From: Boaz Harrosh <bharrosh@panasas.com>
To: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>,
Chris Mason <chris.mason@oracle.com>,
Gregory Farnum <gregory.farnum@dreamhost.com>,
Bernd Schubert <bernd.schubert@itwm.fraunhofer.de>,
Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
<linux-scsi@vger.kernel.org>,
Sven Breuner <sven.breuner@itwm.fraunhofer.de>,
Chuck Lever <chuck.lever@oracle.com>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
<lsf-pc@lists.linux-foundation.org>
Subject: Re: [Lsf-pc] [LSF/MM TOPIC] end-to-end data and metadata corruption detection
Date: Thu, 2 Feb 2012 01:03:00 +0200 [thread overview]
Message-ID: <4F29C4A4.8020005@panasas.com> (raw)
In-Reply-To: <yq1d39ys9n1.fsf@sermon.lab.mkp.net>
On 02/01/2012 08:15 PM, Martin K. Petersen wrote:
>>>>>> "James" == James Bottomley <James.Bottomley@HansenPartnership.com> writes:
>
> IOW, the filesystem should only ever act as a conduit. The only real
> challenge as far as I can tell is how to handle concurrent protected and
> unprotected updates to a page. If a non-PI-aware app updates a cached
> page which is subsequently read by an app requesting PI that means we
> may have to force a write-out followed by a read to get valid PI. We
> could synthesize it to avoid the I/O but I think that would be violating
> the premise of protected transfer. Another option is to have an
> exclusive write access mechanism that only permits either protected or
> unprotected access to a page.
Yes. a protected write implies a byte-range locking on the file.
(And can be implemented with one)
Also the open() call demands an O_PROTECT option. Protection is
then a file attribute as well.
>
Boaz
WARNING: multiple messages have this Message-ID (diff)
From: Boaz Harrosh <bharrosh-C4P08NqkoRlBDgjK7y7TUQ@public.gmane.org>
To: "Martin K. Petersen"
<martin.petersen-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>
Cc: James Bottomley
<James.Bottomley-d9PhHud1JfjCXq6kfMZ53/egYHeGw8Jk@public.gmane.org>,
Chris Mason <chris.mason-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
Gregory Farnum
<gregory.farnum-OZUH0SiS3Izby3iVrkZq2A@public.gmane.org>,
Bernd Schubert
<bernd.schubert-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>,
Linux NFS Mailing List
<linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
linux-scsi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Sven Breuner
<sven.breuner-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>,
Chuck Lever <chuck.lever-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
linux-fsdevel
<linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
lsf-pc-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org
Subject: Re: [Lsf-pc] [LSF/MM TOPIC] end-to-end data and metadata corruption detection
Date: Thu, 2 Feb 2012 01:03:00 +0200 [thread overview]
Message-ID: <4F29C4A4.8020005@panasas.com> (raw)
In-Reply-To: <yq1d39ys9n1.fsf-+q57XtR/GgMb6DWv4sQWN6xOck334EZe@public.gmane.org>
On 02/01/2012 08:15 PM, Martin K. Petersen wrote:
>>>>>> "James" == James Bottomley <James.Bottomley-d9PhHud1JfjCXq6kfMZ53/egYHeGw8Jk@public.gmane.org> writes:
>
> IOW, the filesystem should only ever act as a conduit. The only real
> challenge as far as I can tell is how to handle concurrent protected and
> unprotected updates to a page. If a non-PI-aware app updates a cached
> page which is subsequently read by an app requesting PI that means we
> may have to force a write-out followed by a read to get valid PI. We
> could synthesize it to avoid the I/O but I think that would be violating
> the premise of protected transfer. Another option is to have an
> exclusive write access mechanism that only permits either protected or
> unprotected access to a page.
Yes. a protected write implies a byte-range locking on the file.
(And can be implemented with one)
Also the open() call demands an O_PROTECT option. Protection is
then a file attribute as well.
>
Boaz
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
WARNING: multiple messages have this Message-ID (diff)
From: Boaz Harrosh <bharrosh-C4P08NqkoRlBDgjK7y7TUQ@public.gmane.org>
To: "Martin K. Petersen"
<martin.petersen-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>
Cc: James Bottomley
<James.Bottomley-d9PhHud1JfjCXq6kfMZ53/egYHeGw8Jk@public.gmane.org>,
Chris Mason <chris.mason-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
Gregory Farnum
<gregory.farnum-OZUH0SiS3Izby3iVrkZq2A@public.gmane.org>,
Bernd Schubert
<bernd.schubert-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>,
Linux NFS Mailing List
<linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
<linux-scsi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
Sven Breuner
<sven.breuner-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>,
Chuck Lever <chuck.lever-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
linux-fsdevel
<linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
<lsf-pc-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org>
Subject: Re: [Lsf-pc] [LSF/MM TOPIC] end-to-end data and metadata corruption detection
Date: Thu, 2 Feb 2012 01:03:00 +0200 [thread overview]
Message-ID: <4F29C4A4.8020005@panasas.com> (raw)
In-Reply-To: <yq1d39ys9n1.fsf-+q57XtR/GgMb6DWv4sQWN6xOck334EZe@public.gmane.org>
On 02/01/2012 08:15 PM, Martin K. Petersen wrote:
>>>>>> "James" == James Bottomley <James.Bottomley-d9PhHud1JfjCXq6kfMZ53/egYHeGw8Jk@public.gmane.org> writes:
>
> IOW, the filesystem should only ever act as a conduit. The only real
> challenge as far as I can tell is how to handle concurrent protected and
> unprotected updates to a page. If a non-PI-aware app updates a cached
> page which is subsequently read by an app requesting PI that means we
> may have to force a write-out followed by a read to get valid PI. We
> could synthesize it to avoid the I/O but I think that would be violating
> the premise of protected transfer. Another option is to have an
> exclusive write access mechanism that only permits either protected or
> unprotected access to a page.
Yes. a protected write implies a byte-range locking on the file.
(And can be implemented with one)
Also the open() call demands an O_PROTECT option. Protection is
then a file attribute as well.
>
Boaz
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2012-02-01 23:03 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-17 20:15 [LSF/MM TOPIC] end-to-end data and metadata corruption detection Chuck Lever
2012-01-17 20:15 ` Chuck Lever
2012-01-26 12:31 ` Bernd Schubert
2012-01-26 12:31 ` Bernd Schubert
2012-01-26 14:53 ` Martin K. Petersen
2012-01-26 14:53 ` Martin K. Petersen
2012-01-26 16:27 ` Bernd Schubert
2012-01-26 16:27 ` Bernd Schubert
2012-01-26 23:21 ` James Bottomley
2012-01-31 19:16 ` Bernd Schubert
2012-01-31 19:16 ` Bernd Schubert
2012-01-31 19:21 ` Chuck Lever
2012-01-31 20:04 ` Martin K. Petersen
2012-01-31 2:10 ` Martin K. Petersen
2012-01-31 19:22 ` Bernd Schubert
2012-01-31 19:28 ` Gregory Farnum
2012-02-01 16:45 ` [Lsf-pc] " Chris Mason
2012-02-01 16:52 ` James Bottomley
2012-02-01 17:41 ` Chris Mason
2012-02-01 17:41 ` Chris Mason
2012-02-01 17:59 ` Bernd Schubert
2012-02-01 18:16 ` James Bottomley
2012-02-01 18:30 ` Andrea Arcangeli
2012-02-02 9:04 ` Bernd Schubert
2012-02-02 19:26 ` Andrea Arcangeli
2012-02-02 19:46 ` Andreas Dilger
2012-02-02 19:46 ` Andreas Dilger
2012-02-02 22:52 ` Bernd Schubert
2012-02-02 22:52 ` Bernd Schubert
2012-02-01 18:15 ` Martin K. Petersen
2012-02-01 23:03 ` Boaz Harrosh [this message]
2012-02-01 23:03 ` Boaz Harrosh
2012-02-01 23:03 ` Boaz Harrosh
[not found] ` <DE0353DF-83EA-480E-9C42-1EE760D6EE41@dilger.ca>
2012-01-31 2:22 ` Martin K. Petersen
2012-01-31 2:22 ` Martin K. Petersen
2012-01-26 15:36 ` Martin K. Petersen
2012-01-26 15:36 ` Martin K. Petersen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4F29C4A4.8020005@panasas.com \
--to=bharrosh@panasas.com \
--cc=James.Bottomley@HansenPartnership.com \
--cc=bernd.schubert@itwm.fraunhofer.de \
--cc=chris.mason@oracle.com \
--cc=chuck.lever@oracle.com \
--cc=gregory.farnum@dreamhost.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=lsf-pc@lists.linux-foundation.org \
--cc=martin.petersen@oracle.com \
--cc=sven.breuner@itwm.fraunhofer.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.