Unreachable server after power outage

[PP44]

What type of hardware are you using: Old laptop or computer
What YunoHost version are you running: 12.0.17
How are you able to access your server: Direct access via physical keyboard/screen
Are you in a special context or did you perform specific tweaking on your YunoHost instance ?: I use custom DSN resolver

Describe your issue

Recently, I had a short power outage at home. My server restarted, but some services (mastodon if I remember correctly) stopped working. I checked the web admin, managed to connect, not through the domain, but with the local ip, but after a few minutes on the web admin, it froze, so I tried ssh. Same thing, would not connect to the domain, but worked through ssh. I figured something was wrong with the DNS resolver, and I remembered that I use the FDN resolver that recently ended. So I changed to dnsforge.de’s DNS. I thought I would solve the issue, but when I run the diagnosis, first of all, it stalls, but it still report an issue for internet connectivity about DNS still being broken “for some reason”. Also, nginx won’t start, and I can’t connect through ssh even via the local IP address. When I look at the nginx logs after trying to restart it, I get warnings about ingnoring “ssl_stapling”, then a “start-pre opperation timed out. Terminating”

Any ideas ? I cannot copy paste the logs because I cannot connect through ssh. Thank you very much for your help <3

Share relevant logs or error messages

Cannot copy paste them

[PP44]

Anyone have any idea ? I am very afraid of doing a whole new reinstall with the backups and everything…

[otm33]

Hello @pp44t,

You can generate logs:

• from the web admin: Tools > Services > relevant service(s) > Share with yunopaste

• from the command line using this syntax as an example:

less /var/log/nginx/error.log | yunopaste

How did you change the DNS?
Have you tried running sudo yunohost tools regen-conf dnsmasq nginx ssh or sudo yunohost tools regen-conf?

[PP44]

Sorry for the late answer, I am kind of panicking and avoiding the issue, but now I have no choice but to face it

@otm33

less /var/log/nginx/error.log | yunopaste

this returns no input received from less

@otm33

Have you tried running sudo yunohost tools regen-conf dnsmasq nginx ssh or sudo yunohost tools regen-conf?

this return :

Warning : WARNING Could not indentify correctly the dns zone for domain mydomain.tld, returning mydomain.tld.

This appear twice for my main domain and all subdomains.
I usually access my machine through a VPN, maybe the VPN client is not connected, creating an issue ?

Still 1 significant issue related to internet connectivity…
When I ask for human readable diagnostic results, I get a “the server does not seem to be connected to the internet at all !?”

So I tried to ping my router, or local machines, everything works, but if I ping outside of my local network, I get nothing ! My desktop, connected to the same router, has no issues.

[otm33]

@PP44

if I ping outside of my local network, I get nothing !

… Une éventuelle différence entre ping google.com et ping 8.8.8.8 ?
Question bête mais que je pose quand même: ta box n’aurait-elle pas pu faire sauter quelques lignes dans certaines tables (Nat, dyndns) suite à son redémarrage ? Je parle d’expérience…

[PP44]

@otm33

Une éventuelle différence entre ping google.com et ping 8.8.8.8 ?

Non, j’ai essayer sur les deux machine fdn.fr et 80.67.169.52 et même résultat, ping sur mon desktop ok, ping sur le server rien du tout…

@otm33

Question bête mais que je pose quand même: ta box n’aurait-elle pas pu faire sauter quelques lignes dans certaines tables (Nat, dyndns) suite à son redémarrage ? Je parle d’expérience…

Pas bête puisque effectivement tout a sauté. Je me disais que le UPnP aurait rattrapé le coup, mais dans le doute j’ai tout remis à la main, ça n’a pas l’air d’aidé beaucoup…

Cependant du nouveau, après sudo yunohost tools regen-conf dnsmasq nginx ssh, pas de changement, plus le temps hier, donc je redémarre mon serveur et me barre. Aujourd’hui, de retour dessus, je refait la redirection de port susmentionnée, puis je teste l’interface web, et elle est de retour via l’adresse IP. Le serveur n’a toujours pas l’air d’avoir accès à internet…

Du coup quelques trucs que j’ai remarqué aujourd’hui qui sont bizarres :

• Ping le serveur via la commande ping dans un terminal : pas de réponse, mais Angry IP Scanner trouve le serveur, avec un ping de 2002 ms, de manière très régulière, des variations de quelques millisecondes entre les tests
• Angry ip scanner trouve bien mon port SSH ouvert (j’utilise un autre port que le 22), mais la commande SSH n’obtient pas de réponse…
• When I look into my /etc/resolv.conf file, the only name server I get is 127.0.0.1, is that normal ?

[otm33]

@PP44

I usually access my machine through a VPN, maybe the VPN client is not connected, creating an issue ?

Et que retourne ip route ?

[PP44]

@otm33

Et que retourne ip route ?

default cia 192.168.0.1 dev enp2s0
192.168.0.0/24 dev enp2s0 proto kernel scope link src 192.168.0.15

[otm33]

Déconnecte temporairement le pare-feu de yunohost pour voir si cela change qqch.

Il faudrait en savoir un peu plus sur la configuration de ton réseau (box ? box + routeur? VPN ? Règles de routage et/ou de pare feu, filtrage?)

PS : est-ce que tu as fait un sudo yunohost tools regen-confcomplet ?