Issue Additional IP addresses configuration randomly breaks

[JulienP]

Hi,

We have a recurring problem with our server where all ip addresses become “improperly configured” after a random interval of time.

It can work for weeks without any problem and suddenly break apart (most of time in the middle of the night…).

The server is running on Debian 9.12 with Plesk 18.

When the error occurs all websites become unavailable. If we look at the Plesk interface, we can see all ip addresses (except the original one that came with the server) on error with the following message :
“This IP address is not configured properly”.

Plesk offers a “Repair” link for each ip which solves the problem.
Restarting the server also fixes the issue.

I didn’t find anything in the system’s logs, but I’ve found these errors in the Plesk’s logs:

stderr:

nginx: [emerg] bind() to XX.XXX.XXX.XXX:443 failed (99: Cannot assign requested address)

nginx: configuration file /etc/nginx/nginx.conf test failed



[2020-09-10 05:52:55.520] ERR [util_exec] proc_close() failed ['/opt/psa/admin/bin/nginx-config' '-t'] with exit code [1]

[2020-09-10 05:52:55.520] ERR [panel] '/opt/psa/admin/bin/nginx-config' '-t' failed with code 1.

There is one error of this kind for each IP address.

I don’t know if it may be related, but just before the first error in the log, there is this error:

[2020-09-10 05:50:07.606] ERR [panel] 2020-09-10 05:50:03 INFO: pum is called with arguments: ['--list', '--repo-info', '--json']

2020-09-10 05:50:07 ERROR: W:GPG error: http://ppa.launchpad.net/ondrej/php/ubuntu focal InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 4F4EA0AAE5267A6C, E:The repository 'http://ppa.launchpad.net/ondrej/php/ubuntu focal InRelease' is not signed., W:Updating from such a repository can't be done securely, and is therefore disabled by default., W:See apt-secure(8) manpage for repository creation and user configuration details., W:GPG error: https://repo.cloudlinux.com/kernelcare-debian/6 stable InRelease: The following signatures were invalid: E58FE9B3FE3B0470525159BD6DC3D600CDEF74BB

2020-09-10 05:50:07 ERROR: Exited with returncode 1.

[2020-09-10 05:52:53.165] ERR [util_exec] proc_close() failed ['/opt/psa/admin/bin/nginx-config' '-t'] with exit code [1]

[2020-09-10 05:52:53.165] ERR [panel] '/opt/psa/admin/bin/nginx-config' '-t' failed with code 1.

It’s related to a custom repository that we added months ago. At first glance it doesn’t seem related but I don’t know, the error is only 2 minutes ahead of the first IP address related error.

Do you have any insight to share?

Thanks for your help.

 

[IgorG]

I'd encourage you to try it out:

Nginx on Plesk server fails to start: 99: Cannot assign requested address

Applicable to: Plesk for Linux Symptoms nginx fails to start with the following error: CONFIG_TEXT: Cannot assign requested address The error below may appear in the Home page in Plesk: PLE...

support.plesk.com

 

[JulienP]

Hi,

Thank for your response but we already do what the article suggests when the problem occurs.
But they are two main drawbacks to this:

1) it requires a manual intervention
2) it takes time to re-sync the ips because we have to do it one by one

It becomes a bit critical for us as it now happens several times a day, making the server unusable.

By looking back at the logs, I've see an error related to SSL certificates. The certificate of one of our sites fails to renew:

[2020-09-16 17:26:07.056] ERR [extension/letsencrypt] Domain validation failed for site.com: Invalid response from https://acme-v02.api.letsencrypt.org/acme/authz-v3/7261872379.
Details:
Type: urn:ietf:params:acme:error:unauthorized
Status: 403
Detail: Invalid response from https://ssl-challenges.infomaniak.com/acme-challenge/-xpf-cZQKwU2EKeqaPPA9zZifZ33PAWCth5zA_ZEYFE [185.127.15.2]: 404
[2020-09-16 17:26:07.064] ERR [extension/letsencrypt] Domain validation failed: Invalid response from https://acme-v02.api.letsencrypt.org/acme/authz-v3/7261872379.
Details:
Type: urn:ietf:params:acme:error:unauthorized
Status: 403
Detail: Invalid response from https://ssl-challenges.infomaniak.com/acme-challenge/-xpf-cZQKwU2EKeqaPPA9zZifZ33PAWCth5zA_ZEYFE [185.127.15.2]: 404
[2020-09-16 17:26:07.090] ERR [extension/sslit] Unable to renew domain {domainName} wildcard certificate automatically Invalid response from https://acme-v02.api.letsencrypt.org/acme/authz-v3/7261872379.
Details:
Type: urn:ietf:params:acme:error:unauthorized
Status: 403
Detail: Invalid response from https://ssl-challenges.infomaniak.com/acme-challenge/-xpf-cZQKwU2EKeqaPPA9zZifZ33PAWCth5zA_ZEYFE [185.127.15.2]: 404

I don't understand how an error with Let's Encrypt can cause the ips to desynchronize but I don't know.
Do you think it can be related?

Thanks.