Nginx-docker-container dies after a few seconds


#1

Hello,

I have installed bitwarden with local docker containers on a virtual machine running debian 9 with proxmox/kvm as virtualisation solution. I have run bitwarden.sh install and started it with “bitwarden start”. The nginx container appears for a few seconds and shows port 443 and 80 opening, but then it stops and is caught in a loop.
bitwarden/nginx:1.30.4 “/entrypoint.sh” 5 minutes ago Restarting (1) 10 seconds ago bitwarden-nginx
Port 80 and 443 are now closed. I have provided working certificates and set a dns entry for bitwarden. As long as the container is restarting I can’t spawn a shell and continue to debug inside the container.
There is one thing which may cause problems. I am using an external domain name, mainly because getting certificates for .local domains, which are trusted in all the browsers, smartphones etc, proved way to much work. and a local IP, which is accessible only from inside our network.
This is fine with me, I only want local access from our network. If someone needs his passwords on the road he/she will have to use VPN. Might this cause problems with nginx?

So how can I get the nginx container to run? And another question: I have no experience with mssql (and wonder a bit why mariadb was not used instead). Is there a way to dump the contents of a mssql-database for backup purposes like it can be done with automysqlbackup and mariadb?

Yours Stefan


#2

Good Evening,
I had a similar issue when the letsEncrypt SSL cert was due for renewal. It looks like nginx is going to try to renew the cert and if you closed ports 80 and 443 from the internet letsEncrypt will fail validating your external name causing the nginx to crash…
Solution reopen ports 80 AND 443 from the internet to your local IP (at least while you restart/update your birwarden to renew the cert).

I also had a similar issue when my docker was apparently too small for bitwarden… after increasing the docker RAM and restarting it was stable again.