glusterfs.git/api, branch v3.7.15

gfapi: Fix IO error caused when there is consecutive graph switches

2016-08-25T04:42:50+00:00

Backport of http://review.gluster.org/#/c/14722/

This is part 2 of the fix, the part 1 can be found at:
http://review.gluster.org/#/c/14656/

Problem:
=======
Consider a race between, __glfs_active_subvol() and graph_setup().
Lets say @TIME T1:
fs->active_subvol = A
fs->next_subvol = B
__glfs_active_subvol()                //under lock fs->mutex
{
  ....
  new_subvol = fs->next_subvol       //which is B
  ....                               //Start migration from A to B
  __glfs_first_lookup(){
     ....
     unlock fs->mutex                //@TIME T2
     network fop
     lock fs->mutex
     ....
  }
  ....                                //migration continue on B
  fs->active_subvol = fs->next_subvol //which is C (explained below)
  ....
}

@Time T2, lets say in another thread, graph_setup() is called with C,
note that at T2, fs->mutex is unlocked.

graph_stup(C...)
{
  lock fs->mutex
  ....
  if (fs->next_subvol)                // which is B
      destroy subvol (fs->next_subvol)
  ....
  fs->next_subvol = C
  ....
  unlock fs->mutex
}

Thus at the end of this,
fs->old_subvol = A;
fs->active_subvol = C;
fs->next_subvol = NULL;
which is wrong, as B completed migration, but was destroyed by
graph_setup, and C never was migrated.

Solution:
=========
Any new graph can be in one of the 2 states:
- Picked for migration, migration in progress (fs->mip_subvol)
- Not picked so far for migration (fs->next_subvol)
graph_setup() updates fs->next_subvol only, __glfs_active_subvol()
moves fs->next_subvol to fs->mip_subvol and fs->next_subvol = NULL
atomically, and then once the migration is complete, make that the
fs->active_subvol

> Reviewed-on: http://review.gluster.org/14722
> Smoke: Gluster Build System 
> CentOS-regression: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> Reviewed-by: Raghavendra Talur 
> Reviewed-by: Rajesh Joseph 
> Reviewed-by: Niels de Vos 

BUG: 1367294
Change-Id: Ib6ff0565105c5eedb912a43da4017cd413243612
Signed-off-by: Poornima G 
Signed-off-by: Oleksandr Natalenko 
Reviewed-on: http://review.gluster.org/15167
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Kaushal M

gfapi: Fix IO error caused when there is consecutive graph switches

2016-08-25T04:38:22+00:00

Issue:
Consider a simple situation, where glfs_init() is done, i.e. initial
graph is up. Now perform 2 volume sets that results in 2 client side
graph changes. After this perform some IO, the IO fails with ENOTCON.
The only way to recover this client is i guess another graph switch
or restart.

What actually is happening from code perspective:
Initial graph lets say A, followed by 2 consecutive graph switches
to B and C without any IO those two switches.

- graph_setup (A) as a result of GF_EVENT_CHILD_UP, and
fs->next_subvol = A

- glfs_init() results in fs->active_subvol = A, fs->next_subvol = NULL

- graph_setup (B) as a result of GF_EVENT_CHILD_UP, and
fs->next_subvol = B

- graph_setup (C) as a result of GF_EVENT_CHILD_UP, and
fs->next_subvol = C. It also sees that the previous graph B was never
set as fs->active_subvol, i.e. no IO or anything happened on B, so
can safely send GF_EVENT_PARENT_DOWN (by calling glfs_subvol_done(B)).
This parent down on B, results in child_down(B), which is fine.
But child_down also triggers graph_setup(B).

- graph_setup(B) as a result of GF_EVENT_CHILD_DOWN, and
fs->next_subvol = B, and GF_EVENT_PARENT_DOWN on C as explained
above. This again leads to GF_EVENT_CHILD_DOWN on C.

- graph_setup(C) as a result of GF_EVENT_CHILD_DOWN, and
fs->next_subvol = C, and GF_EVENT_PARENT_DOWN on B as explained
above.

Thus both the graphs B and C are disconnected, and hence the ENOTCON

Solution:
Remove the call to graph_setup() when the event is GF_EVENT_CHILD_DOWN.
It don't see any reason why graph_setup should be called when there is
child_down. Not sure what the original reason was, to have graph_setup
in child_down. git hostory shows the first patch itself had this call.

> Reviewed-on: http://review.gluster.org/14656
> Smoke: Gluster Build System 
> CentOS-regression: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> Reviewed-by: Jeff Darcy 

BUG: 1367294
Change-Id: I9de86555f66cc94a05649ac863b40ed3426ffd4b
Signed-off-by: Poornima G 
Signed-off-by: Oleksandr Natalenko 
Reviewed-on: http://review.gluster.org/14835
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Kaushal M

gfapi: use const qualifier for glfs_*timens()

2016-08-12T11:20:04+00:00

glfs_*timens() functions have the last argument
of struct timespec type that is supposed to be
const. Now using glfs_*timens() in fops, implemented
on top of FUSE library, leads to compiler-time warning
about discarding const qualifier.

Introducing const qualifier does not break ABI,
so let's just fix it.

> Reviewed-on: http://review.gluster.org/15119
> Smoke: Gluster Build System 
> CentOS-regression: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> Reviewed-by: Niels de Vos 

BUG: 1365869
Change-Id: Iea2a1018de7dce67f67a8229671a5978246de800
Signed-off-by: Oleksandr Natalenko 
Reviewed-on: http://review.gluster.org/15137
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY 
Reviewed-by: Prashanth Pai

gfapi: add missing glfs_truncate

2016-08-12T11:19:31+00:00

> Reviewed-on: http://review.gluster.org/13927
> Smoke: Gluster Build System 
> CentOS-regression: Gluster Build System 
> NetBSD-regression: NetBSD Build System 
> Reviewed-by: Kaleb KEITHLEY 

BUG: 1366286
Change-Id: I80b016090a4d9d86278a0a5144dd58c0cbfe9bb2
Signed-off-by: Jeff Darcy 
Signed-off-by: Oleksandr Natalenko 
Reviewed-on: http://review.gluster.org/15150
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Kaleb KEITHLEY 
CentOS-regression: Gluster Build System 
Reviewed-by: Prashanth Pai

gfapi : Avoid double freeing of dict in glfs__getxattr

2016-08-03T11:24:56+00:00

The dict variable "xattr" is passed to glfs_getxattr_process() and
glfs_listxattr_process() in glfs_*_*getxattrs(). This variable is
unrefed by both functions and again in caller function which may
result in segfault. So it is wrong to call dict_unref() in both
glfs_*xattr_process functions.

Backport reference :
>Change-Id: I227f55ebc3169f58910863c04ae536a8d789e80e
>BUG: 1247603 Signed-off-by: Jiffin Tony Thottan 
>Reviewed-on: http://review.gluster.org/13483
>Smoke: Gluster Build System 
>NetBSD-regression: NetBSD Build System 
>CentOS-regression: Gluster Build System 
>Reviewed-by: Niels de Vos 

Change-Id: I2e574ff4b7a095749540bdb9d3593bc1d6275e56
BUG: 1311407
Signed-off-by: Jiffin Tony Thottan 
Signed-off-by: Oleksandr Natalenko 
Reviewed-on: http://review.gluster.org/13505
Reviewed-by: Prashanth Pai 
NetBSD-regression: NetBSD Build System 
Smoke: Gluster Build System 
Reviewed-by: Niels de Vos 
CentOS-regression: Gluster Build System

gfapi/upcall: Fix a ref leak

2016-08-03T09:31:11+00:00

inode_find (used to create the handle) takes a reference
of the inode. This needs to be un'refernced to avoid leak.

This is backport of below master patch -
   http://review.gluster.org/14984

>Change-Id: I22f03577a8f1d9608cfc62d57202cfc4c2ba12b3
>BUG: 1358608
>Signed-off-by: Soumya Koduri 
>Reviewed-on: http://review.gluster.org/14984
>Reviewed-by: jiffin tony Thottan 
>Reviewed-by: Jeff Darcy 
>(cherry picked from commit bb48eb46910085928efbd7fb491c5b2db25bba98)

Change-Id: Iaa91ee757e3497e25d8669c1592106b6266057a0
BUG: 1362010
Signed-off-by: Soumya Koduri 
Reviewed-on: http://review.gluster.org/15057
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Tested-by: Oleksandr Natalenko 
CentOS-regression: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY 
Reviewed-by: Niels de Vos

glfs/upcall: entries should be removed under mutex lock

2016-08-02T09:18:54+00:00

During poll, upcall entries should be removed from the
upcall_list only under upcall_list_mutex lock. Otherwise
it could result in the list corruption if there are entries
being added during poll resulting in memory leak.

Also addressed a probable leak during any failures with upcall
entry addition.

This is backport of below master patch -
   http://review.gluster.org/14972

>Change-Id: I468183f961eb6faed9a0a1bcb783705f711641fc
>BUG: 1358608
>Signed-off-by: Soumya Koduri 
>Reviewed-on: http://review.gluster.org/14972
>Reviewed-by: Shyamsundar Ranganathan 
>Reviewed-by: Jeff Darcy 
>(cherry picked from commit 89dee8b46e126bc1d7541da90fa60844aa83451e)

Change-Id: Ib6702911ca1fa09a3f1a493b27195665603854d3
BUG: 1362010
Signed-off-by: Soumya Koduri 
Reviewed-on: http://review.gluster.org/15058
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Tested-by: Oleksandr Natalenko 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos

gfapi: update count when glfs_buf_copy is used

2016-07-07T09:01:01+00:00

Backport of http://review.gluster.org/#/c/14854

glfs_buf_copy collates all iovecs into a iovec with count=1. If
gio->count is not updated it will lead to dereferencing of invalid
address.

Change-Id: I7c58071d5c6515ec6fee3ab36af206fa80cf37c3
BUG: 1352482
Signed-off-by: Raghavendra Talur 
Signed-off-by: Poornima G 
Reported-By: Lindsay Mathieson 
Reported-By: Dmitry Melekhov 
Reported-By: Tom Emerson 
Reviewed-on: http://review.gluster.org/14859
Smoke: Gluster Build System 
Reviewed-by: Prashanth Pai 
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System

gfapi/handleops: Avoid using glfd during create

2016-07-03T11:16:01+00:00

To avoid leaking glfd while creating a file using handleops and
since application shall not be interested in it, use the 'fd'
object directly which can be un'refed post create.

This is backport of below mainline patch -
  http://review.gluster.org/14532

> Change-Id: I119874ffb63fb4aa18f846ba1fdbe77874b66a54
> BUG: 1339553
> Signed-off-by: Soumya Koduri 
> Signed-off-by: Jiffin Tony Thottan 
> Reviewed-on: http://review.gluster.org/14532
> NetBSD-regression: NetBSD Build System 
> Reviewed-by: Niels de Vos 
> CentOS-regression: Gluster Build System 
> Tested-by: Gluster Build System 
> Smoke: Gluster Build System 
(cherry picked from commit 763ed1017b0011934ad2414d7396c46e528ea5b3)

Change-Id: I119874ffb63fb4aa18f846ba1fdbe77874b66a54
BUG: 1351877
Signed-off-by: Soumya Koduri 
Reviewed-by: Kaleb S KEITHLEY 
Reviewed-on: http://review.gluster.org/14840
NetBSD-regression: NetBSD Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos 
Smoke: Gluster Build System

gfapi : check the value "iovec" in glfs_io_async_cbk only for read

2016-06-29T12:47:19+00:00

The glfs_io_async_cbk() is called from the cbk of all the async ops
such as write, read, fsync, ftruncate. In all other cases, expect for
read the value for "iovec" is NULL. From the code, glfs_io_async_cbk
checks the value in common routine which may end up in failures.

Thanks Joe Julian for finding issue and suggesting the fix.

Cherry picked from commit 61d72b3d91f2655b04de4ef29262f738a8cf7369:
> Change-Id: I0be0123da68f9d8fbb5d94ede2d45566a9add6a5
> BUG: 1349276
> Signed-off-by: Jiffin Tony Thottan 
> Reported-by: Joe Julian 
> Reviewed-on: http://review.gluster.org/14779
> Reviewed-by: Niels de Vos 
> Smoke: Gluster Build System 
> Tested-by: Kaleb KEITHLEY 
> NetBSD-regression: NetBSD Build System 
> CentOS-regression: Gluster Build System 
> Reviewed-by: Joe Julian 

Change-Id: I0be0123da68f9d8fbb5d94ede2d45566a9add6a5
BUG: 1350880
Reported-by: Joe Julian 
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/14822
Smoke: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Kaleb KEITHLEY 
CentOS-regression: Gluster Build System 
Reviewed-by: jiffin tony Thottan

glusterfs.git/api, branch v3.7.15

gfapi: Fix IO error caused when there is consecutive graph switches

gfapi: Fix IO error caused when there is consecutive graph switches

gfapi: use const qualifier for glfs_*timens()

gfapi: add missing glfs_truncate

gfapi : Avoid double freeing of dict in glfs_*_*getxattr

gfapi/upcall: Fix a ref leak

glfs/upcall: entries should be removed under mutex lock

gfapi: update count when glfs_buf_copy is used

gfapi/handleops: Avoid using glfd during create

gfapi : check the value "iovec" in glfs_io_async_cbk only for read

gfapi : Avoid double freeing of dict in glfs__getxattr