[HELP (Not Launching/ Crashing)]

[Francisco-Escobedo]

Checklist

• I have given details of my install including Distribution, Wayland/ XOrg, Parameters Used, echo $XAUTHORITY, etc.
• I have provided logs showing any errors, if available (use DEBUG=1 zwift)
• I have filled out the issue template to the best of my ability.
• This issue only contains 1 issue (if you have multiple issues, open one issue for each issue).
• This issue is not a duplicate issue of previous issue.

Describe the issue

Login fails with “Failed to log in. Check internet connection.” (Hyprland / Wayland / rootful Podman)

Environment

Distro: EndeavourOS (Arch-based)

Desktop: Hyprland (Wayland)

XDG_SESSION_TYPE=wayland

NVIDIA GPU (using --gpus=all)

Container tool: rootful sudo-podman

Networking mode: host

Pi-hole temporarily disabled during testing (no change)

System clock synchronized (timedatectl confirms NTP active)

Zwift version shown in client:
1.107.0 (158475)

Latest zwift.sh confirmed via checksum.

Issue Description

Zwift launches successfully inside the container, but login consistently fails with:

“Failed to log in. Check internet connection.”

Manual login on https://zwift.com/ works using the same credentials.

The failure occurs whether:

Credentials are provided via environment variables

Credentials are entered manually in the UI

The error sometimes briefly shows:

“Wrong email/password”

then switches to

“Check internet connection”

Distribution Details

EndeavourOS running Wayland (Hyprland) -- no issues until most recent update

Reproduction steps

Reproduction Steps

Fresh install:

sudo bash -c "$(curl -fsSL https://raw.githubusercontent.com/netbrain/zwift/master/bin/install.sh)"

Launch:
xhost +si:localuser:root
zwift

Enter valid credentials - login fails with “Check internet connection”.

What Has Been Tested
Network Connectivity Inside Container

Verified from container:

sudo podman run --rm --network=host docker.io/curlimages/curl:latest -I https://secure.zwift.com

Result: HTTP/2 404 (TLS + connectivity working)

sudo podman run --rm --network=host docker.io/curlimages/curl:latest -I https://launcher.zwift.com

Result: successful TLS handshake

DNS resolution inside container confirmed working.

[netbrain]

could you try running interactive mode with zwift and using zwift's own container for testing out curl? i.e

INTERACTIVE=1 zwift
then
curl -I https://secure.zwift.com/

[Francisco-Escobedo]

I updated ~/.config/zwift/config to include:

• DONT_CHECK=1
• CONTAINER_TOOL=podman
• NETWORKING=host
• CONTAINER_EXTRA_ARGS=--gpus=all

Then ran:

xhost +si:localuser:$(whoami)
zwift

This got me up and running for today's ride.

[glennvl]

I updated ~/.config/zwift/config to include:

* DONT_CHECK=1

* CONTAINER_TOOL=podman

* NETWORKING=host

* CONTAINER_EXTRA_ARGS=--gpus=all

Then ran:

xhost +si:localuser:$(whoami) zwift

This got me up and running for today's ride.

Could you show the DRYRUN=1 output of the configuration that does not work and the configuration that works? Because the config you show looks a bit weird tot me.

• DONT_CHECK=1: Disables updates for the zwift.sh script. There were no changes to this script in the meantime, so it does not make a difference. It's a bad idea in general though to disable scirpt updates, it could easily break things for you in the future.
• CONTAINER_TOOL=podman: This is the default if podman is installed and the tool is not explicitly set.
• NETWORKING=host: You mentioned in your original post that you were already using host networking, so this is not a new item you added? If it is, this may be what fixed it for you.
• CONTAINER_EXTRA_ARGS=--gpus=all: You mentioned in your original post that zwift launches and you can enter credentials in the ui, so you did not have issues with your graphics card? If you want to change the graphics flags, you should use VGA_DEVICE_FLAG instead of CONTAINER_EXTRA_ARGS. The former will replace the default vga flags with the ones you provide, the latter appends. In the case of nvidia with podman, the default is --device="nvidia.com/gpu=all". What you end up with now is --device="nvidia.com/gpu=all" --gpus=all.