MySQL addon not starting

[robi]

@stratus41298 for that you need a web service that will have a 200 response on that /healthcheck endpoint

[stratus41298]

@robi I'm not really sure where you're going with that so I can't comment.

As for the continued troubleshooting, I was able to get postgresql up manually with extensive help of ChatGPT. We ran into a DNS issue trying to do an APT reinstall of docker even after adjusting the resolv.conf. If anyone is interested I can give a full writeup, but i'll summarize here:

---

🧩 Cloudron 8.3.1 Upgrade – PostgreSQL & DNS Issues Summary

1. PostgreSQL Socket Failure

• After updating to Cloudron 8.3.1, the PostgreSQL service showed as down in the Cloudron dashboard.

• However, logs showed the database was running fine and accepting connections over TCP.

• Cloudron healthchecks (and dependent services) failed due to the absence of the expected Unix socket at /var/run/postgresql/.s.PGSQL.5432.

• Investigation showed:

  • The actual socket was created in /run/postgresql/
  • The symlink /var/run/postgresql → /run/postgresql was missing inside the PostgreSQL container

• Fix: Re-created the symlink inside the container:

  ln -s /run/postgresql /var/run/postgresql
  

  Followed by:

  supervisorctl restart postgresql
  

• After this, PostgreSQL healthcheck passed, and dependent services began recovering.

---

2. APT Repository DNS/TLS Failure

• While attempting to reinstall the Cloudron CLI to recover from a broken docker recreate, the APT repo at https://apt.cloudron.io was unreachable.

• DNS (dig apt.cloudron.io) returned SOA records but no A records — even after:

  • Disabling systemd-resolved
  • Stopping unbound
  • Overriding /etc/resolv.conf to use 8.8.8.8
  • Confirming DNS worked for other domains (cloudron.io, google.com)

• Attempts to work around with /etc/hosts resulted in a TLS handshake failure:

  error:0A000410:SSL routines::sslv3 alert handshake failure
  

• Root cause appears to be Cloudflare (which fronts apt.cloudron.io) rejecting connections when:

  • DNS is bypassed via /etc/hosts, or
  • Something in the server environment breaks SNI or TLS negotiation

• Result: Unable to reinstall cloudron CLI using APT despite proper DNS config and working CA certs.

---

🧠 Context

• This all occurred during/after a failed cloudron docker recreate attempt (as suggested on the forums).
• That left the system partially repaired, with missing CLI tools and broken DNS.

---

So that's pretty much it. At this point i'm going to just set up a restore either through VM backup or Cloudron backup. I haven't decided yet. I won't do it until tomorrow though in case there's anything you'd like me to provide for future maintenance or research.

[nebulon]

So whatever you have posted here, makes no sense in the cloudron context, some of those commands don't even exist, nor is there any cloudron apt repository (apt.cloudron.io does and should not exist). Much AI hallucination happening here it seems

Looks like we have to start from first principles, first mysql didn't work then postgres and maybe redis...? Also it is probably not clear by now in which state the server is and what would have to be undone or not. Given all that, if you have recent backups a fresh start is probably the fastest and safest option to get to some well known state.

[stratus41298]

That makes sense that there was no apt repo, but the fact was that we couldn't even get out to apt due to the DNS issue so i'd caution an immediate dismissal of AI hallucination. We were also able to get the healthcheck up and running for postgresql after it had gotten stuck during the docker recreate (see forked post) using AI as well so i'm not sure how many of the commands were inappropriate as they all went through. We did indeed disable resolved, stopped unbound, overrode resolv.conf (which i've done before for pihole many times), digged DNS, bypassed DNS using hosts, and I myself had previously performed a docker recreate. All of those commands exist except I guess the apt reinstall, but that's something I myself might have tried, if i'm being honest as it's a fairly common command on other apps.

After the 8.3.1 update many services wouldn't come up and nothing worked on this instance. I guess if you have all the information you care about on this issue, i'll go ahead and restore since i've been down for quite a while at this point.

I get the whole "AI = bad" thing, but I spent about 2 hours troubleshooting with GPT 4o (the paid version) and made significant progress until we hit a wall that we couldn't get around, which at least gave me satisfaction that we tried everything we could.

Take care.

[nebulon]

I am generally not in the group of AI = bad, however it likely just lacks lots of knowledge about the Cloudron internals, which make recommendations often not very useful in this area (yet).

Overall I am still not sure based on this thread why DNS was involved here and given that you hit 3 different addons with apparently diverse issues, maybe something unrelated to them individually caused the trouble.

We will add a few more helpers to the cloudron-support script for the future, to hopefully get down to such issues faster.

If you haven't deleted the server yet, maybe we could take a closer look via SSH, since you already invested much time in debugging. If you want that, please enable remote ssh support and send a mail to support@cloudron.io with your Cloudron details.