glusterfs.git/xlators, branch release-3.5

glusterd: fix glusterd_add_inode_size_to_dict to avoid glusterd to crash

2016-03-11T17:00:05+00:00

Backport of
        http://review.gluster.org/#/c/8134/
        http://review.gluster.org/#/c/8492/

There were couple of backports from mainline got missed and due to which
glusterd crashes if the underlying file system doesn't fail under list of
supported file systems. This patch takes care of handling this negative
scenario.

Reported-by: Michael Martel 
Change-Id: I6f601a4421869bbd7fc26e31f4ca4ffe075c0924
BUG: 1316116
Signed-off-by: Atin Mukherjee 
Reviewed-on: http://review.gluster.org/13661
NetBSD-regression: NetBSD Build System 
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
Reviewed-by: Niels de Vos

quota : avoid "contribution node is empty" error logs

2016-02-12T08:28:36+00:00

In versions older than 3.7, "contribution node list is empty which is
an error" gets logged numerous number of times. It is completely fixed
in 3.7. Since the entire marker and quota code has almost been
refactored, it is hard to backport the complete fix and also it could
not be fixed with the older approach. As a temporary fix, to avoid
numerous logs, the patch just supresses the log level.

3.6 Patch -> http://review.gluster.org/#/c/13304/

Change-Id: Ie666ba99c7bb16b9ce249b581e09857734589f51
BUG: 1288195
Signed-off-by: Manikandan Selvaganesh 
Reviewed-on: http://review.gluster.org/12990
Smoke: Gluster Build System 
CentOS-regression: Gluster Build System 
NetBSD-regression: NetBSD Build System 
Reviewed-by: Niels de Vos 
Reviewed-by: Vijaikumar Mallikarjuna

protocol/client: prevent use-after-free of frame->root

2015-12-12T10:01:34+00:00

A regression failure generated a coredump on the glusterfs-client side:

  (gdb) f 0
  #0  0x00007fba6cd76432 in client_submit_request (this=0x7fba68006fc0,
                  req=0x7fba6579aa70, frame=0x7fba5c0058cc,
                  prog=0x7fba6cfb53c0 , procnum=41,
                  cbkfn=0x7fba6cd9206d ,
                  iobref=0x0, rsphdr=0x0, rsphdr_count=0,
                  rsp_payload=0x0, rsp_payload_count=0, rsp_iobref=0x0,
                  xdrproc=0x7fba79801075 ) at
  /home/jenkins/root/workspace/rackspace-regression-2GB-triggered/xlators/protocol/client/src/client.c:324
  324                   frame->root->ngrps = ngroups;
  (gdb) l
  319                   gf_msg_debug (this->name, 0, "rpc_clnt_submit failed");
  320           }
  321
  322           if (!conf->send_gids) {
  323                   /* restore previous values */
  324                   frame->root->ngrps = ngroups;
  325                   if (ngroups <= SMALL_GROUP_COUNT)
  326                           frame->root->groups_small[0] = gid;
  327           }
  328
  (gdb) p *frame->root
  Cannot access memory at address 0x64185df000000000

After looking at this in more detail, the flow is like this:

  client_submit_request()
    |
    '- rpc_clnt_submit() // on line 314
         |
         '- cbkfn() // = client3_3_release_cbk
              |
              :- STACK_DESTROY (frame->root);
         .----'
    .----'
    |
    :- frame->root->ngrps = ngroups; // on line 324
    '

So, there is a use-after-free, and it is not needed to restore the
previous groups in frame->root.

Cherry picked from commit dc3aa7524e4974f9d02465e2e5dd6ed9b6d319e1:
> Change-Id: I9e7d712183692ed92cfc2f75cd3c2781a9db20e2
> BUG: 1281285 (was incorrect in original patch)
> Signed-off-by: Niels de Vos 
> Reviewed-on: http://review.gluster.org/12575
> Reviewed-by: Dan Lambright 
> Tested-by: NetBSD Build System 
> Reviewed-by: Jeff Darcy 

Change-Id: I9e7d712183692ed92cfc2f75cd3c2781a9db20e2
BUG: 1283691
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/12666
Tested-by: Gluster Build System

build: drop incorrect usage of 'inline'

2015-07-23T14:33:03+00:00

Newer versions of gcc complain loudly about the incorrect usage of
'inline'.

These changes have been part of commit a3cb38e3 in other branches. There
is no need to backport everything from that commit, but silencing the
warnings would be good.

Change-Id: I2319d9682b47e8533e9c179056a5152f75aeddc3
BUG: 1244147
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/11711
Tested-by: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY

nfs: a unix-domain-socket should not be created as fifo

2015-07-23T12:59:20+00:00

Cherry picked from commit 381afd4f480533f212e1f62a0ccfd9dcf7248e17:
> Change-Id: Ic6a23165df1703b330636a059967c3c674dbde57
> BUG: 1235231
> Signed-off-by: Niels de Vos 
> Reviewed-on: http://review.gluster.org/11355
> Tested-by: NetBSD Build System 
> Reviewed-by: Kaleb KEITHLEY 
> Tested-by: Gluster Build System 

Change-Id: Ic6a23165df1703b330636a059967c3c674dbde57
BUG: 1244118
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/11710
Tested-by: Gluster Build System 
Reviewed-by: Kaleb KEITHLEY

nfs: make it possible to disable nfs.mount-rmtab

2015-07-07T16:11:58+00:00

When there are many NFS-clients doing very often mount/unmount actions,
the updating of the 'rmtab' can become a bottleneck and cause delays. In
these situations, the output of 'showmount' may be less important than
the responsiveness of the (un)mounting.

By setting 'nfs.mount-rmtab' to the value "/-", the cache file is not
updated anymore, and the entries are only kept in memory.

Cherry picked from commit 40407afb529f6e5fa2f79e9778c2f527122d75eb:
> Cherry picked from commit 331ef6e1a86bfc0a93f8a9dec6ad35c417873849:
>> BUG: 1169317
>> Change-Id: I40c4d8d754932f86fb2b1b2588843390464c773d
>> Reported-by: Cyril Peponnet 
>> Signed-off-by: Niels de Vos 
>> Reviewed-on: http://review.gluster.org/9223
>> Tested-by: Gluster Build System 
>> Reviewed-by: soumya k 
>> Reviewed-by: jiffin tony Thottan 
>> Reviewed-by: Kaleb KEITHLEY 
>
> This change also contains the fixes to the test-case from:
>>
>> nfs: fix spurious failure in bug-1166862.t
>>
>> In some environments, "showmount" could return an NFS-client that does
>> not start with "1". This would cause the test-case to fail. The check is
>> incorrect, the number of lines should get counted instead.
>>
>> Also moving the test-case to the .../nfs/... subdirectory.
>>
>> Cherry picked from commit ee9b35a780607daddc2832b9af5ed6bf414aebc0:
>> BUG: 1166862
>> Change-Id: Ic03aa8145ca57d78aea01564466e924b03bb302a
>> Signed-off-by: Niels de Vos 
>> Reviewed-on: http://review.gluster.org/10419
>> Tested-by: Gluster Build System 
>> Reviewed-by: Vijay Bellur 
>>
>
> Change-Id: I40c4d8d754932f86fb2b1b2588843390464c773d
> BUG: 1215385
> Signed-off-by: Niels de Vos 
> Reviewed-on: http://review.gluster.org/10379
> Tested-by: NetBSD Build System
> Tested-by: Gluster Build System 
> Reviewed-by: Vijay Bellur 

GLUSTERD_WORKDIR has been added to tests/include.rc and is not
configurable through ./configure like on newer branches. It is not
suitable to change the GlusterD working directory in an update for a
stable release.

Change-Id: I40c4d8d754932f86fb2b1b2588843390464c773d
BUG: 1166862
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/11336
Tested-by: Gluster Build System 
Reviewed-by: jiffin tony Thottan 
Reviewed-by: Kaleb KEITHLEY

quota: Validate NULL inode from the entries received in readdirp_cbk

2015-06-28T19:52:19+00:00

This is a backport of http://review.gluster.org/#/c/10416/

> In quota readdirp_cbk, inode ctx filled for the all entries
> received.
> In marker readdirp_cbk, files/directories are inspected for
> dirty

> There is no guarantee that entry->inode is populated.
> If entry->inode is NULL, this needs to be treated as readdir
>
> Change-Id: Id2d17bb89e4770845ce1f13d73abc2b3c5826c06
> BUG: 1215550
> Signed-off-by: vmallika 
> Reviewed-on: http://review.gluster.org/10416
> Tested-by: NetBSD Build System
> Tested-by: Gluster Build System 
> Reviewed-by: Raghavendra G 
> Tested-by: Raghavendra G 

Change-Id: If6578984ab1a33a3bdbaf0ec169f7c9548991144
BUG: 1217433
Signed-off-by: vmallika 
Reviewed-on: http://review.gluster.org/10472
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

posix: handle failure from posix_resolve

2015-06-28T19:51:01+00:00

This is backport of http://review.gluster.org/#/c/9941/

> When building ancestory, posix_resolve gets the inode
> from the gfid. We need to handle the failure case from
> this function
>
> Change-Id: I19f0f0c739686b1b0ef96309212aa1c7911b3589
> BUG: 1203629
> Signed-off-by: vmallika 
> Reviewed-on: http://review.gluster.org/9941
> Reviewed-by: Raghavendra Bhat 
> Tested-by: Gluster Build System 
> Reviewed-by: Pranith Kumar Karampuri 
> Tested-by: Pranith Kumar Karampuri 

Change-Id: If3f9a910be3d08293199748c9e04b3be2d8633c7
BUG: 1217432
Signed-off-by: vmallika 
Reviewed-on: http://review.gluster.org/10471
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

fuse: squash 64-bit inodes in readdirp when enable-ino32 is set

2015-06-03T20:23:55+00:00

The structures returned by readdirp contain the inode 2x. Only one of
them was squashed into 32-bits when enable-ino32 is enabled.

Backport of:
> Change-Id: I33a6d28fb118bb23971f918ffeb983d7f033106e
> BUG: 1223889
> Reviewed-on: http://review.gluster.org/10881
> Signed-off-by: Niels de Vos 
> Tested-by: Cyril Peponnet  [on release-3.5]

Change-Id: I33a6d28fb118bb23971f918ffeb983d7f033106e
BUG: 1222150
Signed-off-by: Niels de Vos 
Tested-by: Cyril Peponnet 
Reviewed-on: http://review.gluster.org/10884
Tested-by: Gluster Build System

cluster/afr: Make read child match check in afr optional

2015-03-18T11:21:28+00:00

        Backport of: http://review.gluster.org/#/c/9917

Having this particular check which was introduced by
commit bb2df4e63fa8a5d65f18b4a5efc757e8d475fbff causes a drop in
performance in readdirp. So the behavior is made configurable with this
patch.

Change-Id: I9012a6bb955229a0cbb48f06e4e2edc0782dfead
BUG: 1202675
Signed-off-by: Krutika Dhananjay 
Reviewed-on: http://review.gluster.org/9924
Reviewed-by: Pranith Kumar Karampuri 
Reviewed-by: Atin Mukherjee 
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos