16.1 LAB HAProxy not starting

ssubramanian1

I am following Lab 16.1 to install HAProxy for load balancing but HAProxy is not starting getting below error.
`
sudo systemctl status haproxy.service
● haproxy.service - HAProxy Load Balancer
Loaded: loaded (/lib/systemd/system/haproxy.service; enabled; vendor preset: enabled)
Active: failed (Result: exit-code) since Tue 2019-10-08 05:30:34 UTC; 10s ago
Docs: man:haproxy(1)
file:/usr/share/doc/haproxy/configuration.txt.gz
Process: 3398 ExecStart=/usr/sbin/haproxy -Ws -f $CONFIG -p $PIDFILE $EXTRAOPTS (code=exited, status=1/FAILURE)
Process: 3389 ExecStartPre=/usr/sbin/haproxy -f $CONFIG -c -q $EXTRAOPTS (code=exited, status=0/SUCCESS)
Main PID: 3398 (code=exited, status=1/FAILURE)

Oct 08 05:30:34 ip-47-228-8-69 systemd[1]: haproxy.service: Service hold-off time over, scheduling restart.
Oct 08 05:30:34 ip-47-228-8-69 systemd[1]: haproxy.service: Scheduled restart job, restart counter is at 5.
Oct 08 05:30:34 ip-47-228-8-69 systemd[1]: Stopped HAProxy Load Balancer.
Oct 08 05:30:34 ip-47-228-8-69 systemd[1]: haproxy.service: Start request repeated too quickly.
Oct 08 05:30:34 ip-47-228-8-69 systemd[1]: haproxy.service: Failed with result 'exit-code'.
Oct 08 05:30:34 ip-47-228-8-69 systemd[1]: Failed to start HAProxy Load Balancer.
`

chrispokorni

Hi @ssubramanian1,
I experienced similar errors when I started HAproxy based on the configuration presented in the PDF lab exercise alone. When I used the haproxy.cfg file available in the Solutions package I was able to successfully start it.
Did you use the default haproxy.cfg config file?

Regards,
-Chris

ssubramanian1

I tried with solution cfg file now HAProxy does not start throws below error

`ubuntu@ip-:~$ sudo systemctl restart haproxy.service
Job for haproxy.service failed because the control process exited with error code.
See "systemctl status haproxy.service" and "journalctl -xe" for details.
ubuntu@ip-:~$ sudo systemctl status haproxy.service
● haproxy.service - HAProxy Load Balancer
Loaded: loaded (/lib/systemd/system/haproxy.service; enabled; vendor preset: enabled)
Active: failed (Result: exit-code) since Fri 2019-10-11 04:10:32 UTC; 33s ago
Docs: man:haproxy(1)
file:/usr/share/doc/haproxy/configuration.txt.gz
Process: 23785 ExecStart=/usr/sbin/haproxy -Ws -f $CONFIG -p $PIDFILE $EXTRAOPTS (code=exited, status=1/FAILURE)
Process: 23776 ExecStartPre=/usr/sbin/haproxy -f $CONFIG -c -q $EXTRAOPTS (code=exited, status=0/SUCCESS)
Main PID: 23785 (code=exited, status=1/FAILURE)

Oct 11 04:10:31 ip-xx-xxx-8-xx systemd[1]: haproxy.service: Main process exited, code=exited, status=1/FAILURE
Oct 11 04:10:31 ip-xx-xxx-8-xx systemd[1]: haproxy.service: Failed with result 'exit-code'.
Oct 11 04:10:31 ip-xx-xxx-8-xx systemd[1]: Failed to start HAProxy Load Balancer.
Oct 11 04:10:32 ip-xx-xxx-8-xx systemd[1]: haproxy.service: Service hold-off time over, scheduling restart.
Oct 11 04:10:32 ip-xx-xxx-8-xx systemd[1]: haproxy.service: Scheduled restart job, restart counter is at 5.
Oct 11 04:10:32 ip-xx-xxx-8-xx systemd[1]: Stopped HAProxy Load Balancer.
Oct 11 04:10:32 ip-xx-xxx-8-xx systemd[1]: haproxy.service: Start request repeated too quickly.
Oct 11 04:10:32 ip-xx-xxx-8-xx systemd[1]: haproxy.service: Failed with result 'exit-code'.
Oct 11 04:10:32 ip-xx-xxx-8-xx systemd[1]: Failed to start HAProxy Load Balancer.`

chrispokorni

How did you provision your nodes? Are they local or cloud?
Are all the port numbers in the haproxy.cfg file available for HAproxy to use and accessible (not blocked by a firewall) on your proxy node?
In the haproxy.cfg, did you edit the master server name and master IP with your own? The file shows defaults:

backend k8sServers
   balance roundrobin
   server master1  10.128.0.24:6443 check
   ...

Yours should be different (name and IP here are just an example):

backend k8sServers
   balance roundrobin
   server my-k8s-master-1  10.138.4.15:6443 check
   ...

Regards,
-Chris

jzimnowo

Hi @chrispokorni
I also have an issue while starting haproxy on my master node.
After inspecting journalct, I learnt that haproxy is not able to bind stocket:
Starting frontend proxynode: cannot bind socket [0.0.0.0:6443]

It makes sense, because kube-apiserver Pod is bound to hostIP 10.164.0.7 and port 6443.

I am wondering if haproxy should bind to a specific IP (e.g. Virtual IP) ?

Any clue?

chrispokorni

Hi @jzimnowo,

The HAproxy is not supposed to run on any of the master nodes. It is running on a separate/dedicated host.
Read the exercises carefully as they describe the additional resources you need.

Regards,
-Chris

jzimnowo

Hi @chrispokorni
Aha, it works now! Thanks

kumardineshwar

Hi @jzimnowo

have you configure it on other infra node or master nodes.

Regards,
DK

sumersingh

Hi,

I am trying to install and configure the HAPROXY on separate ec2 instance on AWS. I opened the port on security group, disable the firewall, enable the port forwarding and exactly followed the same steps as in pdf but still getting the error.

Dec 12 11:35:27 ip-172-25-4-23 haproxy[26931]: Proxy k8servers started.
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26946]: [ALERT] 345/113528 (26946) : Starting frontend haproxynode: cannot bind socket [0.0.0.0:80]
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26946]: Proxy k8servers started.
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26946]: Proxy k8servers started.
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26970]: [ALERT] 345/113528 (26970) : Starting frontend haproxynode: cannot bind socket [0.0.0.0:80]
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26970]: Proxy k8servers started.
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26970]: Proxy k8servers started.
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26988]: [ALERT] 345/113528 (26988) : Starting frontend haproxynode: cannot bind socket [0.0.0.0:80]
Dec 12 11:35:28 ip-172-25-4-23 haproxy[26988]: Proxy k8servers started.

Let me know if any one have the solution ?

chrispokorni

Apache may be running on port 80 already. Stop it and/or remove it and try the haproxy again.

Regards,
-Chris