[dm-devel] multipath-tools: release uxsocket and resource when cancel thread

[Martin Wilck]

On Wed, 2017-11-29 at 09:35 +0000, Wuchongyun wrote:
> Hi ,
> 
> Issue description: 
> when multipathd initilaze and call uxsock_listen to create unix
> domain socket, but return -1 and the errno is 98 
> and then the uxsock_listen return null. After multipathd startup we
> can't receive any multipathd commands to 
> finish the new multipath creation anymore!
> 
> We found that uxlsnr thread's cleanup function not close the sockets
> and also not release the clients when 
> cancel thread, the domain socket will be release by the system. In
> any special environment like themachine's 
> load is very heavy, the system may not close the old domain socket
> when we try to create and bind the domain
> socket may return 98 (Address already in use).
> 
> And also we make some experiments:
> In uxsock_cleanup if we close the ux_sock first and then immdediately
> call ux_socket_listen to create new ux_sock
> and initialization will be OK; If we don't close the ux_sock and call
> ux_socket_listen will return -1 and errno = 98.
> 
> So we believe that close uxsocket and release clients when cancel
> thread might making new starting multipathd 
> thread can create uxsocket successfully, and might receiving
> multipathd commands properly.
> And also this path can fix clients' memory leak.

I think this is correct. But I have some remarks, see below.

> 
> Thanks,
> Chongyun
> 
> Signed-off-by: wu chongyun <wu.chongyun at h3c.com>
> ---
> multipathd/uxlsnr.c |   16 +++++++++++++++-
> 1 file changed, 15 insertions(+), 1 deletion(-)
> 
> diff --git a/multipathd/uxlsnr.c b/multipathd/uxlsnr.c
> index 98ac25a..de6950b 100644
> --- a/multipathd/uxlsnr.c
> +++ b/multipathd/uxlsnr.c
> @@ -139,6 +139,20 @@ void check_timeout(struct timespec start_time,
> char *inbuf,
> 
>  void uxsock_cleanup(void *arg)
> {
> +       struct client *client_loop;
> +       struct client *client_tmp;
> +       int ux_sock = (int)arg;
> +
> +       /*
> +       * no need to protect clients, 
> +       * because all operations to clients are only from one
> thread(uxlsnr)
> +       */
> +       list_for_each_entry_safe(client_loop, client_tmp, &clients,
> node) {
> +                dead_client(client_loop);

This takes and releases the client_lock in every loop iteration.
I'd rather take the lock, loop over all clients freeing the resources,
and relase.

> +       }
> +
> +       close(ux_sock);
> +
>        cli_exit();
>        free_polls();
> }
> @@ -162,7 +176,7 @@ void * uxsock_listen(uxsock_trigger_fn
> uxsock_trigger, void * trigger_data)
>                 return NULL;
>        }
> 
> -        pthread_cleanup_push(uxsock_cleanup, NULL);
> +       pthread_cleanup_push(uxsock_cleanup, (void*)ux_sock);

This patch has whitespace issues.

Regards,
Martin

-- 
Dr. Martin Wilck <mwilck at suse.com>, Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)