Advanced Access Configuration>Captive Portal section. But, I 
can't find anything in the cluster documentation that would indicate that's how 
it works.

Cheers,
CHRIS



_______________________________________________
PacketFence-users mailing list
[email protected]<mailto:[email protected]>
https://lists.sourceforge.net/lists/listinfo/packetfence-users

_______________________________________________
PacketFence-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/packetfence-users

Reply via email to