Can I reduce the CIFS client connection timeout?

Can I reduce the CIFS client connection timeout?

[Josh Pettett]

Hello,

We have a use case in which we want the the CIFS mount to quickly
timeout and return an error on the client side if the remote server is
down.  This error would then trigger failure-mode logic in our
application. It appears that an error is returned if the server is
unreachable for 120 seconds, but we want the timeout to occur after
about 15 seconds.  Is this value configurable anywhere?

Re: Can I reduce the CIFS client connection timeout?

[Sachin Prabhu]

Hello Josh,

The client will wait 2 * echo interval before marking a server
unresponsive. The default echo interval is set to 60 seconds. You can
tune this to a lower value using the echo_interval=n mount option. In
your case, you can set echo_interval to 8 so that the server is marked
unresponsive after 15 seconds of the server being unavailable.

Upstream patch which added this option.
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=adfeb3e00e8e1b9fb4ad19eb7367e7c272d16003

Explanation of the code before that patch was added:
http://sprabhu.blogspot.in/2015/08/investigation-into-effects-of-server.html

Sachin Prabhu

On Thu, Jan 4, 2018 at 11:15 AM, Josh Pettett <jpettett-qZWgbgh2FR7QT0dZR+AlfA@public.gmane.org> wrote:
> Hello,
>
> We have a use case in which we want the the CIFS mount to quickly
> timeout and return an error on the client side if the remote server is
> down.  This error would then trigger failure-mode logic in our
> application. It appears that an error is returned if the server is
> unreachable for 120 seconds, but we want the timeout to occur after
> about 15 seconds.  Is this value configurable anywhere?
> --
> To unsubscribe from this list: send the line "unsubscribe linux-cifs" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html