glusterfs.git/libglusterfs/src/glusterfs.h, branch v3.5.6

cluster/dht: Modified logic of linkto file deletion on non-hashed

2014-09-17T15:34:54+00:00

Currently whenever dht_lookup_everywhere gets called, if in
dht_lookup_everywhere_cbk, a linkto file is found on non-hashed
subvolume, file is unlinked. But there are cases when this file
is under migration. Under such condition, we should avoid deletion
of file.

When  some other rebalance process changes the layout of parent
such that dst_file (w.r.t. migration) falls on non-hashed node,
then may be lookup could have found it as linkto file but just
before unlink, file  is under migration or already migrated
In such cased unlink can be avoided.

Race:
-------
If we have two bricks (brick-1 and brick-2) with initial file "a"
under BaseDir which is hashed as well as cached on (brick-1).

Assume "a"  hashing gives 44.

                              Brick-1              Brick-2

Initial Setup:               BaseDir/a             BaseDir
                             [1-50]                [51-100]

Now add new-brick Brick-3.

1. Rebalance-1 on node Node-1 (Brick-1 node) will reset
the BaseDir Layout.

2. After that it will perform
a)  Create linkto file on  new-hashed (brick-2)
b)  Perform file migration.

1.Rebalance-1 Fixes the base-layout:
                 Brick-1             Brick-2           Brick-3
                 ---------         ----------         ------------
                 BaseDir/a            BaseDir           BaseDir
                  [1-33]              [34-66]           [67-100]

2. Only a) is     BaseDir/a          BaseDir/a(linkto)   BaseDir
   performed                         Create linktofile

Now rebalance 2 on node-2 jumped in and it will perform
step 1 and 2-a.

After (rebal-2, step-1), it changes the layout of the BaseDir.
                    BaseDir/a     BaseDir/a(link)    BaseDir
                    [67-100]           [1-33]        [34-66]

For  (rebale-2, step-2), It will perform lookup at Brick-3 as w.r.t new
layout 44 falls for brick-3. But lookup will fail.
So  dht_lookup_everywhere gets called.

NOTE: On brick-2 by rebalance-1, a linkto file was created.

Currently that linkto files gets deleted by rebalance-2 lookup as it
is considered as stale linkto file.  But  with patch if rebalance is
already in progress or rebalance is over,  linkto file will not be
unlinked. If rebalance is in progress fd will be  open and if rebalance
is over then linkto file wont be set.

Change-Id: I3fee0d28de3c76197325536a9e30099d2413f079
BUG: 1129541
Signed-off-by: Venkatesh Somyajulu 
Reviewed-on: http://review.gluster.org/8345
Tested-by: Gluster Build System 
Reviewed-by: Raghavendra G 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Vijay Bellur 
(cherry picked from commit 966997992bdbd5fffc632bf705678e287ed50bf7)
Signed-off-by: Nithya Balachandran 
Reviewed-on: http://review.gluster.org/8719
Reviewed-by: Jeff Darcy

gfapi: glfs_set_volfile_server() now entertains multiple calls

2014-05-10T02:06:36+00:00

Backport from - http://review.gluster.org/7317

Previous API:

glfs_set_volfile_server (..., const char *host, ...) - single call

New API's:

glfs_set_volfile_server (..., const char *host1, ...)
glfs_set_volfile_server (..., const char *host2, ...)

Multiple calls to this function with different volfile servers,
port or transport-type would create a list of volfile servers
which would be polled during `volfile_fetch_attempts()`

glfs_unset_volfile_server (..., const char *host, ...) to remove
a server from the list (this is provided for future usage)

>> Change-Id: I313efbd3efbd0214e2a71465f33195788df406cc
>> BUG: 986429
>> Signed-off-by: Harshavardhana 
>> Reviewed-on: http://review.gluster.org/7317
>> Tested-by: Gluster Build System 
>> Reviewed-by: Niels de Vos 
>> Reviewed-by: Anand Avati 

Change-Id: If0020f6b8775bdbe987563247c83d59a2d3744ad
BUG: 986429
Signed-off-by: Harshavardhana 
Reviewed-on: http://review.gluster.org/7384
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

features/quota: Metadata cleanup

2014-01-29T06:39:43+00:00

Quota and marker uses 'trusted.glusterfs.quota*' and 'trusted.pgfid*' xattrs to
store its configurations and accounting information and also to build the
parent inode chain in case of absense of path.

Problem:
After disabling and then enabling quota back, the xattrs may contain stale data
leading to impaired accounting and thus improper enforcement.

Solution:
Clean up all the quota related xattrs after quota disable.

Marker xlator implements a virtual xattr to cleanup quota and pgfid xattrs. In
this approach glusterd mounts an auxiliary mount and sends the below command to
all the files by crawling the mountpoint.

 #setfattr -n "glusterfs.quota-xattr-cleanup" -v 1 

Credit:
Krishnan Parthasarathi 
Varun Shastry          
Change-Id: I9380eca58a285dc27dd572de1767aac8f2cd8049
BUG: 969461
Signed-off-by: Varun Shastry 
Reviewed-on: http://review.gluster.org/6369
Reviewed-by: Raghavendra G 
Tested-by: Gluster Build System 
Reviewed-by: Krishnan Parthasarathi 
Reviewed-by: Vijay Bellur 
Reviewed-on: http://review.gluster.org/6838

locks: various fixes

2014-01-14T12:37:56+00:00

- implement ref/unref of entry locks (and fix bad pointer deref crashes)
- code cleanup and deleted various data types
- fix improper read/write lock conflict detection in entrylk
- fix indefinite hang of blocked locks on disconnect
- register locks in client_t synchronously, fix crashes in disconnect path

Change-Id: Id273690c9111b8052139d1847060d1fb5a711924
BUG: 849630
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/6695
Tested-by: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY 
Reviewed-by: Vijay Bellur

cluster/dht: instruct marker whenever it shouldn't do accounting

2013-11-26T18:27:21+00:00

This is needed for two reasons:

* since dht-linkfiles are internal, they shouldn't be accounted.
* hardlink handling in marker is broken. link/unlink of hardlinks
  present in same directory can break marker accounting. Hence, if src
  and dst are in same directory in case of rename, dht - if it breaks
  rename into link/unlink operations - should instruct marker to not to
  do accounting.

Change-Id: I9c9f7384569f75a2792f6450ee7a5279bf751ae7
BUG: 1022995
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/6203
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

core: add dht_is_linkfile helper procedure.

2013-11-26T18:26:44+00:00

components other than distribute (like marker to exclude linkfiles
from being accounted) also need awareness of what constitutes a
linkfile. Hence its good to separate out this functionality into
core.

Change-Id: Ib944eeacc991bb1de464c9e73ee409fc7a689ff1
BUG: 1022995
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/6152
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

features/quota: Improvements to quota

2013-11-26T18:24:02+00:00

* Two stages of quota enforcement is done:

  Soft and hard quota Upon reaching soft quota limit on the directory
  it logs/alerts in the quota daemon log (ie DEFAULT_LOG_DIR/quotad.log)
  and no more writes allowed after hard
  quota limit. After reaching the soft-limit the daemon alerts the
  user/admin repeatively for every 'alert-time', which is
  configurable.

* Quota enforcer is moved to server-side.

  It  takes care of enforcing quota. Since enforcer doesn't have the
  cluster view, it relies on another service called
  quota-aggregator. Aggregator, on query can return the size of a
  directory based on the cluster view.

  Enforcer is always loaded in the server graph and is by passed if
  the feature is not enabled.

  Options specific to enforcer:

  server-quota - Specifies whether the feature is on/off. It is used
  to by pass the quota if turned off.

  deem-statfs - If set to on, it takes quota limits into consideration
  while estimating fs size. (df command). The algorithm followed is,
  i.   Adjust statvfs based on limit configured on root.
  ii.  If limit is set on the inode passed, use size/limits on that inode to
       populate statvfs. Otherwise, use size/limits configured on root.
  iii. Upon statvfs, update the ctx->size on the inode.
  iv.  Don't let DHT aggregate, instead take the maximum of the usages from the
       subvols of the DHT, since each of it contains the complete information.

  Enforcer also makes use of gfid-to-path conversion functionality to
  work correctly when a client like nfs predominently relies on
  nameless lookups.

* Quota Aggregator acts as a thin client to provide cluster view

  Its a lightweight *gluster client* process with no mount point,
  started upon enabling quota or restarting the volume. This is a
  single process run on each brick, which can answer queries on all
  volumes in the cluster. Its volfile stored in
  GLUSTERD_DEFAULT_WORKING_DIR/quotad/quotad.vol.

Credits:
Raghavendra Bhat        
Varun Shastry           
Shishir Gowda           
Kruthika Dhananjay      
Brian Foster            
Krishnan Parthasarathi  

Change-Id: Id1cb25b414951da34c665a55f77385d482e0f9de
BUG: 969461
Signed-off-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/5952
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

posix: placeholders for GFID to path conversion

2013-11-26T18:22:40+00:00

what?
=====
    The following is an attempt to generate the paths of a file when
    only its gfid is known.

    To find the path of a directory, the symlink handle to the
    directory maintained in  the ".glusterfs" backend directory is
    read. The symlink handle is generated using the gfid of the
    directory. It (handle) contains the directory's name and parent
    gfid, which are used to recursively construct the absolute path as
    seen by the user from the mount point.

    A similar approach cannot be used for a regular file or a symbolic
    link since its hardlink handle, generated using its gfid, doesn't
    contain its parent gfid and basename. So xattrs are set to store
    the parent gfids and the number of hardlinks to a file or a
    symlink having the same parent gfid.  When an user/application
    requests for the paths of a regular file or a symlink with
    multiple hardlinks, using the parent gfids stored in the xattrs,
    the paths of the parent directories are generated as mentioned
    earlier. The base names of the hardlinks (with the same parent
    gfid) are determined by matching the actual backend inode numbers
    of each entry in the parent directory with that of the hardlink
    handle.

    Xattr is set on a regular file, link, and symbolic link as
    follows, Xattr name : trusted.pgfid. Xattr value :
    

    If a regular file, hard link, symbolic link is created then an
    xattr in the above format is set in the backend.

how to use?
===========
    This functionality can be used through getxattr interface. Two
    keys - glusterfs.ancestry.dentry and glusterfs.ancestry.path - enable
    usage of this functionality. A successful getxattr will have the
    result stored under same keys. Values will be,

    glusterfs.ancestry.dentry:
    --------------------------
    A linked list of gf-dirent structures for all possible paths from
    root to this gfid. If there are multiple paths, the linked-list
    will be a series of paths one after another. Each path will be a
    series of dentries representing all components of the path. This
    key is primarily for internal usage within glusterfs.

    glusterfs.ancestry.path:
    ------------------------
    A string containing all possible paths from root to this gfid.
    Multiple hardlinks of a file or a symlink are displayed as a colon
    seperated list (this could interfere with path components
    containing ':').

    e.g. If there is a file "file1" in root directory with two hardlinks,
         "/dir2/link2tofile1" and "/dir1/link1tofile1", then

         [root@alpha gfsmntpt]# getfattr -n glusterfs.ancestry.path -e text
          file1
          glusterfs.ancestry.path="/file1:/dir2/link2tofile1:/dir1/link1tofile1"

    Thanks Amar, Avati and Venky for the inputs.

Original Author: Ramana Raja 
BUG: 990028
Signed-off-by: Raghavendra G 
Change-Id: I0eaa9101e333e0c1f66ccefd9e95944dd4a27497
Reviewed-on: http://review.gluster.org/5951
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

mount/fuse: handle --gid-timeout=0 properly

2013-11-20T22:42:29+00:00

Fix the bug which was using the timeout value as a flag to indicate
if it was set (and hence would fail when timeout=0 would evaluate
as False)

Change-Id: Ie9a8f28d35603458cdac26c9a4e0343e7eda7344
BUG: 1032438
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/6308
Reviewed-by: Raghavendra Talur 
Reviewed-by: Harshavardhana 
Tested-by: Gluster Build System

bd: Add BD support to other xlators

2013-11-13T19:38:55+00:00

Make changes to distributed xlator to work with BD xlator. Unlike files,
a block device can't be removed when its opened. So some part of the
code were moved down to avoid this situation. Also before truncating a
BD file its BD_XATTR should be set otherwise truncate will result in
truncating posix file. So file is created with needed BD_XATTR and
truncate is invoked. Also enables BD xlator in stripe volume type.

Change-Id: If127516e261fac5fc5b137e7fe33e100bc92acc0
BUG: 1028672
Signed-off-by: M. Mohan Kumar 
Reviewed-on: http://review.gluster.org/5235
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati