Nginx autostart does not work properly

[asterixus]

Hi,

My Plesk server installation has following issue:

2015/11/11 11:16:27 [emerg] 1675#0: bind() to <IPv4>:80 failed (99: Cannot assign requested address)
2015/11/11 11:47:15 [emerg] 1722#0: bind() to [IPv6]:80 failed (99: Cannot assign requested address)

Nginx works fine, if I start it manually. But when I reboot the server, it does not!
I tried the solution from this thread: http://talk.plesk.com/threads/solve...-37-187-211-80-443-failed.333998/#post-784024
but it did not help!

Hope, someone can give me advice!

Thanks!

 

[jnarvaez]

Same problem here on Centos 7 fresh new server.

 

[UFHH01]

Hi,

such issues may appear, when systemd boots very fast, but the network is not yet ready and fully initialized.

You might be intersted in reading: http://www.freedesktop.org/wiki/Software/systemd/NetworkTarget/ ( external link, please inform me, when the link goes dead, so I could adjust the link with another one! )


Consider to investigate your "/var/log/messages" - log and use systemd - commands ( as for example "systemd-analyze" ), to investigate your issue:

http://www.freedesktop.org/software/systemd/man/systemd-analyze.html ( external link, please inform me, when the link goes dead, so I could adjust the link with another one! )

 

[asterixus]

Well, I was able to solve ipv4 issue with:
net.ipv4.ip_nonlocal_bind=1
The server indeed might start to fast for the network interfaces to become ready, because of the HW raid1+ssd, and it also uses NetworkManager, but in my opinion it is still a bug, that I cannot understand. My other server, same config Centos7+plesk125, but with SOFT RAID1 + Sata HDD works fine with no such modification for ipv4. I am still not able to find a solution for this ipv6 issue.

Nov 12 22:54:49 SRV systemd: Starting Startup script for nginx service...
Nov 12 22:54:49 SRV nginx: nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
Nov 12 22:54:49 SRV nginx: nginx: [emerg] bind() to [ipv6]:80 failed (99: Cannot assign requested address)

So I assume, it is not the boot time, which causes the issue(or is a part of the problem).

 

[UFHH01]

Hi asterixus,

sorry, but without any investigations, you can't "assume" conflicts, failures, issues and problems. You never will get to the root cause, if you "assume" something, just by having a look at the nginx error and by compairing two different servers hardware... but ... it's up to you...