Re: zero-length files in snapshots

2010-02-13 Thread Sander
Mike Fedyk wrote (ao): > On Fri, Feb 12, 2010 at 8:32 AM, Josef Bacik wrote: > > Creating a file is a metadata operation, and _any_ metadata operation has > > to be > > committed to disk when the transaction commits in order to maintain a > > coherent > > fs. ??Thanks, > > What I still don't un

Re: [PATCH 2/2] btrfs: a new tool to manage a btrfs filesystem - man page

2010-02-13 Thread Daniel J Blueman
Hi Geffredo, On Fri, Feb 12, 2010 at 7:01 PM, Goffredo Baroncelli wrote: > This is the man page of the btrfs command. > > diff --git a/man/Makefile b/man/Makefile > index 4e8893b..4a90b75 100644 > --- a/man/Makefile > +++ b/man/Makefile > @@ -7,13 +7,16 @@ mandir = $(prefix)/man >  man8dir = $(ma

Btrfs - Unable to mount, can't read superblock?

2010-02-13 Thread Victor Hooi
heya, This is on kernel 2.6.32.8-1, on Arch Linux 64-bit. I have an external harddisk, with a btrfs filessystem on it. Just now, after a reboot, I seem to be unable to mount it. KDE gives a message about being unable to find the superblock, trying to mount it from the command-line gives something

Re: zero-length files in snapshots

2010-02-13 Thread Mike Fedyk
On Sat, Feb 13, 2010 at 3:25 AM, Sander wrote: > Mike Fedyk wrote (ao): >> On Fri, Feb 12, 2010 at 8:32 AM, Josef Bacik wrote: >> > Creating a file is a metadata operation, and _any_ metadata operation has >> > to be >> > committed to disk when the transaction commits in order to maintain a >>

btrfs for clustered/distributed storadge (Isilon, Panasas, ceph)

2010-02-13 Thread rk
.. it would be great if btrfs could be expended about abilities that OneFS file system has (FS by Isilon) .. there is ceph open source clustered/distributed storage system that is using btrfs, maybe work with those people would be a good start http://ceph.newdream.net/about/ http://ceph.newdream.ne

Re: [PATCH V2] Btrfs: Full direct I/O and AIO read implementation.

2010-02-13 Thread jim owens
Josef Bacik wrote: On Wed, Feb 10, 2010 at 01:53:50PM -0500, jim owens wrote: + /* traditional 512-byte device sector alignment is the +* minimum required. if they have a larger sector disk +* (possibly multiple sizes in the filesystem) and need + * a larger alignment for

[Regression] Filesystem I/O is CPU-bound in rc7 and rc8

2010-02-13 Thread James Cloos
Sometime between rc6 and rc7 all filesystem I/O started using 100% CPU, usually on the order of 60% sys, 40% user. I've tried this with each of ext4, jfs and btrfs filesystems. All show the same issue. Using dd(1) to read from the block specials directly works as well and as fast as it always ha

[PATCH] bcp: fix off-by-one errors in path handling

2010-02-13 Thread Eduard - Gabriel Munteanu
This fixes a bug which causes the first character of each filename in the destination to be omitted. Signed-off-by: Eduard - Gabriel Munteanu --- bcp |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/bcp b/bcp index 5729e91..c6b4bef 100755 --- a/bcp +++ b/bcp @@ -137,7