2022-03-10 16:01:04

by Christoph Hellwig

[permalink] [raw]
Subject: Re: [PATCH] block: aoe: handle device timeouts atomic-safe

On Thu, Mar 10, 2022 at 04:17:37PM +0100, Valentin Kleibel wrote:
> If an aoe device does not respond to any packet on the network layer within
> aoe_deadsecs, the driver tries to mark the device as down, fail all I/O and
> clean out the queue.

I think the problem is that aoe even tries to queue up request
internally. It really needs to switch to use BLK_MQ_F_BLOCKING and
stop queuing up the requests internall at all.


2022-03-28 22:51:41

by Valentin Kleibel

[permalink] [raw]
Subject: Re: [PATCH] block: aoe: handle device timeouts atomic-safe

On 10/03/2022 16:29, Christoph Hellwig wrote:
> On Thu, Mar 10, 2022 at 04:17:37PM +0100, Valentin Kleibel wrote:
>> If an aoe device does not respond to any packet on the network layer within
>> aoe_deadsecs, the driver tries to mark the device as down, fail all I/O and
>> clean out the queue.
>
> I think the problem is that aoe even tries to queue up request
> internally. It really needs to switch to use BLK_MQ_F_BLOCKING and
> stop queuing up the requests internall at all.

Unfortunately I'm not able to assess this.
It would be a more complicated change than I'd be able to provide and
the proposed patch fixes the existing issue.

Cheers,
Valentin