[Drbd-dev] [PATCH] drbd: fix throttling on newly created DM backing devices

[Lars Ellenberg]

On Wed, Sep 03, 2014 at 03:02:36PM +0200, Imre Palik wrote:
> From: "Palik, Imre" <imrep at amazon.de>
> 
> If the drbd backing device is a new device mapper device (e.g., a
> dm-linear mapping of an existing block device that contains data), the
> counters are initially 0 even though the device contains useful
> data. This causes throttling until something accesses the drbd device
> or the backing device.
> 
> The patch disables throttling, as long as only resync is responsible
> for disk activity on a freshly created device.

You are right, the !device->rs_last_events was supposed to be an
indicator of "just started resync, please call once to initialize".
But if curr_events stays 0, which is quite possible given the bursty
nature of the DRBD resync, that will stay 0, and keep throttling.

I'm not sure I like your patch, though.
We may want to properly initialize rs_last_events instead.

I'll think about that...

	Lars Ellenberg

> Reported-by: Mikhail Sugakov <msugakov at amazon.de>
> Cc: Matt Wilson <msw at amazon.com>
> Signed-off-by: Imre Palik <imrep at amazon.de>
> ---
>  drivers/block/drbd/drbd_receiver.c |    3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
> index 9342b8d..511d2eb 100644
> --- a/drivers/block/drbd/drbd_receiver.c
> +++ b/drivers/block/drbd/drbd_receiver.c
> @@ -2482,7 +2482,8 @@ bool drbd_rs_c_min_rate_throttle(struct drbd_device *device)
>  			atomic_read(&device->rs_sect_ev);
>  
>  	if (atomic_read(&device->ap_actlog_cnt)
> -	    || !device->rs_last_events || curr_events - device->rs_last_events > 64) {
> +		|| (curr_events && (!device->rs_last_events
> +					|| curr_events - device->rs_last_events > 64))) {
>  		unsigned long rs_left;
>  		int i;
>  
> -- 
> 1.7.9.5
>