[pve-devel] [PATCH qemu-server v3] cd rom handling: return a clearer error when there is no cd rom drive

[Thomas Lamprecht]

On 19/12/2022 13:23, Stefan Sterz wrote:
> when a vm is configured to use a physical cd rom drive but there is no
> such drive a cryptic "uninitialized value" error is thrown. this is
> due to `$path` not being defined in `sub
> print_drive_commandline_full` in this case. warn that no cd rom drive
> is available instead.
> 
> note that the error was basically cosmetic as the vm would start just
> fine.
> 
> forum thread: https://forum.proxmox.com/threads/119592/
> 
> Signed-off-by: Stefan Sterz <s.sterz at proxmox.com>
> ---
> changes from v2 (thanks @ Fiona Ebner):
> 
> * added newline to message
> * cache the fact that no drive is available so we warn only once
> * adjusted error message again
> 
> changes from v1: 
> 
> * make the error message more concise. 
> * use shorter forum link for better formatting in commit message 
>   (thanks @ Fiona Ebner).
> 
>  PVE/QemuServer.pm | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
> index dd6ea3e..11626cf 100644
> --- a/PVE/QemuServer.pm
> +++ b/PVE/QemuServer.pm
> @@ -1287,11 +1287,14 @@ sub option_exists {
>  my $cdrom_path;
>  sub get_cdrom_path {
>  
> -    return  $cdrom_path if $cdrom_path;
> +    return  $cdrom_path if defined($cdrom_path);
>  
>      return $cdrom_path = "/dev/cdrom" if -l "/dev/cdrom";
>      return $cdrom_path = "/dev/cdrom1" if -l "/dev/cdrom1";
>      return $cdrom_path = "/dev/cdrom2" if -l "/dev/cdrom2";
> +
> +    warn "no physical CD-ROM available\n";

why not use `log_warn` so that it shows up in the task worker log
prominently (and handles the trailing newline itself)?

also maybe end the sentence with "available for auto-selection" ?

> +    return $cdrom_path = '';

I know you did not start this, but I really do not like returning the result of
an assignment in dynamically-"typed" languages like perl, it's totally unclear
what side effects could be for most; anyhow can be addressed independent of
this patch.

>  }
>  
>  sub get_iso_path {