Received-SPF: pass (google.com: domain of linux-kernel+bounces-71705-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Precedence: bulk
MIME-Version: 1.0
References: <CAO9zADyveGGBNKsD-JFoOeGZ3om3wMhkqvPc2mvxp-yBzk5mrg@mail.gmail.com>
 <f8a92dd7-8509-428d-8982-4ba94063081c@infradead.org>
In-Reply-To: <f8a92dd7-8509-428d-8982-4ba94063081c@infradead.org>
From: Justin Piszcz <jpiszcz@lucidpixels.com>
Date: Mon, 19 Feb 2024 11:49:00 -0500
Message-ID: <CAO9zADxtxd6ZSBcy79fCuCRbZFha6i4d-XW13f5smn0J0Zay8A@mail.gmail.com>
Subject: Re: Linux FS6712X-EB92 5.13.x - page allocation failure followed by a
 controller reset and the drive drops out of the array
To: Randy Dunlap <rdunlap@infradead.org>
Cc: LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sat, Feb 17, 2024 at 11:34=E2=80=AFPM Randy Dunlap <rdunlap@infradead.or=
g> wrote:
>
> Hi Justin,

[ .. ]

> I can't answer your question. I suggest that you ask it on the
> linux-raid@vger.kernel.org  mailing list.

Thanks, I will follow up there.

>
> Also, I have one question:
>
> in this log fragment:
>
> [1698614.263935] SLUB: Unable to allocate memory on node -1,
> gfp=3D0x800(GFP_NOWAIT)
> [1698614.271680]   cache: skbuff_head_cache, object size: 224, buffer
> size: 256, default order: 0, min order: 0
> [1698614.281979]   node 0: slabs: 32, objs: 512, free: 64
> [1933116.236646] nvme nvme9: I/O 119 QID 2 timeout, aborting
> [1933116.242365] nvme nvme9: I/O 120 QID 2 timeout, aborting
> [1933141.324640] nvme nvme9: I/O 1 QID 0 timeout, reset controller
> [1933146.444701] nvme nvme9: I/O 119 QID 2 timeout, reset controller
> [1933215.826997] nvme nvme9: Device not ready; aborting reset, CSTS=3D0x1
>
> there are roughly 4 days between the 1698614 log entry and the
> 1933116 log entry. Is that (logging) accurate? Did it miss anything?
> It just seems odd to me.
Not that I had seen within the logs, when I tried to access the device
via Windows file share it had been awhile since it was last accessed;
however, there is no sleep setting for the NVME SSDs (only
external/USB HDDs).