nGinX gets shut down and then won't restart.

InoT

One of my servers is having some weird issue lately. nGinX service just stops and then won't restart. Sometimes I have to restart the server to make nGinX work again. Other times clicking on the start button instead of restart also helps.

2020/08/18 06:57:12 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 06:57:12 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 06:57:12 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 06:57:12 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 06:57:12 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 06:57:12 [emerg] 816#0: still could not bind()
2020/08/18 17:37:52 [emerg] 820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:37:52 [emerg] 820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:37:52 [emerg] 820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:37:52 [emerg] 820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:37:52 [emerg] 820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:37:52 [emerg] 820#0: still could not bind()
2020/08/18 17:38:12 [emerg] 819#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:38:12 [emerg] 819#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:38:12 [emerg] 819#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:38:12 [emerg] 819#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:38:12 [emerg] 819#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/18 17:38:12 [emerg] 819#0: still could not bind()
2020/08/21 14:19:26 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:26 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:26 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:26 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:26 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:26 [emerg] 821#0: still could not bind()
2020/08/21 14:19:48 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:48 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:48 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:48 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:48 [emerg] 816#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/21 14:19:48 [emerg] 816#0: still could not bind()
2020/08/22 08:08:23 [emerg] 17749#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:23 [emerg] 17749#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:23 [emerg] 17749#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:23 [emerg] 17749#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:23 [emerg] 17749#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:23 [emerg] 17749#0: still could not bind()
2020/08/22 08:08:34 [emerg] 17820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:34 [emerg] 17820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:34 [emerg] 17820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:34 [emerg] 17820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:34 [emerg] 17820#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:08:34 [emerg] 17820#0: still could not bind()
2020/08/22 08:09:08 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:08 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:08 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:08 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:08 [emerg] 821#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:08 [emerg] 821#0: still could not bind()
2020/08/22 08:09:35 [emerg] 1411#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:35 [emerg] 1411#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:35 [emerg] 1411#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:35 [emerg] 1411#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:35 [emerg] 1411#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:35 [emerg] 1411#0: still could not bind()
2020/08/22 08:09:45 [emerg] 1472#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:45 [emerg] 1472#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:45 [emerg] 1472#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:45 [emerg] 1472#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:45 [emerg] 1472#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:09:45 [emerg] 1472#0: still could not bind()
2020/08/22 08:10:48 [emerg] 2252#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:10:48 [emerg] 2252#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:10:48 [emerg] 2252#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:10:48 [emerg] 2252#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:10:48 [emerg] 2252#0: bind() to 0.0.0.0:80 failed (98: Address already in use)
2020/08/22 08:10:48 [emerg] 2252#0: still could not bind()

This is the nGinX error log.
Is there some way to auto restart nGinX if it get's shutdown by any chance? Or is there anyway to find out why it got shut down in the first place so that I can try and fix that issue.

Trying to restart using the restart button shows this error (Same as the log above):

@aaPanel_Jose please look into this. This server holds a WordPress website with WooCommerce only. No other website.

Server configuration:
CentOs 7.8.2003
2 GB RAM, 1 V-CPU, 60 GB SSD.

idflorin

you can try https://stackoverflow.com/questions/14972792/nginx-nginx-emerg-bind-to-80-failed-98-address-already-in-use?answertab=votes#tab-top

InoT

idflorin, Turbine

Thank you for giving solutions for how to start is back again. But, before that I would like to know the reason behind the issue and fix it so that it doesn't repeats again. Else until I manually restart it, the website will be down.

Turbine

sudo killall nginx

InoT

For the time being I have added a cron job with this script code:

`#!/bin/bash

systemctl -q is-active "nginx.service"
status=$?
if [ "$status" == 0 ]; then
echo "OK"
else
systemctl start "nginx.service"
fi`

This will solve my issue of downtime to some extent. But still @aaPanel_Jose it will be really helpful if you can help me to get to the root of the issue.

aaPanel_Jose

InoT
Thanks for your feedback, we will check this script

InoT

aaPanel_Jose That;s good but what about helping me find the root cause of the issue first? Why nGinX is getting stopped? Is it something related to the server itself or if aaPanel is having some issue?

JAY

InoT
why dont you try to change to apache or OLS and see either its happening with nginx or with other webservers too

aaPanel_Jose

InoT
Can you provide me with the error log of nginx?

InoT

JAY Its a live server. I don't have the liberty to try and test theories without affecting the users experience. Otherwise I would have tried your suggestion.

If you can suggest something where I can keep running the website as is and also test apache or ols I am all ears.

InoT

aaPanel_Jose I provided everything on the 1st post.

hklcf

run netstat -anp |grep 80 when nginx start error

InoT

hklcf Thanks for the suggestion. But right now I am more concerned on why is happened and how to stop it from happening again. The fixing part has been taken care of hopefully.
Yet I will keep your's suggestion handy in case my fix fails to work next time the issue occurs.

hklcf

InoT the reason is bind() to 0.0.0.0:80 failed (98: Address already in use), it means port 80 already in use.

hklcf

maybe you have apache installed at the same server?

InoT

hklcf I have nGinX + Php7.4/7.3 + mySQL + phpMyAdmin. Apache is not installed and was never installed.

BTW, I am not complaining that aaPanel itself is at fault. I am just trying to figure out what is causing the issue. Whether its the aaPanel, the OS itself or something else altogether.

aaPanel_Jose

InoT
This may be caused by a process that has not been released after nginx is shut down, causing port 80 to be occupied all the time, or other reasons may cause nginx to suspended animation. The log you provided only shows that port 80 is occupied.

InoT

aaPanel_Jose I am afraid I have that log only.