Error 1000 and tunnel

I created a tunnel last week to allow access to one of our internal servers for external developers to work against.
I set up the tunnel on the server using cloudflared, configured it all, allowed the cloudflared to create the CNAME entry for the host etc… and all worked brilliantly for a week.

Last night I started getting reports of “Error 1000” being displayed which makes no sense given that there is no A record for the host.
If I take the tunnel down I get an Error 1030? so it is recognising the tunnel but as soon as I bring the tunnel back up again I get the Error 1000 again.
I’ve tried setting it up with different hostnames, deleted the coudflared software and reinstalled, no matter what I do it seems I get the Error 1000 every time.

Nothing changed on the setup prior to it failing.

Support has just sent me a template email suggesting I remove the A record… which doesn’t exist and suggested I ask the community and stopped responding so I’m hoping someone here can help.

The ONLY thing that’s different on this setup to a similar setup I’ve used on another account is that this account has two users. Not sure that’s relevant but in the teams dashboard I’m not seeing the tunnel listed, even when I follow the step by step guide.

This morning I recreated everything again using the super admin account and the tunnel now appears in the list as you’d expect… unfortunately the original problem still persists.

This is what I see when I close the tunnel down…

When I bring the tunnel back up I see…

In the tunnels list I now see…

Hi @david.white,

Do you have any Cloudflare Load Balancers in your setup?

No, just a tunnel, cname record and a self-hosted server.

That’s odd, so it’s literally just a DNS record pointing directly to the Tunnel with nothing in between?

I suspect this is going to have to be escalated, do you have a ticket with Cloudflare Support about this?

Yes, couldn’t really be any simpler. The fact that it worked fine for around a week and then “fell over” of it’s own accord makes it even stranger.

I did have a ticket and think it’s still open, I’ve updated it a number of times but they’ve ignored it since the “You need to delete the A record” initial response. It’s really frustrating as I’ve been telling everyone how great tunnels are as a way to allow access to internal servers without putting holes in the firewall and then this happens !

I’ve just been through the tunnel creation process again, using a different domain on a different account and it works perfectly. Fortunately as it’s a dev site we’re not too fussed about the domain we use and, for now at least, it’s working and the devs can get back to work after 36 hours down time.

I’m guessing it’s hitting some issue with the account and bouncing, interestingly the address given as “your IP” in the Error 1000 message isn’t my address… I suspect it’s an internal Cloudflare address as it’s ipv6 and I’m stuck on ipv4 from this location. I think Error 1000 is where all the untracked errors end up !

The domain that works is on a free Cloudflare plan, the one that’s broken is on a paid plan, not sure if that makes a difference.

That’s very odd if you have an identical setup that’s working, usually it’s a Load Balancer or something on the host passing connections back through Cloudflare for a second time.

Can you share your ticket number here?

It was [2371515]

I can see how that sort of loop would cause issues, one of the reasons for trying it on a different domain/account was to at least rule in/out the server configuration.

Thanks, I’ve escalated that ticket.

Thanks, I can set it all back up again in it’s faulty state if needed.

It will probably be needed to troubleshoot it, perhaps you can reproduce it on a subdomain to leave the main one working?

Hello David,

I’ve replied to your case, and I’ll keep an eye out on your case, and if you reply soon with the information, I can personally take a look at your follow up.

Thank you for working on this issue with us.

Gabriel from Support


I’ve updated the ticket with the info and established/attached the tunnel giving the error.


Added the log file to the ticket.

In case anyone else runs into this, it took a while but in the end Daan from Cloudflare support identified a potential issue in the logs that turned out to be the cause. All is now working as it should !

The domain I was having the problem with has been with Cloudflare for a long time and had “Railgun” enabled in the Speed settings. As soon as I turned this off the pipe started working without any further issues. I’m not sure what Railgun actually did/does but turning it off doesn’t seem to have had any negative effect on the sites hosted under that domain.

Whatever Railgun does it seems it’s not compatible with tunnels.

1 Like

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.

Right, Railgun is not compatible with Cloudflare Tunnel , and expect that using both will cause problems.