similar to: [RFC] big fat transaction ioctl

Hi Chris, This is the extent of my current queue of Btrfs snapshot/subvol/commit stuff. Most of these were posted several months ago. Can be sent upstream during this merge window? Not having this functionality is becoming a bit of a roadblock for our efforts to keep the Ceph data in a consistent state. These patches are also available from

Hi everyone, This patchset is the latest approach I''m using for the Ceph storage daemon to keep track of which data has safely committed to disk. The basic idea is to not use the (problematic) user transaction ioctls at all. Instead, the daemon quiesces its own write requests, initiates an async snapshot, and then continues. The snapshot approach is nice because it provides rollback.

The freeze rwsem is taken by sb_start_intwrite() and dropped during the commit_ or end_transaction(). In the async case, that happens in a worker thread. Tell lockdep the calling thread is releasing ownership of the rwsem and the async thread is picking it up. Josef and I worked out a more complicated solution that made the async commit thread join and potentially get a later transaction, but

We need to truncate page cache pages for the clone ioctl target range or else we''ll confuse ourselves to no end. If the old data was cached, we''ll still see it (until remount). If it was dirty we''ll get a mix of old and new data if the page(s) are partially updated. Signed-off-by: Sage Weil <sage@newdream.net> --- fs/btrfs/ioctl.c | 4 ++++ 1 files changed,

Hi Chris- Here''s a simpler version of the patch that drops the unrelated sync_fs stuff. thanks- sage The ''flushoncommit'' mount option forces any data dirtied by a write in a prior transaction to commit as part of the current commit. This makes the committed state a fully consistent view of the file system from the application''s perspective (i.e., it

2011/10/26 Sage Weil <sage@newdream.net>: > On Wed, 26 Oct 2011, Christian Brunner wrote: >> >> > Christian, have you tweaked those settings in your ceph.conf?  It would be >> >> > something like ''journal dio = false''.  If not, can you verify that >> >> > directio shows true when the journal is initialized from your osd log?

* Sage Weil (sage@newdream.net) [091216 17:55]: > On Wed, 16 Dec 2009, Li Dongyang wrote: > > > Have a look at line 998, ioctl.c, inside btrfs_ioctl_clone(), > > the src->i_size(the size of the testfile created by touch) is just 0, and this > > will cause btrfs_ioctl_clone just return -EINVAL. > > I''m not sure if it makes sense to clone a file which

This check for a user transaction is a bogus: fsync() with an open user transaction is verboten. If the caller is the same process who started a user transaction, and this is a different file, we will deadlock. If it''s the same file, this check means we won''t deadlock, but the transaction will be prematurely closed out. Any user transaction ioctl users should be careful to

Hi I am hitting this Warning reproducible, the workload is a ceph osd, kernel ist 3.1.0-rc5. Best Regards, martin [ 5472.099766] ------------[ cut here ]------------ [ 5472.099833] WARNING: at fs/btrfs/inode.c:2193 btrfs_orphan_commit_root+0xb0/0xc0 [btrfs]() [ 5472.099838] Hardware name: MS-96B3 [ 5472.099842] Modules linked in: radeon ttm drm_kms_helper drm i2c_algo_bit psmouse sp5100_tco

Add a mount option user_subvol_rm_allowed that allows users to delete a (potentially non-empty!) subvol when they would otherwise we allowed to do an rmdir(2). We duplicate the may_delete() checks from the core VFS code to implement identical security checks (minus the directory size check). Signed-off-by: Sage Weil <sage@newdream.net> --- fs/btrfs/ctree.h | 1 + fs/btrfs/ioctl.c |

Hi everyone, I'm setting up a clustered Samba+CTDB in front of CephFS and am running into an issue during failover. For the most part everything seems to work: the IP moves quickly, smbd is started on the right node, etc, but if there is an IO load from a client during failover (e.g., copying a big directory full of files in File Explorer), it pauses for a couple of seconds and then pops up

I am testing a Btrfs root file system with Debian (kernel 2.6.32) under KVM. jason@vrtl:~$ touch testfile jason@vrtl:~$ cp --reflink testfile /tmp cp: failed to clone `/tmp/testfile'': Invalid argument This is with GNU Coreutils 8.0 taken from debian Sid. Is this a Coreutils issue, a Btrfs problem or something in my local configuration? All the other issues that I have encountered so

After running ceph on XFS for some time, I decided to try btrfs again. Performance with the current "for-linux-min" branch and big metadata is much better. The only problem (?) I''m still seeing is a warning that seems to occur from time to time: [87703.784552] ------------[ cut here ]------------ [87703.789759] WARNING: at fs/btrfs/inode.c:2103

Hi all I have run into a case where I don't understand why predict.lrm and predict.glm don't yield the same results. My data look like this: set.seed(1) library(Design); ilogit <- function(x) { 1/(1+exp(-x)) } ORDER <- factor(sample(c("mc-sc", "sc-mc"), 403, TRUE)) CONJ <- factor(sample(c("als", "bevor", "nachdem",

Hi, we are running a ceph cluster with btrfs as it''s base filesystem (kernel 3.0). At the beginning everything worked very well, but after a few days (2-3) things are getting very slow. When I look at the object store servers I see heavy disk-i/o on the btrfs filesystems (disk utilization is between 60% and 100%). I also did some tracing on the Cepp-Object-Store-Daemon, but I''m

Hi-- This is a question with a trivial and obvious answer, I'm sure, but I can't seem to find it in the help files and books that I have handy. I have a dataframe consisting of two columns, "Gene_Name," a list of gene symbols, and "Number," a numeric measure of how frequently a tag representing that gene showed up in a SAGE library. Several of the genes are

All, I am curious what is used typically for the file system replication and how do you make sure that it is consistent. So for example when using large 3TB+ sata/NL-sas drives. Is is typical to replicate three times to get similar protection to raid 6? Also what is typically done to ensure that all replicas are in place and consistent? A cron that stats of ls's the file system from a

Hi! I heard many things over ogg, but all what i can see is shit. O.K. The oggdrop.exe play with me, have this tool a version number? I can't see it. Every other tool don't run under my system (Win98; Athlon1,5GHz; 768MB RAM). For example every version of oggenc produce the error: "OGGENC.EXE kann nicht ausgef?hrt werden"! There are a german mailing list? An how can i register

Hi all, I am attempting to get Sage to support their software if the files are on Samba (Linux). They (like a lot of people I've heard) say "We don't support Linux, Goodbye." even when the client is on Win98/NT. We have been having trouble on a particular site, and they refuse to support Sage Line 50 if the files are on Linux which is absurd, because Linux is much more

I'm trying to build R 2.10.1 on a Sun Blade 1000 running Solaris 10 (03/05 release). I've installed iconv 1.13.1 and used: CPPFLAGS="-I /export/home/drkirkby/sage-4.3.3.alpha0/local/include" (which is where iconv is) LDFLAGS= -R/export/home/drkirkby/sage-4.3.3.alpha0/local/lib -L/export/home/drkirkby/sage-4.3.3.alpha0/local/lib The build of R fails as below. gcc