Linode reboot prevents access to Dokploy panel

My Linode instance needed to reboot, and after successfully rebooting, I can no longer access the Dokploy dashboard. All my applications and databases are still working, but when I try to access the subdomain I have associated with my dashboard, I receive a Bad Gateway error from Cloud flare. I cannot seem to figure out why this is happening. I tried to get service logs for Dokploy, but the command returns nothing (no logs) how can I fix this without having to start all over?
No description
17 Replies
MantisInABox
MantisInABoxOP5mo ago
No description
Sang Dang
Sang Dang5mo ago
this is quite big issue, could you give my friend a help @Siumauricio (sorry for pinging directly)
Henrik
Henrik5mo ago
What does docker ps give you?
Shree
Shree5mo ago
@MantisInABox Looks like the docker installed on your linode instance is not starting on startup. Since dokploy instance runs on as a docker container, means the server will be down and cloudflare will return Bad Gateway, cause they don't know where to redirect the traffic to. How to enable : https://docs.docker.com/engine/install/linux-postinstall/#configure-docker-to-start-on-boot-with-systemd
Shree
Shree5mo ago
No description
MantisInABox
MantisInABoxOP5mo ago
It only lists the applications that I installed through Dokploy, as those are still accessible after the reboot.
Henrik
Henrik5mo ago
No dokploy/traefik/postgres?
MantisInABox
MantisInABoxOP5mo ago
Everything is running after the reboot, except the Dokploy control panel
Shree
Shree5mo ago
even the docker engine?
MantisInABox
MantisInABoxOP5mo ago
Traefik/Postgres are there. I can get you a screenshot in just a moment
Henrik
Henrik5mo ago
I have had a similar experience, but it was traefik that was killed. When I checked journalctl I saw OOM had killed the container. Are you running very resource constrained?
MantisInABox
MantisInABoxOP5mo ago
No description
Henrik
Henrik5mo ago
If dokploy image still exists on the system you can start it up again. I’m not sure what parameter it’ll need, but those can be found by a «donor» VPS
MantisInABox
MantisInABoxOP5mo ago
I tried restarting the image and it started giving connection errors, so, I'm sure I missed something when it comes to the parameters
MantisInABox
MantisInABoxOP5mo ago
I guess the cron cleanups kicked in, and cleaned up the image, because it can't even be found now
No description
MantisInABox
MantisInABoxOP5mo ago
I guess I'll just spin up a new machine and migrate everything over to it, it's all deployed from GitHub so it shouldn't be too bad
Henrik
Henrik5mo ago
That also works, but if you, in the future, need how the conatiner was started, you can run
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock:ro \
assaflavie/runlike <RUNNING-CONTAINER>
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock:ro \
assaflavie/runlike <RUNNING-CONTAINER>
From a donor VPS

Did you find this page helpful?