Warp CLI: DNS at 127.0.2.2 stops responding after a few minutes on Linux

Zero Trust WARP
1

I used warp-cli on Ubuntu last year for about six months with zero problems. I’ve recently started using it again and I find that the DNS connection to 127.0.2.2 stops responding after a few minutes. I can still see IP addresses—it’s just the DNS resolution that stops working.

Disconnecting and reconnecting is the only way to get it working again.

Any idea what is causing this? I’m currently on Ubuntu 22.04. I have the latest warp-cli and have uninstalled and reinstalled it completely.

Another colleague on Linux has the same problem, but those running a Mac do not.

2

I have noticed that if I egregiously hack /etc/resolv.conf to this, it works for a while longer:

# This file was generated by cloudflare-warp.
nameserver 127.0.0.53
nameserver 127.0.2.2
nameserver 127.0.2.3
nameserver fd01:db8:1111::2
nameserver fd01:db8:1111::3

But eventually it will get automatically rewritten back to Cloudflare’s setting.

3

Still seems to be an issue on warp-cli 2023.3.398

1 Like
4

I noticed that when I am not on any VPN I occasionally get temporary ERR_NETWORK_CHANGED errors in Chrome. These last for a second or two before the page reloads just fine. I have seen others correlate this issue with ip6, so I turned off ip6 entirely. I have not had any VPN problems since then:

sudo sysctl -w net.ipv6.conf.all.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.default.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.lo.disable_ipv6=1

Maybe a coincidence, maybe not.

5

I can confirm the same issue, with the latest release of Fedora SilverBlue, for me though, It just kept trying for connection after failure every 10/20 seconds and fails with unable to connect. Reason: Connectivity check failed due to DNS Lookup failed. For the time being, I’m reverted to 2023.1.133 as both version after that had problems.

1 Like