lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <57e91c7f7bc1d3e0cd511f828ec8140b@mail.gmail.com>
Date:	Tue, 2 Dec 2014 18:48:18 +0530
From:	Sreekanth Reddy <sreekanth.reddy@...gotech.com>
To:	Jan Vesely <jvesely@...hat.com>, linux-scsi@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	"James E.J. Bottomley" <JBottomley@...allels.com>
Cc:	Tomas Henzl <thenzl@...hat.com>, hch@...radead.org
Subject: RE: [PATCH 0/2] mpt{2,3}sas remove disconnected phys on topology change

Hi James/Chris

We are observing below issue due to this patch set code changes

Issue Description:
     Drives connected Enclosure/Expander won't be visible to the OS if any
one remove and add expander cable with in DMD (Device Missing Delay) time
period (also if any power-off and power-on the Enclosure with in the DMD
period).

i.e.
Due to this code changes; driver will unregister the Enclosure port or
target PHY with SCSI transport layer, whenever driver receives "SAS
TOPOLOGY CHANGE LIST EVENT" with reason code set to "PHY link status
change" and phy link rate is Zero. Firmware usually send this type of "SAS
TOPOLOGY CHANGE LIST EVENT" event whenever drive is missing. but still
here DMD timer has not yet expired so normally driver should not
unregister the target with the SCSI transport layer instead it will move
the state of the  drive to blocked state (so it won't receive any IOs from
SML for this drive).

Normally, Driver should only unregister the target with SCSI transport
layer only when driver receives "SAS TOPOLOGY CHANGE LIST EVENT" with
reason code set to "Target Remove". Firmware will send this event when the
DMD timer expires for the missing drive. If drive comes back with in the
DMD period then firmware won't send this target remove event and so driver
won't unregister the target drive with the SCSI transport layer. Driver
will just unblock the target drive to start IOs to this drive.

So, can you please revert this patch set changes back in the latest
upstream kernel and if possible on stable kernels.
I checked with the Redhat team and there are fine to revert back these
patch set changes.

Regards,
Sreekanth

>-----Original Message-----
>From: Jan Vesely [mailto:jvesely@...hat.com]
>Sent: Thursday, August 01, 2013 3:31 PM
>To: linux-scsi@...r.kernel.org; linux-kernel@...r.kernel.org
>Cc: Jan Vesely; Nandigama, Nagalakshmi; Reddy, Sreekanth; Tomas Henzl
>Subject: [PATCH 0/2] mpt{2,3}sas remove disconnected phys on topology
>change
>
>From: Jan Vesely <jvesely@...hat.com>
>
>These two patches add phy removal on link loss. This change keeps sysfs
>up- to-date with actually connected phys. Without these patches,
>disconnected phys remain listed under their former ports.
>
>tested on both mpt2sas and mpt3sas hw.
>
>CC: Nagalakshmi Nandigama <Nagalakshmi.Nandigama@....com>
>CC: Sreekanth Reddy <Sreekanth.Reddy@....com>
>CC: Tomas Henzl <thenzl@...hat.com>
>Signed-off-by: Jan Vesely <jvesely@...hat.com>
>
>Jan Vesely (2):
>  mpt2sas: Remove phys on topology change.
>  mpt3sas: Remove phys on topology change
>
> drivers/scsi/mpt2sas/mpt2sas_transport.c |    5 ++++-
> drivers/scsi/mpt3sas/mpt3sas_transport.c |    5 ++++-
> 2 files changed, 8 insertions(+), 2 deletions(-)
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ