Anyone else having an issue when an update is installed where the tunnel goes down and has to be manually started?

It happened today but has happened numerous times before, I setup an alert in Cloudflare because of it so I get an email alert if the tunnel goes down now, this is what it showed in the log today :-

2026-10-06T12:27:10Z INF cloudflared has been updated version=2026.10.0

2026-10-06T12:27:10Z INF Restarting service managed by SysV...

2026-10-06T12:27:10Z INF PID of the new process is 42

2026-10-06T12:27:10Z ERR Initiating shutdown error="cloudflared has been updated to version 2026.10.0"

2026-10-06T12:27:10Z INF Metrics server stopped

2026-10-06T12:27:10Z INF Tunnel server stopped

cloudflared has been updated to version 2026.10.0

Hello, could you please update this container? I received an email from Cloudflare warning me to update due to the eMudhra private key issue.

7 minutes ago, bybes said:

Hello, could you please update this container? I received an email from Cloudflare warning me to update due to the eMudhra private key issue.

It should be easily updatable by yourself. The email states:
"If you installed with a package manager, upgrade with that same package manager and then restart your service."

I saw mine was still running on 2026.9.1 and after restarting the service it pulled down 2026.10.4 automatically.

1 hour ago, Gyakusetsu said:

It should be easily updatable by yourself. The email states:
"If you installed with a package manager, upgrade with that same package manager and then restart your service."

I saw mine was still running on 2026.9.1 and after restarting the service it pulled down 2026.10.4 automatically.

Same here - restarting the container triggered an update to 2026.10.4

Edited October 26Oct 26 by GarethRR
double post

Just came to ask the same thing.
Thanks for posting, @Gyakusetsu !

9 hours ago, Gyakusetsu said:

It should be easily updatable by yourself. The email states:
"If you installed with a package manager, upgrade with that same package manager and then restart your service."

I saw mine was still running on 2026.9.1 and after restarting the service it pulled down 2026.10.4 automatically.

I've stopped my container and restarted it a bunch of times and it still isn't working. I've also tried the "force update" function. I originally installed via community apps. How can I verify what version I'm running and get it to update?

Edit: I seem to have had some greater connectivity issue crop up out of the blue? Chatgpt helped me diagnose but it seems like my Adguard Home container and Tailscale were no longer playing nicely for some reason. I enabled "Host access to custom networks" and i was able to connect to CA and update my containers. it also fixed mounting SMB. not sure what happened there.

Edited October 27Oct 27 by Randyd718

Same here after a stop/start. I didn't see a change on unraid but happened to have the Cloudflare Tunnels page open and as soon as the docker started the version changed to 2026.10.4 on that page.

OK I'm back. Mine has seemingly updated and Cloudflare Tunnel page reports "2026.10.5", but the container keeps stopping itself after a while. I don't even know what the issue is, anyone have any suggestions?

On 6/10/2026 at 11:36 AM, gacpac said:

i'm having the same problem. I'm monitoring the docker now with uptimekuma. and nothing happened last night at 3am.

What could i provide from my end because i don't see logs either anywhere, docker just stopped. Ideally yes i could simply have a script to start a docker manually like a watchdog, but that's just a band aid not the real fix.

have you (or anyone) figured the cause of this? mine has just started crashing itself recently also. it was running fine for years until now.

9 hours ago, Randyd718 said:

have you (or anyone) figured the cause of this? mine has just started crashing itself recently also. it was running fine for years until now.

What is your NO_AUTOUPDATE variable set to? If it is set to false and cloudflare releases an update, cloudflared will kill the process which kills the container. I am experimenting with new ways of handling this. For now you can add --restart unless-stopped in the `extra parameters` section if you want to keep it on false. or set it to true and have it restart from your appdata backup process (if u use this) or a cron script that restarts the container every so often.

The updates on cloudflared github releases page line up with when your container is crashing but let me know if you have other information that might be causing this.

Edited Saturday at 01:20 AM5 days by Figro

On 10/30/2026 at 7:17 PM, Figro said:

What is your NO_AUTOUPDATE variable set to? If it is set to false and cloudflare releases an update, cloudflared will kill the process which kills the container. I am experimenting with new ways of handling this. For now you can add --restart unless-stopped in the `extra parameters` section if you want to keep it on false. or set it to true and have it restart from your appdata backup process (if u use this) or a cron script that restarts the container every so often.

The updates on cloudflared github releases page line up with when your container is crashing but let me know if you have other information that might be causing this.

NO_AUTOUPDATE is set to "true"

I will say it seems to have been running the past day or two without interruption...still keeping an eye on it.

I am however also running "Cloudflare-DDNS" from CA and that is still stopping itself. The log references some errors but I'm totally lost on interpreting them, and the support link for that one just goes to a github issues page that doesnt seem to be active. I might need something else to keep my dynamic IP current with CF if anyone has suggestions