Restart works, reload fails

[heck0045]

Hi - I’m running HAProxy version 2.4.3-4dd5a5a, released 2021/08/17 and am able to run:

systemctl restart haproxy.service

without issue, but when I run:

systemctl reload haproxy.service

The service always fails with the below systemctl status haproxy.service:

haproxy.service - SYSV: HAProxy is a TCP/HTTP reverse proxy which is particularly suited for high availability environments.
   Loaded: loaded (/etc/rc.d/init.d/haproxy; bad; vendor preset: disabled)
   Active: failed (Result: signal) since Tue 2021-08-31 18:16:17 UTC; 2s ago
     Docs: man:systemd-sysv-generator(8)
  Process: 9849 ExecStop=/etc/rc.d/init.d/haproxy stop (code=exited, status=0/SUCCESS)
  Process: 9994 ExecReload=/etc/rc.d/init.d/haproxy reload (code=exited, status=0/SUCCESS)
  Process: 9908 ExecStart=/etc/rc.d/init.d/haproxy start (code=exited, status=0/SUCCESS)
 Main PID: 10003 (code=killed, signal=KILL)

Aug 31 18:16:09 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: Starting SYSV: HAProxy is a TCP/HTTP reverse proxy which is particularly suited for high availability environments....
Aug 31 18:16:09 ip-10-0-6-173.us-west-2.compute.internal haproxy[9908]: Starting haproxy: [  OK  ]
Aug 31 18:16:09 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: Started SYSV: HAProxy is a TCP/HTTP reverse proxy which is particularly suited for high availability environments..
Aug 31 18:16:17 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: Reloading SYSV: HAProxy is a TCP/HTTP reverse proxy which is particularly suited for high availability environments..
Aug 31 18:16:17 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: Reloaded SYSV: HAProxy is a TCP/HTTP reverse proxy which is particularly suited for high availability environments..
Aug 31 18:16:17 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: haproxy.service: main process exited, code=killed, status=9/KILL
Aug 31 18:16:17 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: Unit haproxy.service entered failed state.
Aug 31 18:16:17 ip-10-0-6-173.us-west-2.compute.internal systemd[1]: haproxy.service failed.

I’d love to get any guidance. Thanks!

[lukastribus]

You need to remove the init.d script and install the systemd unit file from contrib/systemd:

Failing to start with systemd and unexplained issues

Do not use the init.d file. You are using systemd, you have use the systemd unit file and wrapper, not the init.d script. rm /etc/init.d/haproxy cd contrib/systemd cp haproxy.service.in /lib/systemd/system/ systemctl daemon-reload systemctl enable haproxy systemctl start haproxy systemctl status haproxy Note: Haproxy 1.8 and later needs to be compiled with USE_SYSTEMD=1 for the provided systemd unit file to work (needs systemd header files, that would be libsystemd-dev on Debian/Ubuntu or sys…

[heck0045]

Thank you - well that set me in the right direction. Here is what I did:

1. rm /etc/init.d/haproxy
2. Recompiled haproxy from source using make USE_SYSTEMD=1
3. cd /root/haproxy-2.4.3/admin/systemd (my haproxy source directly is /root/haproxy-2.4.3) – note that contrib seems to be the old dir name, now the systemd file lives in admin)
4. Run make – this copies the haproxy.service.in file into haproxy.service and replaces @SBINDIR@ with your local dir (/usr/local)
5. cp haproxy.service /lib/systemd/system
6. systemctl daemon-reload
7. systemctl start haproxy.service
8. systemctl reload haproxy.service ← and it works!

Thanks