Nginx dont start, after reboot - IPv6

[B4c4rd1]

TITLE:

Nginx dont start, after reboot - IPv6​

PRODUCT, VERSION, OPERATING SYSTEM, ARCHITECTURE:

Plesk Onyx, Version 17.0.17 Update #20, Ubuntu 16.04.2 LTS‬, 64 Bit VM​

PROBLEM DESCRIPTION:

I have currently the problem that nginx after a restart does not start automatically. At least he starts too early.​

STEPS TO REPRODUCE:

Add 20 IPv6 Addresses and reboot!​

ACTUAL RESULT:

2017/03/30 20:16:40 [emerg] 1651#0: bind() to [XXXXIPV6::10]:80 failed (99: Cannot assign requested address)

The nginx service dont start after a reboot.

With service nginx start, the service starts.​

EXPECTED RESULT:

Nginx startet after reboot​

ANY ADDITIONAL INFORMATION:

I fixed it by appending net.ipv6.conf.eth0.accept_dad = 0 to /etc/sysctl.conf​

YOUR EXPECTATIONS FROM PLESK SERVICE TEAM:

Confirm bug​

 

[UFHH01]

Hi B4c4rd1,

I rather think, that your service "systemd-networkd" isn't running on your server, or you still use "network-manager".

Consider to start the services with for example:

systemctl start systemd-networkd.service
systemctl start systemd-resolved.service

... and enable these services with

systemctl enable systemd-networkd.service
systemctl enable systemd-resolved.service

... so that they will start on each reboot / system start.​

More informations for "systemd" can be found at for example: => NetworkTarget

 

[B4c4rd1]

Hi @UFHH01

Thanks for your response. I already tried this option. But this does not work in a VM when the IPv6 is forwarded.

root@srv-02 ~ # service nginx status
● nginx.service - Startup script for nginx service
   Loaded: loaded (/lib/systemd/system/nginx.service; enabled; vendor preset: enabled)
   Active: failed (Result: exit-code) since Fr 2017-03-31 00:54:03 CEST; 13s ago
  Process: 1604 ExecStartPre=/usr/sbin/nginx -t (code=exited, status=1/FAILURE)
  Process: 1591 ExecStartPre=/usr/bin/test $NGINX_ENABLED = yes (code=exited, status=0/SUCCESS)

Mär 31 00:54:03 srv-02 systemd[1]: Starting Startup script for nginx service...
Mär 31 00:54:03 srv-02 nginx[1604]: nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
Mär 31 00:54:03 srv-02 nginx[1604]: nginx: [emerg] bind() to [IPv6::10]:80 failed (99: Cannot assign requested address)
Mär 31 00:54:03 srv-02 nginx[1604]: nginx: configuration file /etc/nginx/nginx.conf test failed
Mär 31 00:54:03 srv-02 systemd[1]: nginx.service: Control process exited, code=exited status=1
Mär 31 00:54:03 srv-02 systemd[1]: Failed to start Startup script for nginx service.
Mär 31 00:54:03 srv-02 systemd[1]: nginx.service: Unit entered failed state.
Mär 31 00:54:03 srv-02 systemd[1]: nginx.service: Failed with result 'exit-code'.

root@srv-02 ~ # systemctl status systemd-networkd.service
● systemd-networkd.service - Network Service
   Loaded: loaded (/lib/systemd/system/systemd-networkd.service; enabled; vendor preset: enabled)
   Active: active (running) since Fr 2017-03-31 00:54:01 CEST; 47s ago
     Docs: man:systemd-networkd.service(8)
 Main PID: 520 (systemd-network)
   Status: "Processing requests..."
    Tasks: 1
   Memory: 1.1M
      CPU: 5ms
   CGroup: /system.slice/systemd-networkd.service
           └─520 /lib/systemd/systemd-networkd

root@srv-02 ~ # systemctl status systemd-resolved.service
● systemd-resolved.service - Network Name Resolution
   Loaded: loaded (/lib/systemd/system/systemd-resolved.service; enabled; vendor preset: enabled)
   Active: active (running) since Fr 2017-03-31 00:54:03 CEST; 55s ago
     Docs: man:systemd-resolved.service(8)

 

[UFHH01]

Hi B4c4rd1,

your described issue is not a bug of Plesk products or components. Neither "ifupdown", nor "inetutils-tools", or "systemd-networkd" are shipped with Plesk.
Your described nginx - issue depends on a slow, or delayed network-service-start, which you should forward to Ubuntu ( Suggestion: => Bugs : ifupdown package : Ubuntu )

 

[B4c4rd1]

Yes, you are absolutely right It is rather a problem of Hetzner and their ipv6 settings.

In any case, this error triggers a series of other errors. For example, some folders can not be created in /var/run/ and there is a shm error in the panel.

 

[LTUser]

Confirmed. So i disabled IPV6 on my Hetzner-Server
echo 1 > /proc/sys/net/ipv6/conf/all/disable_ipv6

Now all is fine