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

build: remove unneeded include for FreeBSD

2016-04-28T14:11:37+00:00

The include  causes a conflicting definition of an RPC
'struct pmap':

    --- fuse-helpers.lo ---
    In file included from /usr/include/rpc/rpc.h:73:0,
                     from ../../../../libglusterfs/src/glusterfs-fops.h:35,
                     from /usr/home/jenkins/root/workspace/freebsd-smoke/libglusterfs/src/glusterfs.h:32,
                     from /usr/home/jenkins/root/workspace/freebsd-smoke/xlators/mount/fuse/src/fuse-bridge.h:22,
                     from /usr/home/jenkins/root/workspace/freebsd-smoke/xlators/mount/fuse/src/fuse-helpers.c:26:
    /usr/include/rpc/pmap_prot.h:89:8: error: redefinition of 'struct pmap'
     struct pmap {
            ^
    In file included from /usr/include/vm/pmap.h:90:0,
                     from /usr/include/sys/user.h:52,
                     from /usr/home/jenkins/root/workspace/freebsd-smoke/xlators/mount/fuse/src/fuse-helpers.c:19:
    /usr/include/machine/pmap.h:299:8: note: originally defined here
     struct pmap {
            ^

It seems that building on FreeBSD still functions without any additional
warnings or errors, even when the include is removed.

Change-id I98fc8cf7e4b631082c7b203b5a0a77111bec1fb9 identified this
issue, and this build-fix is needed for applying I98fc8cf7.

BUG: 1198849
Change-Id: Ib8241b7dc47eb2c3593d2f8ea1d196178e63d02d
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/14093
NetBSD-regression: NetBSD Build System 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Jeff Darcy

mount/fuse: report ESTALE as ENOENT

2016-04-19T05:23:46+00:00

When the inode/gfid is missing, brick report back as an ESTALE
error. However, most of the applications don't accept ESTALE as an
error for a file-system object missing, changing their behaviour.

For eg., rm -rf ignores ENOENT errors during unlink of
files/directories. But with ESTALE error it doesn't send rmdir on a
directory if unlink had failed with ESTALE for any of the files or
directories within it.

Thanks to Ravishankar N , here is a link as to
why we split up ENOENT into ESTALE and ENOENT.
http://review.gluster.org/#/c/6318/

Change-Id: I467df0fdf22734a8ef20c79ac52606410fad04d1
BUG: 1245065
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/13816
Smoke: Gluster Build System 
Reviewed-by: Jeff Darcy 
Reviewed-by: N Balachandran 
Reviewed-by: Niels de Vos 
Tested-by: N Balachandran 
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

mount/fuse: cleanup an additional inode_ref()

2016-03-16T17:21:03+00:00

commit ca515db0127 introduced a check in
fuse_resolve_inode_simple(). This results in an additional
ref being held on inodes which were obtained through readdirp.
As a result, the inode table keeps growing and entries remain in
the active list even after deletion of such inodes.

Change-Id: I780ec5513990d6ef00ea051ec57ff20e4428081e
BUG: 1317948
Signed-off-by: Vijay Bellur 
Reviewed-on: http://review.gluster.org/13689
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Smoke: Gluster Build System 
Reviewed-by: Niels de Vos

fuse: Address the review comments in the backport

2016-03-10T03:15:33+00:00

Backport @ http://review.gluster.org/#/c/13626/3

Fix a typo error, consolidate the selinux and capability
check in getxattr and setxattr.

Change-Id: I4303de3d4dd00853169b07577311e03cbb912ed7
BUG: 1316327
Signed-off-by: Poornima G 
Reviewed-on: http://review.gluster.org/13653
NetBSD-regression: NetBSD Build System 
Tested-by: Vijay Bellur 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Vijay Bellur

fuse: Add a new mount option capability

2016-03-08T05:57:28+00:00

Originally all security.* xattrs were forbidden if selinux is disabled,
which was causing Samba's acl_xattr module to not work, as it would
store the NTACL in security.NTACL. To fix this http://review.gluster.org/#/c/12826/
was sent, which forbid only security.selinux. This opened up a getxattr
call on security.capability before every write fop and others.

Capabilities can be used without selinux, hence if selinux is disabled,
security.capability cannot be forbidden. Hence adding a new mount
option called capability.

Only when "--capability" or "--selinux" mount option is used,
security.capability is sent to the brick, else it is forbidden.

Change-Id: I77f60e0fb541deaa416159e45c78dd2ae653105e
BUG: 1309462
Signed-off-by: Poornima G 
Reviewed-on: http://review.gluster.org/13540
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Raghavendra G

glusterd: Bug fixes for IPv6 support

2016-02-20T17:16:42+00:00

Problem:
Glusterd not working using ipv6 transport. The idea is with proper glusterd.vol configuration,
1. glusterd needs to listen on default port (240007) as IPv6 TCP listner.
2. Volume creation/deletion/mounting/add-bricks/delete-bricks/peer-probe
   needs to work using ipv6 addresses.
3. Bricks needs to listen on ipv6 addresses.
All the above functionality is needed to say that glusterd supports ipv6 transport and this is broken.

Fix:
When "option transport.address-family inet6" option is present in glusterd.vol
file, it is made sure that glusterd creates listeners using ipv6 sockets only and also the same information is saved
inside brick volume files used by glusterfsd brick process when they are starting.

Tests Run:
Regression tests using ./run-tests.sh
    IPv4: Ran manually till tests/basic/rpm.t .
    IPv6: (Need to add the above mentioned config and also add an entry for "hostname ::1" in /etc/hosts)
        Started failing at ./tests/basic/glusterd/arbiter-volume-probe.t and ran successfully till here

Unit Tests using Ipv6
    peer probe
    add-bricks
    remove-bricks
    create volume
    replace-bricks
    start volume
    stop volume
    delete volume

Change-Id: Iebc96e6cce748b5924ce5da17b0114600ec70a6e
BUG: 1117886
Signed-off-by: Nithin D 
Reviewed-on: http://review.gluster.org/11988
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Atin Mukherjee 
Reviewed-by: Jeff Darcy

fuse: add support for SEEK_HOLE and SEEK_DATA through lseek()

2016-02-11T06:16:50+00:00

The Linux FUSE kernel module has gained support for passing SEEK_HOLE
and SEEK_DATA on through lseek(). This can greatly improve performance
when working with sparse files.

Linux FUSE introduced support for lseek() with version 4.5. The commit
in mainline Linux is 0b5da8db145bfd44266ac964a2636a0cf8d7c286.

URL: http://thread.gmane.org/gmane.comp.file-systems.fuse.devel/14752
Change-Id: I12496d788e59461a3023ddd30e0ea3179007f77e
BUG: 1220173
Signed-off-by: Ravishankar N 
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/11474
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Raghavendra G

Fuse: Add a check for NULL in fuse_itable_dump

2016-02-08T08:24:44+00:00

Problem: Immediately after starting a disperse volume (2+1)
kill one brick and just after that try to mount it
through fuse. This lead to crash.

Our test scripts use process statedumps to determine various things
like whether they are up, connected to bricks etc. It takes some time
for an active_subvol to be be associated with fuse even after mount 
process is daemonized. This time is normally a function of completion 
of handshake with bricks. So, if we try to take statedump in this time 
window, fuse wouldn't have an active_subvol associated with it leading 
to this crash.

This happened while executing ec-notify.t, which contains above steps.

Solution: Check priv and  priv->active_subvol for NULL before
inode_table_dump. If priv->active_subvol is null its perfectly fine to
skip dumping of inode table as inode table is associated with an
active_subvol. A Null active_subvol indicates initialization in
progress and fuse wouldn't even have started reading requests from 
/dev/fuse and hence there wouldn't be any inodes or file system 
activity.

Change-Id: I323a154789edf8182dbd1ac5ec7ae07bf59b2060
BUG: 1299410
Signed-off-by: Ashish Pandey 
Reviewed-on: http://review.gluster.org/13253
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Raghavendra G

fuse: update fuse_kernel.h to version 23

2016-02-07T05:56:46+00:00

The following changes were made upstream:
 - add FUSE_WRITEBACK_CACHE
 - add time_gran to fuse_init_out
 - add reserved space to fuse_init_out
 - add FATTR_CTIME
 - add ctime and ctimensec to fuse_setattr_in
 - add FUSE_RENAME2 request
 - add FUSE_NO_OPEN_SUPPORT flag

Including these changes will make it easier to backport support for
lseek().

Because the fuse_init_out structure changed its size, older versions of
FUSE would fail initializing. When an older version of FUSE is detected,
the fuse_init_out structure is reduced to the previous size. This is
harmless, as the attributes that are not passed, are not used for
earlier versions anyway.

BUG: 1220173
Change-Id: I58c74e161638b2d4ce12fc91a206fdc1b96de14d
Signed-off-by: Ravishankar N 
[ndevos: splitted from http://review.gluster.org/11474
         old version fuse_init_out size correction]
Reviewed-on: http://review.gluster.org/11537
Smoke: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System