glusterfs.git/xlators, branch v4.1.0rc0 protocol/client: Don't send fops till SETVOLUME is complete 2018-05-31T20:29:45+00:00 Raghavendra G rgowdapp@redhat.com 2018-05-30T04:04:44+00:00 4f591e873c0a38b662910f8e8a5768fd3e9bbeaa An earlier commit set conf->connected just after rpc layer sends RPC_CLNT_CONNECT event. However, success of socket level connection connection doesn't indicate brick stack is ready to receive fops, as an handshake has to be done b/w client and server after RPC_CLNT_CONNECT event. Any fop sent to brick in the window between, * protocol/client receiving RPC_CLNT_CONNECT event * protocol/client receiving a successful setvolume response can end up accessing an uninitialized brick stack. So, set conf->connected only after a successful SETVOLUME. Change-Id: I139a03d2da6b0d95a0d68391fcf54b00e749decf fixes: bz#1584633 Signed-off-by: Raghavendra G <rgowdapp@redhat.com> (cherry picked from commit 430e4872db43447c8a15b638760d23c188257139) 
An earlier commit set conf->connected just after rpc layer sends
RPC_CLNT_CONNECT event. However, success of socket level connection
connection doesn't indicate brick stack is ready to receive fops, as
an handshake has to be done b/w client and server after
RPC_CLNT_CONNECT event. Any fop sent to brick in the window between,
* protocol/client receiving RPC_CLNT_CONNECT event
* protocol/client receiving a successful setvolume response

can end up accessing an uninitialized brick stack. So, set
conf->connected only after a successful SETVOLUME.

Change-Id: I139a03d2da6b0d95a0d68391fcf54b00e749decf
fixes: bz#1584633
Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
(cherry picked from commit 430e4872db43447c8a15b638760d23c188257139)
cluster/dht: Fix rebalance log msg 2018-05-31T13:13:39+00:00 N Balachandran nbalacha@redhat.com 2018-05-21T05:34:45+00:00 67086dbd9e8fbe8ff21ba3cccdf21530ba3f80ff Corrected the name of the xattr and fixed the code to log an error only if op_errno is not ENODATA or ENOATTR. Change-Id: I42c5b1d838eec586ac7bed2471eb1d27ff09a9ea fixes: bz#1583769 Signed-off-by: N Balachandran <nbalacha@redhat.com> 
Corrected the name of the xattr and fixed
the code to log an error only if op_errno
is not ENODATA or ENOATTR.

Change-Id: I42c5b1d838eec586ac7bed2471eb1d27ff09a9ea
fixes: bz#1583769
Signed-off-by: N Balachandran <nbalacha@redhat.com>
posix/ctime: Fix updating mtime to older time 2018-05-25T13:52:47+00:00 Kotresh HR khiremat@redhat.com 2018-05-22T04:34:48+00:00 e36d5c6b8c1c3800ef89a0251854bd2b7e2af924 With ctime feature enabled, the mtime is not updated when it's set to time older than the existing one. Fixed the same. But the ctime is not allowed to change to older dates. Backport of: > Patch: https://review.gluster.org/#/c/20055/ > BUG: 1581035 > Change-Id: If520922df42d6ce084c8df3046c138f8367164e5 (cherry picked from commit e9e3699456e738635685c9f42d1c4206c6177510) fixes: bz#1582531 Change-Id: If520922df42d6ce084c8df3046c138f8367164e5 Signed-off-by: Kotresh HR <khiremat@redhat.com> (cherry picked from commit e9e3699456e738635685c9f42d1c4206c6177510) 
With ctime feature enabled, the mtime is not
updated when it's set to time older than the
existing one. Fixed the same. But the ctime
is not allowed to change to older dates.

Backport of:
 > Patch: https://review.gluster.org/#/c/20055/
 > BUG: 1581035
 > Change-Id: If520922df42d6ce084c8df3046c138f8367164e5
(cherry picked from commit e9e3699456e738635685c9f42d1c4206c6177510)

fixes: bz#1582531
Change-Id: If520922df42d6ce084c8df3046c138f8367164e5
Signed-off-by: Kotresh HR <khiremat@redhat.com>
(cherry picked from commit e9e3699456e738635685c9f42d1c4206c6177510)
afr: fix bug-1363721.t failure 2018-05-25T12:57:45+00:00 Ravishankar N ravishankar@redhat.com 2018-05-18T10:08:29+00:00 c6f21697a52200bbbf9af05dff2b9c3ae0a99b43 Problem: In the .t, when the only good brick was brought down, writes on the fd were still succeeding on the bad bricks. The inflight split-brain check was marking the write as failure but since the write succeeded on all the bad bricks, afr_txn_nothing_failed() was set to true and we were unwinding writev with success to DHT and then catching the failure in post-op in the background. Fix: Don't wind the FOP phase if the write_subvol (which is populated with readable subvols obtained in pre-op cbk) does not have at least 1 good brick which was up when the transaction started. Note: This fix is not related to brick muliplexing. I ran the .t 10 times with this fix and brick-mux enabled without any failures. Change-Id: I915c9c366aa32cd342b1565827ca2d83cb02ae85 updates: bz#1581548 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 985a1d15db910e012ddc1dcdc2e333cc28a9968b) 
Problem:
In the .t, when the only good brick was brought down, writes on the fd were
still succeeding on the bad bricks. The inflight split-brain check was
marking the write as failure but since the write succeeded on all the
bad bricks, afr_txn_nothing_failed() was set to true and we were
unwinding writev with success to DHT and then catching the failure in
post-op in the background.

Fix:
Don't wind the FOP phase if the write_subvol (which is populated with readable
subvols obtained in pre-op cbk) does not have at least 1 good brick which was up
when the transaction started.

Note: This fix is not related to brick muliplexing. I ran the .t
10 times with this fix and brick-mux enabled without any failures.

Change-Id: I915c9c366aa32cd342b1565827ca2d83cb02ae85
updates: bz#1581548
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
(cherry picked from commit 985a1d15db910e012ddc1dcdc2e333cc28a9968b)
cluster/ec: Fix pre-op xattrop management 2018-05-25T02:06:11+00:00 Xavi Hernandez xhernandez@redhat.com 2018-05-15T09:37:16+00:00 16b3830c337acacc1155c6d463c3ced852603e12 Multiple pre-op xattrop can be simultaneously being processed. On the cbk it was checked if the fop was waiting for some specific data (like size and version) and, if so, it was assumed that this answer should contain that data. This is not true, since a fop can be waiting for some data, but it may come from the xattrop of another fop. This patch differentiates between needing some information and providing it. This is related to parallel writes. Disabling them fixed the problem, but also prevented concurrent reads. A change has been made so that disabling parallel writes still allows parallel reads. Backport of: > BUG: 1578325 Fixes: bz#1582056 Change-Id: I74772ad6b80b7b37805da93d5ec3ae099e96b041 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> 
Multiple pre-op xattrop can be simultaneously being processed. On the cbk
it was checked if the fop was waiting for some specific data (like size and
version) and, if so, it was assumed that this answer should contain that
data.

This is not true, since a fop can be waiting for some data, but it may come
from the xattrop of another fop.

This patch differentiates between needing some information and providing it.

This is related to parallel writes. Disabling them fixed the problem, but
also prevented concurrent reads. A change has been made so that disabling
parallel writes still allows parallel reads.

Backport of:
> BUG: 1578325

Fixes: bz#1582056
Change-Id: I74772ad6b80b7b37805da93d5ec3ae099e96b041
Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
storage/posix: use proper FOP for unwinding readdir(p) 2018-05-25T02:06:06+00:00 Raghavendra Bhat raghavendra@redhat.com 2018-05-21T18:51:44+00:00 d8b7ec6b4176981cf8b4fe4fae4da2ca5fca4bdf As of now, even for readdirp, posix is unwinding with readdir signature. Change-Id: I6440c8a253c5d78bbcc97043e4e6e208e3d47cd1 fixes: bz#1582199 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> (cherry picked from commit de9b8c06b89b6129154693f048985554ccc5a7e7) 
As of now, even for readdirp, posix is unwinding with readdir
signature.

Change-Id: I6440c8a253c5d78bbcc97043e4e6e208e3d47cd1
fixes: bz#1582199
Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
(cherry picked from commit de9b8c06b89b6129154693f048985554ccc5a7e7)
Revert "glusterfsd: Memleak in glusterfsd process while brick mux is on" 2018-05-25T02:05:37+00:00 Mohit Agrawal moagrawa@redhat.com 2018-05-23T03:40:11+00:00 b679fd4b73d9ec039029088769722887b61d750a Updates: bz#1582286 This reverts commit 7c3cc485054e4ede1efb358552135b432fb7047a. Change-Id: I831d646112bcfa13d0c2153482ad00ff1b23aa6c Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> 
Updates: bz#1582286
This reverts commit 7c3cc485054e4ede1efb358552135b432fb7047a.
Change-Id: I831d646112bcfa13d0c2153482ad00ff1b23aa6c
Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
Revert "gluster: Sometimes Brick process is crashed at the time of stopping brick" 2018-05-25T02:05:37+00:00 Mohit Agrawal moagrawa@redhat.com 2018-05-23T03:36:04+00:00 7b95d5a4b3988757bf8c91f82dcaf86ed3da6875 Updates: bz#1582286 This reverts commit 0043c63f70776444f69667a4ef9596217ecb42b7. Change-Id: Iab3b4f4a54e122c589e515add93c6effc966b3e0 
Updates: bz#1582286
This reverts commit 0043c63f70776444f69667a4ef9596217ecb42b7.
Change-Id: Iab3b4f4a54e122c589e515add93c6effc966b3e0
Revert "server: fix unresolved symbols by moving them to libglusterfs" 2018-05-25T02:05:37+00:00 Mohit Agrawal moagrawa@redhat.com 2018-05-23T03:34:41+00:00 57dd3692d1a10d446db7fe919497335984e2cd3f Updates: bz#1582286 This reverts commit 408a6d07ababde234ddeafe16687aacd2b810b42. Change-Id: If8247d7980d698141f47130a3c532b942408ec2b 
Updates: bz#1582286
This reverts commit 408a6d07ababde234ddeafe16687aacd2b810b42.
Change-Id: If8247d7980d698141f47130a3c532b942408ec2b
posix/ctime: Fix gfid heal on first lookup 2018-05-24T07:42:39+00:00 Kotresh HR khiremat@redhat.com 2018-05-21T16:57:45+00:00 833808f42247bcb8db1be917f1ffd7841d9e226f With ctime feature enabled, the gfid is not healing on first lookup. The fresh file logic depends on ctime and it was fetching from backend instead of xattr with ctime feature enabled. Fixed the same. Also fixed a possible hang with inode lock Backport of: > Patch: https://review.gluster.org/20052 > BUG: 1580532 > Change-Id: I020875c0462b284d6fa0e68304a422fa3d6a3e73 > Signed-off-by: Kotresh HR <khiremat@redhat.com> fixes: bz#1582080 Change-Id: I020875c0462b284d6fa0e68304a422fa3d6a3e73 Signed-off-by: Kotresh HR <khiremat@redhat.com> 
With ctime feature enabled, the gfid is
not healing on first lookup. The fresh
file logic depends on ctime and it was
fetching from backend instead of xattr
with ctime feature enabled. Fixed the
same.

Also fixed a possible hang with inode lock

Backport of:
 > Patch: https://review.gluster.org/20052
 > BUG: 1580532
 > Change-Id: I020875c0462b284d6fa0e68304a422fa3d6a3e73
 > Signed-off-by: Kotresh HR <khiremat@redhat.com>

fixes: bz#1582080
Change-Id: I020875c0462b284d6fa0e68304a422fa3d6a3e73
Signed-off-by: Kotresh HR <khiremat@redhat.com>