Re: [PLUG] [plug-announce] TOMORROW - Tue, Apr 19, 2016: PLUG North - "Linux Containers" by Jason Plum and Rich Freeman (*6:30 pm* at CoreDial in Blue Bell)

["Keith C. Perry" <kperry@daotechnologies.com>]

"Both zfs and btrfs redundantly store ALL metadata (I know btrfs even
does this by default on a single disk, but with multiple disks it
ensures redundant copies are on different disks), and everything is
checked on read."

That is my understanding as well.  CRC's are currently only applied to the metadata.  This is my point though, is that a small amount of redundancy is useful and continues to improve XFS as a file system.  Even before the feature was added (its a relatively new) XFS was considered a "high performance" file system- a subjective, time-wise contextual but ultimately a throw away statement because its still only one part of the process- pick an fs, rinse and repeat the same thing.  BTRFS and ZFS in this regard offer optimizations (i.e. better redundancy, more advanced features, consolidated management, etc) at the file system level but that is all they can do and when you net out the optimizations throughout the entire process that is how we get to the overall high level of data durability available the Linux world via a number of file systems and other data storage technology.

"The main thing that has kept me away from xfs is the fact that you
can't shrink the filesystem.  That can be a major inconvenience when
you're moving disks around.  I've run it in the past and it works
reasonably well in general, but for my generic reliable filesystem I
stick with ext4."

That's the number 1 reason people say they don't run XFS.  Its a valid point, however, the question then become how often do you shrink file systems **in place** in production systems.  Obviously you and others do.  For me that answer is never.  From the R&D point of view I'm often moving system and/or data volumes around- its one of things I do most especially between VM containers and bare metal (that's why its never an an in place task).  So, even though I always start with 100Gb disk images when I move the file system, it gets imaged onto the appropriate medium in at the appropriate size with the same one-liner.  It doesn't matter whether its a 16Gb USB stick, 128Gb SSD or 1Tb encrypted VM container with system and data volumes.


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 
Keith C. Perry, MS E.E. 
Owner, DAO Technologies LLC 
(O) +1.215.525.4165 x2033 
(M) +1.215.432.5167 
www.daotechnologies.com

----- Original Message -----
From: "Rich Freeman" <r-plug@thefreemanclan.net>
To: "Philadelphia Linux User's Group Discussion List" <plug@lists.phillylinux.org>
Sent: Tuesday, April 26, 2016 6:32:20 AM
Subject: Re: [PLUG] [plug-announce] TOMORROW - Tue, Apr 19, 2016: PLUG North - "Linux Containers" by Jason Plum and Rich Freeman (*6:30 pm* at CoreDial in Blue Bell)

On Mon, Apr 25, 2016 at 11:37 PM, Keith C. Perry
<kperry@daotechnologies.com> wrote:
> The redundancy in the in redundancy bits for the error correction.  See section 4.1 of this paper:
>
> http://pages.cs.wisc.edu/~vshree/xfs.pdf
>

As far as I can tell that section only describes a taxonomy for the
levels of error detection and recovery any hypothetical filesystem
COULD have, not the features of XFS itself.  If you look at the
following sections you'll see that there is little redundancy in the
data that xfs stores (it sounds like it is mainly just the superblock
- as with most filesystems), and when there is redundancy it generally
isn't used automatically for recovery.

If you look at figures 3-6 you'll see that the best form of recovery
they even observed was fixing some errors at time of remount.

Both zfs and btrfs redundantly store ALL metadata (I know btrfs even
does this by default on a single disk, but with multiple disks it
ensures redundant copies are on different disks), and everything is
checked on read.

>
> Finally, an article from January 2012 discussing why XFS will probably become the default file system over EXT4 and why in the future it will probably be BTRFS if all the issues can be worked out.  I only include it to illustrate how people get wrapped in the religious fight while ignoring the fact that there is a lot that goes into the process so saying "my god is more reliable that your god" gets a bit too far into the weeds.  Funny how that always happens.  :D

The main thing that has kept me away from xfs is the fact that you
can't shrink the filesystem.  That can be a major inconvenience when
you're moving disks around.  I've run it in the past and it works
reasonably well in general, but for my generic reliable filesystem I
stick with ext4.

-- 
Rich
___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug
___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug