glusterfs.git/xlators/storage, branch v3.9dev

dht/afr/client/posix: Fail mkdir without gfid-req

2016-04-29T13:57:39+00:00

Do not allow directory creations without gfids as
after the directories are created, operations
on them fail anyway. So it is better to fail mkdir.

BUG: 1317361
Change-Id: I8f8e3b38bbded1960b7215bac0432500f7e78038
Signed-off-by: Pranith Kumar K 
Reviewed-on: http://review.gluster.org/13690
Smoke: Gluster Build System 
Reviewed-by: Krutika Dhananjay 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System

storage/posix: change the conflicting msg-id

2016-04-26T08:24:12+00:00

Change-Id: I11b2ffb73b2358380771921548fa2c51da6ad93f
BUG: 1323040
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/14062
Reviewed-by: N Balachandran 
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Poornima G 
CentOS-regression: Gluster Build System

cluster/distribute: detect stale layouts in entry fops

2016-04-22T17:28:54+00:00

dht_mkdir ()
{
      first-hashed-subvol = hashed-subvol for "bname" in in-memory
                            layout of "parent";
      inodelk (SETLKW, parent, "LAYOUT_HEAL_DOMAIN", "can be any
               subvol, but we choose first-hashed-subvol randomly");
      {
begin:
            hashed-subvol = hashed-subvol for "bname" in in-memory
                            layout of "parent";
            hash-range = extract hashe-range from layout of "parent";

            ret = mkdir (parent/bname, hashed-subvol, hash-range);
            if (ret == "hash-value doesn't fall into layout stored on
                       the brick (this error is returned by posix-mkdir)")
            {
                refresh_parent_layout ();
                goto begin;
            }

      }
      inodelk (UNLCK, parent, "LAYOUT_HEAL_DOMAIN",
               "first-hashed-subvol");

      proceed with other parts of dht_mkdir;
}

posix_mkdir (parent/bname, client-hash-range)
{

       disk-hash-range = getxattr (parent, "dht-layout-key");
       if (disk-hash-range != client-hash-range) {
              fail-with-error ("hash-value doesn't fall into layout
                                stored on the brick");
              return 0;
       }

       continue-with-posix-mkdir;
}

Similar changes need to be done for dentry operations like create,
symlink, link, unlink, rmdir, rename. These will be addressed in
subsequent patches. This patch addresses only mkdir codepath.

This change breaks stripe tests, as on some striped subvols dht layout
xattrs are not set for some reason. This results in failure of
mkdir. Since striped volumes are always created with dht, some tests
associated with stripe also fail. So, I am making following tests
changes (since stripe is out of maintainance):
* modify ./tests/basic/rpc-coverage.t to not to use striped volumes
* mark all (2) tests in tests/bugs/stripe/ as bad tests

Change-Id: Idd1ae879f24a48303dc743c1bb4d91f89a629e25
BUG: 1323040
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/13885
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: N Balachandran

posix: Add lease() fop

2016-04-21T11:32:12+00:00

Change-Id: I20007d7ff4536ae8180425c84a1aa7863a6f2447
BUG: 1319992
Signed-off-by: Poornima G 
Reviewed-on: http://review.gluster.org/11598
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Raghavendra Talur

arbiter: write performance improvement

2016-04-11T11:32:37+00:00

Problem: The throughput for a 'dd' workload was much less for arbiter
configuration when compared to normal replica-3 volume. There were 2
issues:

i)arbiter_writev was using the request dict as response dict while
unwinding, leading to incorect GLUSTERFS_WRITE_IS_APPEND and
GLUSTERFS_OPEN_FD_COUNT values (=4), leading to immediate post-ops
because is_afr_delayed_changelog_post_op_needed() failed due to
afr_are_multiple_fds_opened() check.

ii) The arbiter code in afr was setting local->transaction.{start and len} =0
to take full file locks. What this meant was even for simultaenous but
non-overlapping writevs, afr_transaction_eager_lock_init() was not
happening because afr_locals_overlap() always stays true. Consequently
is_afr_delayed_changelog_post_op_needed() failed due to
local->delayed_post_op not being set.

Fix:
i) Send appropriate response dict values in arbiter_writev.
ii) Modify flock params instead of local->transaction.{start and len} to
take full file locks in the transaction.

Also changed _fill_writev_xdata() in posix to fill rsp_xdata for
whatever key is requested for.

Change-Id: I1c5fc5e98aba49ade540bb441a022e65b753432a
BUG: 1324004
Signed-off-by: Ravishankar N 
Reported-by: Robert Rauch 
Reported-by: Russel Purinton 
Reviewed-on: http://review.gluster.org/13906
Reviewed-by: Pranith Kumar Karampuri 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System

dht: add "nuke" functionality for efficient server-side deletion

2016-04-07T15:07:55+00:00

This turns a special xattr into an rmdir with flags set.  When that hits
the posix translator on the server side, that causes the file/directory
to be moved into the special "landfill" directory.  From there, the
posix janitor thread will take care of deleting it entirely on the
server side - traversing it recursively if necessary.  A couple of
secondary issues were fixed to make this effective.

 * FUSE now ensures that setxattr values are NUL terminated.

 * The janitor thread now gets woken up immediately when something is
   placed in 'landfill' instead of only when file descriptors need to be
   closed.

 * The default landfill-emptying interval was reduced to 10s.

To use the feature, issue a setxattr something like this:

   setfattr -n glusterfs.dht.nuke -v "" /mnt/glusterfs/vol/some_dir

The value doesn't actually matter; the mere receipt of a request with
this key is sufficient.  Some day it might be useful to allow setting a
required value as a sort of password, so that only those who know it can
access the underlying special functionality.

Change-Id: I8a343c2cdb40a76d5a06c707191fb67babb8514f
Signed-off-by: Jeff Darcy 
Reviewed-on: http://review.gluster.org/13878
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Raghavendra G

storage/posix: send proper iatt attributes for the root inode

2016-03-26T16:33:48+00:00

* changes in posix to send proper iatt attributes for the root directory
  when ancestry is built. Before posix was filling only the gfid and the
  inode type in the iatt structure keeping rest of the fields zeros. This
  was cached by posix-acl and used to send EACCES when some fops came on
  that object if the uid of the caller is same as the uid of the object on
  the disk.

* getting and setting inode_ctx in function 'posix_acl_ctx_get' is not atomic
  and can lead to memory leak when there are multiple looups for an
  inode at same time. This patch fix this problem

* Linking an inode in posix_build_ancestry, can cause a race in
  posix_acl.
  When parent inode is linked in posix_build_ancestry, and before
  it reaches posix_acl_readdirp_cbkc, reate/lookup can
  come on a leaf-inode, as parent-inode-ctx not yet updated
  in posix_acl_readdirp_cbk, create/lookup can fail
  with EACCESS. So do the inode linking in the quota xlator

Change-Id: I3101eefb65551cc4162c4ff2963be1b73deacd6d
BUG: 1320818
Signed-off-by: Raghavendra Bhat 
Reviewed-on: http://review.gluster.org/13730
Tested-by: Vijaikumar Mallikarjuna 
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Raghavendra G

storage/posix: Typo in log message

2016-03-17T16:56:55+00:00

Fixed missing spaces between some words in a log message.
Signed-off-by: N Balachandran 

Change-Id: I24398e6dd2ab83956827fc905f43cffd7e699c51
BUG: 1318107
Reviewed-on: http://review.gluster.org/13747
Smoke: Gluster Build System 
Tested-by: N Balachandran 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy

posix: Filter gsyncd stime xattr

2016-03-16T05:00:21+00:00

Filter gsyncd stime xattr in lookup as well.
The value of stime would be different among
replica bricks and EC bricks. AFR and EC
should not take any action on these as it
could be different.

Change-Id: If577f6115b36e036af2292ea0eaae93110f006ba
BUG: 1296496
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/13678
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Pranith Kumar Karampuri

geo-rep: Mask xtime and stime xattrs

2016-02-26T09:52:12+00:00

Allow access to xtime and stime xattrs only
to gsyncd client and mask them for the rest.

This is to prevent afr from performing self
healing on marker xtime and geo-rep stime
xattr which is not expected as each of which
gets updated them from backend brick and
should not be healed.

Change-Id: I24c30f3cfac636a55fd55be989f8db9f8ca10856
BUG: 1296496
Signed-off-by: Kotresh HR 
Reviewed-on: http://review.gluster.org/13242
NetBSD-regression: NetBSD Build System 
Reviewed-by: Vijaikumar Mallikarjuna 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Venky Shankar