[CRIU] [PATCH] restore: wait while processes are dying

[Pavel Emelyanov]

On 10/09/2015 05:41 PM, Andrey Vagin wrote:
> From: Andrew Vagin <avagin at openvz.org>
> 
> If criu restore failed, criu should wait all processes.

... because otherwise ... ... ... what?

> Here we do this only for cases when processes are restored in a pid namespace.
> 
> "Restoring FAILED." will be printed at the end of the log.
> 
> Cc: Nikita Spiridonov <nspiridonov at odin.com>
> Reported-by: Nikita Spiridonov <nspiridonov at odin.com>
> Signed-off-by: Andrew Vagin <avagin at openvz.org>
> ---
>  cr-restore.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/cr-restore.c b/cr-restore.c
> index fb86114..73097a9 100644
> --- a/cr-restore.c
> +++ b/cr-restore.c
> @@ -1940,9 +1940,15 @@ out_kill:
>  	 * otherwise an external proccesses can be killed.
>  	 */
>  	if (root_ns_mask & CLONE_NEWPID) {
> +		int status;
> +
>  		/* Kill init */
>  		if (root_item->pid.real > 0)
>  			kill(root_item->pid.real, SIGKILL);
> +
> +		if (waitpid(root_item->pid.real, &status, 0) < 0)
> +			pr_warn("Unable to wait %d: %s",
> +				root_item->pid.real, strerror(errno));
>  	} else {
>  		struct pstree_item *pi;
>  
>