Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 21 Jun 2016 14:33:53 +0000
From:      bugzilla-noreply@freebsd.org
To:        freebsd-bugs@FreeBSD.org
Subject:   [Bug 210438] NVME controller failure: panic
Message-ID:  <bug-210438-8@https.bugs.freebsd.org/bugzilla/>

next in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D210438

            Bug ID: 210438
           Summary: NVME controller failure: panic
           Product: Base System
           Version: 10.3-RELEASE
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: kern
          Assignee: freebsd-bugs@FreeBSD.org
          Reporter: borjam@sarenet.es

Created attachment 171648
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=3D171648&action=
=3Dedit
backtrace in ddb

I have had a controller failure in a Supermicro server with 10 NVMe drives.

The controller failure triggered a PANIC and the system panics in subsequent
reboots unless the affected drive is pulled from the system. (The drive is
fine, it seems to be a backplane/controller problem).

I am not sure wether this is a bug or an unavoidable problem, but it would =
be
desirable to avoid a panic due to a disk controller failure.

The first symptom was these messages in /var/log/messages at night, when the
controller
failed, apparently:

Jun 17 03:01:33 nvme1 kernel: nvme3: resetting controller
Jun 17 03:01:33 nvme1 kernel: nvme3: nvme_ctrlr_wait_for_ready called with
desired_val =3D 0 but cc.en =3D 1

In subsequent reboots, the system panics. Screen capture of a backtrace from
ddb is attached. I can't get it generate a proper dump, unfortunately

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-210438-8>