Troubleshooting

This page collects rare, host-specific issues that do not affect most Sailwright setups.

Windows: Cygwin Ansible Shadowed by a Host Installation

On Windows with Cygwin, the Ansible installation inside Cygwin can be shadowed by another Python or Ansible installation from the Windows host.

If that happens:

• Do not install Ansible directly on the Windows host.
• Remove conflicting host-level Ansible installations.
• Install and run Ansible through the Cygwin Python environment instead.

macOS: Ansible Fork Safety Crash

Running Ansible on macOS can trigger a fork-safety crash similar to this:

TASK [Gathering Facts] ***************************************************************************************************************************************
objc[9473]: +[NSNumber initialize] may have been in progress in another thread when fork() was called.
objc[9473]: +[NSNumber initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork() child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug.
ERROR! A worker was found in a dead state

Work around it by setting this environment variable before running Ansible:

export OBJC_DISABLE_INITIALIZE_FORK_SAFETY=YES

Linux: virsh domifaddr Returns No IP

Linux libvirt provisioning waits for virsh domifaddr to report a guest IPv4 address. Sailwright checks the guest agent first, then libvirt DHCP leases. If both sources return no address, provisioning cannot build the temporary Ansible inventory.

First confirm the VM is running and query the same libvirt connection Sailwright uses:

virsh --connect qemu:///system domifaddr <domain> --source agent
virsh --connect qemu:///system domifaddr <domain> --source lease

If the agent lookup is empty, wait for the guest to finish booting and make sure the image has qemu-guest-agent installed and running. If the lease lookup is empty, make sure the VM is attached to a libvirt-managed network with visible DHCP leases:

virsh --connect qemu:///system net-info default
virsh --connect qemu:///system net-dhcp-leases default

When the default network exists but is inactive, enable it:

sudo virsh --connect qemu:///system net-start default
sudo virsh --connect qemu:///system net-autostart default

For DEV_ALCHEMY_LIBVIRT_URI=qemu:///session, run the same checks against qemu:///session. Session VMs use libvirt user-mode networking by default, so guest-agent visibility is usually the most reliable IP source.