| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Introduce cluster.favorite-child-policy which when enabled with
[ctime|mtime|size|majority], automatically heals files that are in
split-brian.
The majority policy will not pick a source if there is no majority.
The other three policies pick the first brick with a valid reply and
non-zero ctime/mtime/size as source.
Change-Id: I3c099a0404082213860f74f2c9b4d207cfaedb76
BUG: 1328224
Original-author: Richard Wareing <rwareing@fb.com>
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/14026
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anuradha Talur <atalur@redhat.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
If a named fresh-lookup is done on an loc and the fop fails on one of the
bricks or not sent on one of the bricks, but by the time response comes to afr,
if the brick is up, 'can_interpret' will be set to false in afr_lookup_done(),
this will lead to inode-ctx for that inode to be not set, this can lead to EIO
in case of a transaction as it depends on 'readable' array to be available by
that point.
Fix:
Refresh inode for inode-write fops for the ctx to be set if it is not already
done at the time of named fresh-lookup or if the file is in split-brain where
we need to perform one more refresh before failing the fop to check if the file
is still in split-brain or not.
BUG: 1336612
Change-Id: I5c50b62c8de06129b8516039f7c252e5008c47a5
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/14368
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: I52da41dff5619492b656c2217f4716a6cdadebe0
BUG: 1269461
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/12442
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: During mount, afr waits for response from all its children before
notifying the parent xlator. In a 1x2 replica volume , if one of the nodes is
down, the mount will hang for more than a minute until child down is received
from the client xlator for that node.
Fix:
When parent up is received by afr, start a 10 second timer. In the timer call
back, if we receive a successful child up from atleast one brick, propagate the
event to the parent xlator.
Change-Id: I31e57c8802c1a03a4a5d581ee4ab82f3a9c8799d
BUG: 1054694
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/11113
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Locking schemes in afr-v1 were locking the directory/file completely during
self-heal. Newer schemes of locking don't require Full directory, file locking.
But afr-v2 still has compatibility code to work-well with older clients, where
in entry-self-heal it takes a lock on a special 256 character name which can't
be created on the fs. Similarly for data self-heal there used to be a lock on
(LLONG_MAX-2, 1). Old locking scheme requires heal info to take sh-domain locks
before examining heal-state. If it doesn't take sh-domain locks, then there is
a possibility of heal-info hanging till self-heal completes because of
compatibility locks. But the problem with heal-info taking sh-domain locks is
that if two heal-info or shd, heal-info try to inspect heal state in parallel
using trylocks on sh-domain, there is a possibility that both of them assuming
a heal is in progress. This was leading to spurious entries being shown in
heal-info.
Fix:
As long as there is afr-v1 way of locking, we can't fix this problem with
simple solutions. If we know that the cluster is running newer versions of
locking schemes, in those cases we can give accurate information in heal-info.
So introduce a new option called 'locking-scheme' which if it is 'granular'
will give correct information in heal-info. Not only that, Extra network hops
for taking compatibility locks, sh-domain locks in heal info will not be
necessary anymore. Thus it improves performance.
BUG: 1322850
Change-Id: Ia563c5f096b5922009ff0ec1c42d969d55d827a3
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/13873
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-by: Anuradha Talur <atalur@redhat.com>
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
BUG: 1221737
Change-Id: I0ed71a72f0e33bd733723e00a01cf28378c5534e
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/13755
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Smoke: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch is part two change to prevent data loss
in a replicate volume on doing a add-brick operation.
Problem: After doing add-brick, there is a chance
that self heal might happen from the newly added
brick rather than the source brick, leading to data loss.
Solution: Mark pending changelogs on afr children for
the new afr-child so that heal is performed in the
correct direction.
Change-Id: I11871e55eef3593aec874f92214a2d97da229b17
BUG: 1276203
Signed-off-by: Anuradha Talur <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/12454
Smoke: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It is better to choose local brick as source if possible to prevent
over the wire read thus saving on bandwidth. Also changed code to not
attempt data-heal if 'source' is selected as arbiter.
Change-Id: I9a328d0198422280b13a30ab99545370a301dfea
BUG: 1314150
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/13585
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Tested-by: Krutika Dhananjay <kdhananj@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If a heal is needed after inode refresh (lookup, read_txn), launch it in
the background instead of blocking the fop (that triggered refresh) until the
heal happens.
afr_replies_interpret() is modified such that the heal is
launched only if atleast one sink brick is up.
Max. no of heals that can happen in parallel is configurable via the
'background-self-heal-count' volume option. Any number greater than that
is put in a wait queue whose length is configurable via
'heal-wait-queue-leng' volume option. If the wait queue is also full,
further heals will be ignored.
Default values: background-self-heal-count=8, heal-wait-queue-leng=128
Change-Id: I1d4a52814cdfd43d90591b6d2ad7b6219937ce70
BUG: 1297172
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/13207
Smoke: Gluster Build System <jenkins@build.gluster.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
seek() is like a read(), copied the same semantics.
Change-Id: I100b741d9bfacb799df318bb081f2497c0664927
BUG: 1220173
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Reviewed-on: http://review.gluster.org/11483
Smoke: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Now heal-info does an open() on the file being examined so that
the client at some point sees open-fd count being > 1 and releases
the eager-lock so that heal-info doesn't remain blocked forever
until IO completes.
Change-Id: Icc478098e2bc7234408728b54d8185102b3540dc
BUG: 1297695
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/13326
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Things done :
1) during lookup and inode_refresh as part of read_txn,
request is sent to detect if heal is required or not.
2) If heal is required, be conservative in setting the
readdirp entry inodes to NULL, otherwise don't be.
3) Self-heal-daemon now crawls both indices/xattrop
and indices/dirty directory while healing.
Change-Id: Ic4a4da63fb7e0726eab5f341a200859b29cf7eb7
BUG: 1250803
Signed-off-by: Anuradha Talur <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/12507
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: If a file is created with zeroes ('dd', 'fallocate' etc.) when
a brick is down, the self-heal does not write the zeroes to the sink
after it comes up. Consequenty, there is a mismatch in disk-usage
amongst the bricks of the replica.
Fix: If we definitely know that the file is not sparse, then write the
zeroes to the sink even if the checksums match.
Change-Id: Ic739b3da5dbf47d99801c0e1743bb13aeb3af864
BUG: 1272460
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/12371
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
1. When winding the pre-op, transaction.pre_op[i] is set. If the pre-op fails,
transaction.failed_subvols[i] is set. If if fails on all chidren, we can
directly proceed to unlock (via afr_changelog_post_op_now) without trying
to wind the write, fail and then go to unlock.
2. 'fop_subvols' seems to be an unused variable, hence removing it.
Change-Id: I9525628daf48082e979b0093fa0478934495e61f
BUG: 1272362
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/12368
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Reviewed-by: Anuradha Talur <atalur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
replace-brick setxattr is not performed inside a synctask. This can lead
to hangs if the setxattr is executed by epoll thread, as the epoll
thread will be waiting for replies to come where as epoll thread is the
thread that needs to epoll_ctl for reading from socket and listen.
Fix:
Move replace-brick to synctask to prevent epoll thread hang.
This patch is in line with the fix performed in
http://review.gluster.org/#/c/12163/
Change-Id: I6a71038bb7819f9e98f7098b18a6cee34805868f
BUG: 1262345
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/12169
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On executing `getfattr -n replica.split-brain-status <file>` on mount,
there is a possibility that the mount hangs. To avoid this hang,
fetch the split-brain-status of a file in synctask.
Change-Id: I87b781419ffc63248f915325b845e3233143d385
BUG: 1262345
Signed-off-by: Anuradha Talur <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/12163
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
When xlators above afr do [f]xattrop when one of the bricks is down, after the
brick comes backup, the metadata is not healed because [f]xattrop is not
considered a transaction.
Fix:
Treat [f]xattrop as transaction so that changes done by xlators above afr are
marked for heal when some of the bricks were down at the time of [f]xattrop.
Change-Id: Iea180f9a456509847c3cd8d5d59a0cdc2712d334
BUG: 1248887
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/11809
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: AFR serves statfs from the brick having the least free space
available. Since the size to be allocated to the arbiter brick in a 3
way replica is supposed to be considerably lesser than the other 2
bricks, statfs will be served from this brick which is incorrect.
Fix: Don't serve statfs from the arbiter brick.
Change-Id: I5af098b9c50626f52cf3d7dbb060bf754c797f05
BUG: 1251346
Reported-by: Fredrik Brandt <fredrikb@denlillaplaneten.se>
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/11857
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For directories, block metadata FOPS.
For non-directories, block data and metadata FOPS.
Do not block entry FOPS.
Change-Id: Id7f656f4a513b9d33c457dd7f2d58028dbef8e61
BUG: 1235007
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/11371
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
| |
calculation
Change-Id: I12c1e4f67f4ec4affbe13d7daf871044a8a2a12e
BUG: 1235216
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/11373
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of including config.h in each file, and have the additional
config.h included from the compiler commandline (-include option).
When a .c file tests for a certain #define, and config.h was not
included, incorrect assumtions were made. With this change, it can not
happen again.
BUG: 1222319
Change-Id: I4f9097b8740b81ecfe8b218d52ca50361f74cb64
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Reviewed-on: http://review.gluster.org/10808
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System
Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
1) Provided setfattr command to set timeout for split-brain
choice.
2) If split-brain inspection/resolution is being done
from the mount for a file, ref the inode when
split-brain-choice is set.
This inode will be unconditionally unref-ed after timeout
seconds set by the user/default otherwise.
3) Updated the doc and testcase to reflect the changes.
Change-Id: I15c9037dee28855f21e680e7e3632e1f48dba4e1
BUG: 1209104
Signed-off-by: Anuradha <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/10134
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add logic in afr to work in conjunction with the arbiter xlator when a
replica 3 arbiter volume is created. More specifically, this patch:
* Enables full locks for afr data transaction for such volumes.
* Removes the upfront marking of pending xattrs at the time of pre-op
and defer it to post-op. (This is an arbiter independent change and is made for all afr transactions.)
* After pre-op stage, check if we can proceed with the fop stage without
ending up in split-brain by examining the changelog xattrs.
* Unwinds the fop with failure if only one source was available at the
time of pre-op and the fop happened to fail on particular source brick.
* Skips data self-heal if arbiter brick is the only source available.
* Adds the arbiter-count option to the shd graph.
This patch is a part of the arbiter logic implementation for 3 way AFR
details of which can be found at http://review.gluster.org/#/c/9656/
Change-Id: I9603db9d04de5626eb2f4d8d959ef5b46113561d
BUG: 1199985
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/10258
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Logic for adding the 'glusterd_brickinfo->group' member and using it to
find the brick positon has been taken from http://review.gluster.org/#/c/9919.
Thanks to Jeff Darcy for that.
This patch is a part of the arbiter logic implementation for 3 way AFR
details of which can be found at http://review.gluster.org/#/c/9656/
Change-Id: Idbfe4f29ee8e098e0102def8f38b32314316b188
BUG: 1199985
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/10257
Tested-by: NetBSD Build System
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com>
Tested-by: Krishnan Parthasarathi <kparthas@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Changed the implementation of marker xattr handling to take just a
function which populates important data that is different from
default 'gauge' values and subvolumes where the call needs to be
wound.
- Removed duplicate code I found while reading the code and moved it to
cluster_marker_unwind. Removed unused structure members.
- Changed dht/afr/stripe implementations to follow the new implementation
- Implemented marker xattr handling for ec.
Change-Id: Ib0c3626fe31eb7c8aae841eabb694945bf23abd4
BUG: 1200372
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/9892
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Part 2/2 patch to enable users analyze and resolve
split-brain.
This patch enables :
1) Users to inspect the files in data and metadata split-brain.
2) Resolve the split-brain.
Both using a series of setfattr commands.
Consider a volume "test" with 2 bricks.
1) To inspect a file f1:
setfattr -n replica.split-brain-choice -v test-client-0 f1
After the execution of this command, if no read_subvol
is found, reads will be served from test-client-0 (corresponding
to brick-0).
2) To resolve split-brain :
setfattr -n replica.split-brain-heal-finalize -v test-client-0 f1
Execution of this command will lead to the resolution
of data and metadata split-brain with subvol mentioned in the
command (test-client-0 here) as the source and the rest as sink.
Change-Id: Ia20f3ee5abd3119e3d54fcc599f1e55ac65fd179
BUG: 1191396
Signed-off-by: Anuradha <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/9743
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Having this particular check which was introduced by
commit c78998c39f0857ea7aacba360632c148afc54a55 causes a drop in
performance in readdirp. So the behavior is made configurable with this
patch.
Change-Id: I2858fc18b3539df7aa6d3f489e0d5cfaeb8a9b3c
BUG: 1202669
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/9917
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Provide a way of disabling reads when quorum is not met.
Change-Id: Ic4f57c2b87a0b8514600759de3a7a47e217fe3b5
BUG: 1187885
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/9543
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
These will be used by both afr and ec. Moved syncop_dirfd, syncop_ftw,
syncop_dir_scan functions also into syncop-utils.c
Change-Id: I467253c74a346e1e292d36a8c1a035775c3aa670
BUG: 1177601
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/9740
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-by: Anuradha Talur <atalur@redhat.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch is one part to enable users analyze and resolve
split-brain.
Problem : To know if a file is in data/metadata split-brain
Solution : Performing "getfattr -n afr.split-brain-status
<path-to-file>" from the mount provides this information.
Also provides the list of afr children to analyse to
get more information.
Change-Id: I4d9b429794759a906371416cb84c84a212e2c7b9
BUG: 1191396
Signed-off-by: Anuradha <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/9633
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Extend the AFR heal command to include automated split-brain resolution.
This patch [3/3] is the final patch for afr automated split-brain resolution
implementation.
"gluster volume heal <VOLNAME> [full | statistics [heal-count [replica
<HOSTNAME:BRICKNAME>]] |info [healed | heal-failed | split-brain]| split-brain
{bigger-file <FILE> |source-brick <HOSTNAME:BRICKNAME> [<FILE>]}]"
The new additions being:
1.gluster volume heal <VOLNAME> split-brain bigger-file <FILE>
Locates the replica containing the FILE, selects bigger-file as source and
completes heal.
2.gluster volume heal <VOLNAME> split-brain source-brick <HOSTNAME:BRICKNAME>
<FILE>
Selects <FILE> present in <HOSTNAME:BRICKNAME> as source and completes heal.
3.gluster volume heal <VOLNAME> split-brain <HOSTNAME:BRICKNAME>
Selects all split-brained files in <HOSTNAME:BRICKNAME> as source and completes
heal.
Note: <FILE> can be either the full file name as seen from the root of the
volume (or) the gfid-string representation of the file, which sometimes gets
displayed in the heal info command's output.
Entry/gfid split-brain resolution is not supported.
Example can be found in the test case.
Change-Id: I4649733922d406f14f28ee9033a5cb627b9538b3
BUG: 1136769
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/9377
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Afr winds inodelk calls without any order, so blocking inodelks
from two different mounts can lead to dead lock when mount1 gets
the lock on brick-1 and blocked on brick-2 where as mount2 gets
lock on brick-2 and blocked on brick-1
Fix:
Serialize the inodelks whether they are blocking inodelks or
non-blocking inodelks.
Non-blocking locks also need to be serialized.
Otherwise there is a chance that both the mounts which issued same
non-blocking inodelk may endup not acquiring the lock on any-brick.
Ex:
Mount1 and Mount2 request for full length lock on file f1. Mount1 afr may
acquire the partial lock on brick-1 and may not acquire the lock on brick-2
because Mount2 already got the lock on brick-2, vice versa. Since both the
mounts only got partial locks, afr treats them as failure in gaining the locks
and unwinds with EAGAIN errno.
Change-Id: Ie6cc3d564638ab3aad586f9a4064d81e42d52aef
BUG: 1176008
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/9372
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The purpose of encoding d_off in AFR is to indicate the
selected subvolume for the first readdir, and continue all
further readdirs of the session on the same subvolume. This is
required because, unlike files, dir d_offs are specific to the
backend and cannot be re-used on another subvolume. The d_off
transformation encodes the subvolume id and prevents such
invalid use of d_offs on other servers.
However, this approach could be quite wasteful of precious d_off
bit-space. Unlike DHT, where server id can change from entry to
entry and thus encoding the server id in the transformed d_off
is necessary, we could take a slightly relaxed approach in AFR.
The approach is to save the subvolume where the last readdir
request was sent in the fd_ctx. This consumes constant space (i.e
no per-entry cache), and serves the purpose of avoiding d_off
"misuse" (i.e using d_off from one server on another).
The compromise here is NFS resuming readdir from a non-0 cookie
after an extended delay (either anonymous FD has been reclaimed,
or server has restarted). In such cases a subvolume is picked
freshly. To make this fresh picking more deterministic (i.e, to
pick the same subvolume whenever possible, even after reboots),
the function afr_hash_child (used by afr_read_subvol_select_by_policy)
is modified to skip all dynamic inputs (i.e PID) for the case
of directories.
Change-Id: I46ad95feaeb21fb811b7e8d772866a646330c9d8
BUG: 1163161
Signed-off-by: Anand Avati <avati@redhat.com>
Reviewed-on: http://review.gluster.org/9332
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
gluster volume heal <volname> info command
will now also display if the files listed (in the output
of the command) are in split-brain or possibly being
healed.
This patch also fixes build warning that occurs.
Change-Id: I1fc92e62137f23b2b9ddf6e05819cee6230741d1
BUG: 1163804
Signed-off-by: Anuradha <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/9119
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Afr should ignore quota-size-key as part of self-heal
but should heal quota-limit key.
Change-Id: Ic0b06bd20a563a00d6bfdc2dc5a76c661e533ecb
BUG: 1161106
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/9061
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Added logging for metadata and data self-heals which helped
in debugging this issue.
- Added checks to skip self-heals when no sinks are available to heal
Change-Id: I0d50dceb84cd9ad4fe00e0b749ddf7d4ff42348a
BUG: 1128721
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/8709
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
AFR_STACK_RESET previously didn't cleanup afr_local_t,
leading to memory leaks. With this patch, cleanup is
done.
All credit goes to Pranith Kumar Karampuri.
Change-Id: I3c727ff4bb323dccb81da4b3168ac69bb340d17d
BUG: 1145471
Signed-off-by: Anuradha <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/8821
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
When one of the brick is taken down and brough back up in a replica pair, locks
on that brick will be allowed. Afr returns inodelk success even when one of the
bricks already has the lock taken.
Fix:
If any brick returns EAGAIN return failure to parent xlator.
Change-Id: I5b842d0fc094359cc4231494053d2bfeb606bbbe
BUG: 1141539
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/8710
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Detect and heal mismatching user extended attributes during lookup.
'Forward' port of http://review.gluster.org/#/c/7444/
Change-Id: Id03c9746f083ffd3014711d0b3a2e5a71a45eed4
BUG: 1134691
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/8558
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Based on type of file, set appropriate pending changelogs
for new entries.
Change-Id: Ifd124bf9bc54b996ce83ab9f39d03b3ccca7eb3c
BUG: 1130892
Signed-off-by: Anuradha <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/8555
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
dict_t objects that are ref'd in alloca'd "replies" in
afr_replies_copy() are not unref'd after "replies" go out of scope.
Change-Id: Id5a6ca3c17a8de72b94b3e0f92165609da5a36ea
BUG: 1134221
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/8553
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Counts may give wrong results when the number of bricks is > 2. If the
locks are acquired on one source and sink, but the source accuses even the
down sink then there will be 2 sinks and lock is acquired on 2 bricks so
even when there is a clear source and sink **_finalize_source functions think
the file/directory is in split-brain.
Fix:
Check that all the bricks which are locked are sinks.
Change-Id: Ia43790e8e1bfb5e72a3d0b56bcad94abd0dc58ab
BUG: 1128721
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/8456
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Allowing lookup with 'gfid-req' will lead to assigning gfid at posix layer.
When two mounts perform lookup in parallel that can lead to both bricks getting
different gfids leading to gfid-mismatch/EIO for the lookup.
Fix:
Perform gfid heal inside lock.
BUG: 1129529
Change-Id: I20c6c5e25ee27eeb906bff2f4c8ad0da18d00090
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/8512
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problems with fuse/server:
Fuse loc touch up sets loc->name even when pargfid
is not known. Server lookup does (pargfid, name) based
lookup when name is set ignoring the gfid. Because of this server
resolver finds that the lookup came on (null-pargfid, name) and
fails the lookup with EINVAL.
Fix:
Don't set loc->name in loc_touchup if the pargfid is not known.
Did the same even for server-resolver
Problem with afr:
Lets say there is a directory hierarchy a/b/c/d on the mount and the
user is cd'ed into the directory. Bring down one of the bricks of replica and
remove all directories/files to simulate disk replacement on that brick. Now
this brick is brought back up. Creates on the cd'ed directory fail with ESTALE.
Basically before sending a create of 'f' inside 'd', fuse sends a lookup to
make sure the file is not present. On one of the bricks 'd' is present and
'f' is not so it sends ENOENT as response. On the new brick 'd' itself is not
present. So it sends ESTALE. In afr ESTALE is considered to be special errno on
witnessing which lookup has to fail. And ESTALE is given more priority than
ENOENT. Due to these reasons lookup fails with ESTALE rather than ENOENT. Since
lookup didn't fail with ENOENT, 'create' can't be issued so the command is
failed with ESTALE.
Solution:
Afr needs to consider ESTALE errno normally and ENOENT needs to
be given more priority so that operations like create can proceed even when
only one of the brick is up and running. Whenever client xlator identifies
that gfid-changed, it sets that information in lookup xdata. Afr uses this
information to fail the lookup with ESTALE so that top xlator can send
fresh lookup.
Change-Id: Ica6ce01baef08620154050a635e6f97d51029ef6
BUG: 1106408
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/8015
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change important (from a diagnostics point of view) log messages to use
the gf_msg() framework.
Change-Id: I0a58184bbb78989db149e67f07c140a21c781bc2
BUG: 1075611
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/7784
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Have common place to perform quorum fop wind check
- Check if fop succeeded in a way that matches quorum
to avoid marking changelog in split-brain.
BUG: 1066996
Change-Id: Ibc5b80e01dc206b2abbea2d29e26f3c60ff4f204
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/7600
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
For write fops afr's transaction eager-lock init adds transactions
that can share eager-lock to fdctx list. But if eager-lock finodelk
fop fails the stub remains in the list. This could later lead to
corruption of the list and lead to infinite loop on the list
leading to a mount hang.
Fix:
Remove the stub when finodelk fails.
Change-Id: I0ed4bc6b62f26c5e891c1181a6871ee6e4f4f5fd
BUG: 1063190
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/6944
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Anand Avati <avati@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Remove client side self-healing completely (opendir, openfd, lookup)
- Re-work readdir-failover to work reliably in case of NFS
- Remove unused/dead lock recovery code
- Consistently use xdata in both calls and callbacks in all FOPs
- Per-inode event generation, used to force inode ctx refresh
- Implement dirty flag support (in place of pending counts)
- Eliminate inode ctx structure, use read subvol bits + event_generation
- Implement inode ctx refreshing based on event generation
- Provide backward compatibility in transactions
- remove unused variables and functions
- make code more consistent in style and pattern
- regularize and clean up inode-write transaction code
- regularize and clean up dir-write transaction code
- regularize and clean up common FOPs
- reorganize transaction framework code
- skip setting xattrs in pending dict if nothing is pending
- re-write self-healing code using syncops
- re-write simpler self-heal-daemon
Change-Id: I1e4080c9796c8a2815c2dab4be3073f389d614a8
BUG: 1021686
Signed-off-by: Anand Avati <avati@redhat.com>
Reviewed-on: http://review.gluster.org/6010
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Under the entry self heal, readlink is done at the
source and sink. When readlink is done at the sink,
because link is not present at the sink, afr expects
ENOENT. AFR translator takes decisions for new link
creation based on ENOENT but server translator is modified
to return ESTALE because of which afr xlator is not able
to heal.
Fix:
The check for inode absence at server includes ESTALE as
well.
Change-Id: I319e4cb4156a243afee79365b7b7a5a7823e9a24
BUG: 1046624
Signed-off-by: Venkatesh Somyajulu <vsomyaju@redhat.com>
Reviewed-on: http://review.gluster.org/6599
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anand Avati <avati@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also renamed allow-sh-for-running-transaction -> attempt-self-heal
Change-Id: I134cc79e663b532e625ffc342c59e49e71644ab3
BUG: 1039544
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/6463
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: venkatesh somyajulu <vsomyaju@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|