------- Original Message ------- On Saturday, August 20th, 2022 at 9:50 AM, Willy Tarreau <w...@1wt.eu> wrote:
> Did you notice if it failed to serve anything and ate CPU from start or > if it completely started and only then ate CPU ? It appears to start up normally according to the logs and then it uses 100% CPU. I can get to the stats pages and it serves traffic as expected. I noticed that "systemctl status haproxy" always shows: Active: activating (start) Then after a couple of minutes it restarts the process. It does this continuously. Here is the systemd config. This is directly from the source distro with some modifications: [Unit] Description=HAProxy Load Balancer After=network-online.target rsyslog.service Wants=network-online.target rsyslog.service [Service] EnvironmentFile=-/etc/default/haproxy Environment="CONFIG=/etc/haproxy/haproxy.cfg" "PIDFILE=/run/haproxy.pid" ExecStartPre=/usr/local/sbin/haproxy -Ws -f $CONFIG -c -q $EXTRAOPTS ExecStart=/usr/local/sbin/haproxy -Ws -f $CONFIG -p $PIDFILE $EXTRAOPTS # Checks the config first. ExecReload=/usr/local/sbin/haproxy -Ws -f $CONFIG -c -q $EXTRAOPTS # This saves off the server state before reload which haproxy loads on # (re)start. ExecReload=/opt/bin/save_server_state.sh ExecReload=/bin/kill -USR2 $MAINPID # This should save server state on stop. ExecStop=/opt/bin/save_server_state.sh ExecStopPost=/usr/local/bin/systemd-email-notifier haproxy KillMode=mixed Restart=always SuccessExitStatus=143 Type=notify [Install] WantedBy=multi-user.target We have a staging server that I always deploy to first so I can test anything you want me to there w/o affecting production. Bren
