driver/sle15/nvidia-driver fails to parse correct kernel version on SLES15 SP5
When running container with Podman built from a Dockerfile in the the driver/sle15 directory, the container correctly parses kernel version 5.14.21-150500.55.62.2 but not with version 5.14.21-150500.55.65.1.
The first error seems to occur in the _resolve_kernel_version() function. It seems to parse out "5.14.21-150500.55.65 5.14.21". Note the space and the second value, neither of which I think should be there. In addition, when the failed runs include a line with just the kernel version, including flavor, which aren't shown when it succeeds.
The error that is shown in the container logs is:
Resolving Linux kernel version...
Proceeding with Linux kernel version 5.14.21-150500.55.65
5.14.21-150500.55.65-default
Installing Linux kernel source...
Refreshing service 'container-suseconnect-zypp'.
Loading repository data...
Reading installed packages...
No provider of '5.14.21' found.
FATAL: failed to install kernel packages. Ensure SLES subscription is available.
Stopping NVIDIA persistence daemon...
Unloading NVIDIA driver kernel modules...
Unmounting NVIDIA driver rootfs...
By downgrading the kernel (plus uninstalling all Nvidia packages and rebooting), the container runs without error:
Resolving Linux kernel version...
Proceeding with Linux kernel version 5.14.21-150500.55.62-default
Installing Linux kernel source...
Refreshing service 'container-suseconnect-zypp'.
Loading repository data...
Reading installed packages...
Resolving package dependencies...
The following 26 NEW packages are going to be installed:
...