| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Consider a replica setup, where one mount writes data to a
file and the other mount reads the file. In afr, read operations
are not transaction based, a brick(read subvolume) is chosen as
a part of lookup or other operations, read is always wound only
to the read subvolume, even if there was write from a different client
that failed on this brick. This stale read continues until there is
a lookup or any write operation from the mount point. Currently, this
is not a major issue, as a lookup is issued before every read and it will
switch the read subvolume to a correct one. But with the plan of
increasing md-cache timeout to 600s, the stale read problem will be
more pronounced, i.e. stale read can continue for 600s(or more if cascaded
with readdirp), as there will be no lookups.
Solution:
Afr doesn't have any built-in solution for stale read(without affecting
the performance). The solution that came up, was to use upcall. When a file
on any brick is marked bad for the first time, upcall sends a notification
to all the clients that had recently accessed the file. The solution has
2 parts:
- Identifying when a file is marked bad, on any of the bricks,
for the first time
- Client side actions on recieving the notifications
Identifying when a file is marked bad on any of the bricks for the first time:
-----------------------------------------------------------------------------
The idea is to track xattrop in upcall. xattrop currently comes with 2 afr
xattrs - afr dirty bit and afr pending xattrs.
Dirty xattr is set to 1 before every write, and is unset if write succeeds.
In certain scenarios, dirty xattr can be 0 and still the file could be bad
copy. Hence do not track dirty xattr.
Pending xattr is set on the good copy, indicating the other bricks that have
bad copy. It is still not as simple as, notifying when any of the pending xattrs
change. It could lead to flood of notifcations, in case the other brick is
completely down or consistantly failing. Hence it is important to notify only
once, the first time a good copy is marked bad.
Client side actions on recieving pending xattr change, notification:
--------------------------------------------------------------------
md-cache will invalidate the cache of that file, so that further lookup is
passed down to afr and hence update the read subvolume. Invalidating only in
md-cache is not enough, consider the folling oder of opertaions:
- pending xattr invalidation - invalidate md-cache
- readdirp on the bad read subvolume - fill md-cache
- lookup (served from md-cache)
- read - wound to the old read subvol.
Hence, along with invalidating md-cache, it is very important to reset the
read subvolume for that file, in afr.
Design Credit: Anuradha Talur, Ravishankar N
1. xattrop doesn't carry info saying post op/pre op.
2. Pre xattrop will have 0 value for all pending xattrs,
the cbk of pre xattrop carries the on-disk xattr value.
Non zero indicated healing is required.
3. Post xattrop will have non zero value for any of the
pending xattrs, if the fop failed on any of the bricks.
Change-Id: I469cbc111714c433984fe1c922be2ef113c25804
BUG: 1211863
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15398
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The earlier fix prevented rebalance from
starting even if only a single brick of a
replica set was down.
Change-Id: I8c9a7d4c1fe33f7749568357462f29796e1b832a
BUG: 1376671
Signed-off-by: N Balachandran <nbalacha@redhat.com>
Reviewed-on: http://review.gluster.org/15517
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
correct
Problem : "gluster v heal test statistcs heal-count replica" does not
show correct output.
Solution: After update condition (match brick name) in
_select_hxlator_with_matching_brick, it shows correct output.
BUG: 1325792
Change-Id: I60cc7c68ea70bce267a747570f91dcddbc1d9016
Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
Reviewed-on: http://review.gluster.org/15494
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If connect fails with any other error than EINPROGRESS we cannot get
the error status using getsockopt (... SO_ERROR ... ). Hence we need
to remember the state of connect and take appropriate action in the
event_handler for the same.
As an added note, a event can come where poll_err is HUP and we have
poll_in as well (i.e some status was written to the socket), so for
such cases we need to finish the connect, process the data and then
the poll_err as is the case in the current code.
Special thanks to Kaushal M & Raghavendra G for figuring out the issue.
Change-Id: Ic45ad59ff8ab1d0a9d2cab2c924ad940b9d38528
BUG: 1372356
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Signed-off-by: Shyam <srangana@redhat.com>
Reviewed-on: http://review.gluster.org/15440
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: In a distributed-replicate volume attribute
"replica.split-brain-status" value does not display split-brain
condition though directory is in split-brain.
If directory is in split brain on mutiple replica-pairs
it does not show full list of replica pairs.
Solution: Update the dht_aggregate code to aggregate the xattr
value in this specific condition.
Fix: 1) function getChoices returns the choices from split-brain
status string.
2) function add_opt adding the choices to local buffer to
store in dictionary
3) For the key "replica.split-brain-status" function dht_aggregate
call dht_aggregate_split_brain_xattr to prepare the list.
Test: To verify the patch followed below steps
1) Create a distributed replica volume and create mount point
2) Stop heal daemon
3) Touch file and directories on mount point
mkdir test{1..5};touch tmp{1..5}
4) Down brick process on one of the replica set
pkill -9 glusterfsd
5) Change permission of dir on mount point
chmod 755 test{1..5}
6) Restart brick process on node with force option
7) kill brick process on other node in same replica set
8) Change permission of dir again on mount point
chmod 766 test{1..5}
9) Reexecute same step from 4-9 on other replica set also
10) After check heal status on server it will show dir's are
in split brain on all replica sets
11) After check the replica.split-brain-status attr on mount
point it will show wrong status of split brain.
12) After apply the patch the attribute shows correct value.
BUG: 1368312
Change-Id: Icdfd72005a4aa82337c342762775a3d1761bbe4a
Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
Reviewed-on: http://review.gluster.org/15201
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Post add-brick event the new brick will have permission of 755
by default. If the root directory permission was other than 755,
that does not get healed to the new brick leading to permission
errors/inconsistencies.
For choosing source of attr heal we can trust the subvols which
have layouts with latest ctime(as part of missing directory heal,
we heal the proper attr). In case none of the subvols have layout,
return ESTALE to retrigger a fresh lookup.
Note: This patch heals the permission of the root directories only.
Since, permission healing of directory is not straight forward and
required intrusive fix, those are not addressed here.
Change-Id: If894e3895d070d46b62d2452e52c1eaafcf56c29
BUG: 1368012
Signed-off-by: Susant Palai <spalai@redhat.com>
Reviewed-on: http://review.gluster.org/15195
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
wait for remove brick to complete before attempt for a commit.
Change-Id: I66ea6c48b6a69fe33d79f9d9080b6f2c1462578e
BUG: 1374993
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-on: http://review.gluster.org/15457
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
dht_selfheal_layout_maximize_overlap () does not consider
chunk sizes while calculating overlaps. Temporarily
enabling this operation if only if weighted rebalance is disabled
or all bricks are the same size.
Change-Id: I5ed16cdff2551b826a1759ca8338921640bfc7b3
BUG: 1366494
Signed-off-by: N Balachandran <nbalacha@redhat.com>
Reviewed-on: http://review.gluster.org/15403
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Verify if the unlink, rename and other ops are reflected both on
the current mount and other mounts.
Change-Id: I5a296cdd557194dcf487e65ee4a14bbeaf4be690
BUG: 1211863
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15419
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently md-cache invalidation feature is enabled by setting
"performance.cache-invalidation", but this case was sent when
"features.cache-invalidation" was enabling md-cache invalidation.
Hence, fix the same.
Change-Id: If044f6208179748a120fbe1d63b676367e707f73
BUG: 1372584
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15385
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Raghavendra Talur <rtalur@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
With this patch, there will be additional entries seen in
the profile info:
UPCALL : Total number of upcall events that were sent from
the brick(in brick profile), and number of upcall
notifications recieved by client(in client profile)
Cache invalidation events:
-------------------------
CI_IATT : Number of upcalls that were cache invalidation and
had one of the IATT_UPDATE_FLAGS set. This indicates
that one of the iatt value was changed.
CI_XATTR : Number of upcalls that were cache invalidation, and
had one of the UP_XATTR or UP_XATTR_RM set. This indicates
that an xattr was updated or deleted.
CI_RENAME : Number of upcalls that were cache invalidation,
resulted by the renaming of a file or directory
CI_UNLINK : Number of upcalls that were cache invalidation,
resulted by the unlink of a file.
CI_FORGET : Number of upcalls that were cache invalidation,
resulted by the forget of inode on the server side.
Lease events:
------------
LEASE_RECALL : Number of lease recalls sent by the brick (in
brick profile), and number of lease recalls recieved
by client(in client profile)
Note that the sum of CI_IATT, CI_XATTR, CI_RENAME, CI_UNLINK,
CI_FORGET, LEASE_RECALL may not be equal to UPCALL. This is
because, each cache invalidation can carry multiple flags.
Eg:
- Every CI_XATTR will have CI_IATT
- Every CI_UNLINK will also increment CI_IATT as link count is an
iatt attribute.
Also UP_PARENT_DENTRY_FLAGS is currently not accounted for,
as CI_RENAME and CI_UNLINK will always have the flag
UP_PARENT_DENTRY_FLAGS
Change-Id: Ieb8cd21dde2c4c7618f12d025a5e5156f9cc0fe9
BUG: 1371543
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15193
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Rajesh Joseph <rjoseph@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixed the way shd up check is done to prevent self-heal daemon
not running error when heal full command is executed.
Change-Id: I93c4a0da12316373d62cd4ea74432cd9bf2b090c
BUG: 1370053
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/15341
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Anuradha Talur <atalur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently when a client performs a readdirp it is not stored
in upcall, as one of the clients that have accessed the files.
Hence, when any other client modifies the file, the client that
had performed readdirp will not get any notifications.
Fix this by adding the clients to upcall database when they
perform readdirp.
Change-Id: I7767f1e26bf1bd1f67702a6d01f8aa64526ccc46
BUG: 1369430
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15313
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: soumya k <skoduri@redhat.com>
Reviewed-by: Niels de Vos <ndevos@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently, md-cache only processes IATT_UPDATE_FLAGS, UP_XATTR and
UP_XATTR_RM. We also need to process UP_RENAME_FLAGS, UP_FORGET,
UP_PARENT_DENTRY_FLAGS and UP_NLINK_FLAGS. Otherwise the files
unlinked or renamed will not be reflected on other mounts.
Change-Id: Icb8b03da51482c3fc2e2a7292d16d56e11a341d9
BUG: 1211863
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15324
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: If88efe3db782a6156614af4c650d53b159ade57f
BUG: 1371541
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15354
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The command basically allows replace brick with src and
dst bricks as same.
Usage:
gluster v reset-brick <volname> <hostname:brick-path> start
This command kills the brick to be reset. Once this command is run,
admin can do other manual operations that they need to do,
like configuring some options for the brick. Once this is done,
resetting the brick can be continued with the following options.
gluster v reset-brick <vname> <hostname:brick> <hostname:brick> commit {force}
Does the job of resetting the brick. 'force' option should be used
when the brick already contains volinfo id.
Problem: On doing a disk-replacement of a brick in a replicate volume
the following 2 scenarios may occur :
a) there is a chance that reads are served from this replaced-disk brick,
which leads to empty reads. b) potential data loss if next writes succeed
only on replaced brick, and heal is done to other bricks from this one.
Solution: After disk-replacement, make sure that reset-brick command is
run for that brick so that pending markers are set for the brick and it
is not chosen as source for reads and heal. But, as of now replace-brick
for the same brick-path is not allowed. In order to fix the above
mentioned problem, same brick-path replace-brick is needed.
With this patch reset-brick commit {force} will be allowed even when
source and destination <hostname:brickpath> are identical as long as
1) destination brick is not alive
2) source and destination brick have the same brick uuid and path.
Also, the destination brick after replace-brick will use the same port
as the source brick.
Change-Id: I440b9e892ffb781ea4b8563688c3f85c7a7c89de
BUG: 1266876
Signed-off-by: Anuradha Talur <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/12250
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Use defined HEAL and PROCESS_UP timeouts rather than
hard code them in self-heald.t.
Change-Id: I21586811904c8417b7208bb643f14dff20dc4832
BUG: 1370074
Signed-off-by: Anuradha Talur <atalur@redhat.com>
Reviewed-on: http://review.gluster.org/15316
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Tested-by: Ravishankar N <ravishankar@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently, the volume set option features.cache-invalidation enables upcall
feature on server side and md-cache cache-invalidation on client side.
There are multiple problems that can arise from this:
1. The scenario when user wants to, enable upcall for nfs-ganesha setup,
but do not want to enable md-cache cache-invalidation, as the
nfs-clients have already cached the metadata and upcall is used to
to invalidate the nfs-client cache. In this case, users should have
a way of disabling md-cache invalidation without disabling upcall.
2. Upcall requires a op-version of GD_OP_VERSION_3_7_0, where as
md-cache invalidation requires an op version of GD_OP_VERSION_3_9_0.
Consider a setup where the servers are in op-version GD_OP_VERSION_3_7_0,
and th clients are in op-version GD_OP_VERSION_3_9_0. if there is one
single volume set option, user can enable this feature in this setup.
But it can lead to stale xattr cache as the xattr invalidation was
introduced in upcall only in release 3.8. Hence, we should not be
able to enable md-cache invalidation, if all the servers and clients
are not on opversion >= GD_OP_VERSION_3_9_0.
To solve the above mentioned issues, we have seperate volume options
for enabling md-cache invalidation and upcall. But this can lead to
issues when user enable md-cache invalidation and forgets to enable
upcall. Probably in the next release, these can be enables by default.
Change-Id: Ie70eff97fe12fcb623eec8f4f5861ac065bf483e
BUG: 1211863
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15314
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: soumya k <skoduri@redhat.com>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently there is no existing CLI that can be used to get the
local state representation of the cluster as maintained in glusterd
in a readable as well as parseable format.
The CLI added has the following usage:
# gluster get-state [daemon] [odir <path/to/output/dir>] [file <filename>]
This would dump data points that reflect the local state
representation of the cluster as maintained in glusterd (no other
daemons are supported as of now) to a file inside the specified
output directory. The default output directory and filename is
/var/run/gluster and glusterd_state_<timestamp> respectively. The
option for specifying the daemon name leaves room to add support for
other daemons in the future. Following are the data points captured
as of now to represent the state from the local glusterd pov:
* Peer:
- Primary hostname
- uuid
- state
- connection status
- List of hostnames
* Volumes:
- name, id, transport type, status
- counts: bricks, snap, subvol, stripe, arbiter, disperse,
redundancy
- snapd status
- quorum status
- tiering related information
- rebalance status
- replace bricks status
- snapshots
* Bricks:
- Path, hostname (for all bricks these info will be shown)
- port, rdma port, status, mount options, filesystem type and
signed in status for bricks running locally.
* Services:
- name, online status for initialised services
* Others:
- Base port, last allocated port
- op-version
- MYUUID
Change-Id: I4a45cc5407ab92d8afdbbd2098ece851f7e3d618
BUG: 1353156
Signed-off-by: Samikshan Bairagya <samikshan@gmail.com>
Reviewed-on: http://review.gluster.org/14873
Reviewed-by: Avra Sengupta <asengupt@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
cyclic order
When the bricks are brought offline and then online in cyclic
order while writes are in progress on a file, thanks to inode
refresh in write txns, AFR will mostly fail the write attempt
when the only good copy is offline. However, there is still a
remote possibility that the file will run into split-brain if
the brick that has the lone good copy goes offline *after* the
inode refresh but *before* the write txn completes (I call it
in-flight split-brain in the patch for ease of reference),
requiring intervention from admin to resolve the split-brain
before the IO can resume normally on the file. To get around this,
the patch does the following things:
i) retains the dirty xattrs on the file
ii) avoids marking the last of the good copies as bad (or accused)
in case it is the one to go down during the course of a write.
iii) fails that particular write with the appropriate errno.
This way, we still have one good copy left despite the split-brain situation
which when it is back online, will be chosen as source to do the heal.
Change-Id: I9ca634b026ac830b172bac076437cc3bf1ae7d8a
BUG: 1363721
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/15080
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Oleksandr Natalenko <oleksandr@natalenko.name>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Generate SSL certificates before enabling
management encryption to avoid test failure.
Change-Id: Iab23b36703f4653f1d5bb9d14695e4d3fa63ad61
Signed-off-by: Ashish Pandey <aspandey@redhat.com>
BUG: 1368349
Signed-off-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-on: http://review.gluster.org/15202
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
http://review.gluster.org/#/c/14758/ introduces a check in
glusterd_restart_bricks that makes sure that if server quorum is
enabled and if the glusterd instance has been restarted, the bricks
do not get started. This prevents bricks which have been brought
down purposely, say for maintainence, from getting started
upon a glusterd restart. However this change introduced regression
for a situation that involves multiple volumes. The bricks from
the first volume get started, but then for the subsequent volumes
the bricks do not get started. This patch fixes that by setting
the value of conf->restart_done to _gf_true only after bricks are
started correctly for all volumes.
Change-Id: I2c685b43207df2a583ca890ec54dcccf109d22c3
BUG: 1367478
Signed-off-by: Samikshan Bairagya <samikshan@gmail.com>
Reviewed-on: http://review.gluster.org/15183
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently per xlator loglevel setting doesn't work, due to the
flaw in loglevel checking. Fix the same.
Per xlator logging can be set using the below command:
Eg: setfattr -n trusted.glusterfs.patchy-md-cache.set-log-level -v TRACE /mnt/glusterfs/0
Change-Id: I8ff1d15bd5693b6f682d99bee22a4bbb5eee646c
BUG: 1362520
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/15071
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Niels de Vos <ndevos@redhat.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
triggered
Problem: After add brick to a distribute volume to convert to replica is not
triggering self heal.
Solution: Modify the condition in brick_graph_add_index to set trusted.afr.dirty
attribute in xlator.
Test : To verify the patch followd below steps
1) Create a single node volume
gluster volume create <DIS> <IP:/dist1/brick1>
2) Start volume and create mount point
mount -t glusterfs <IP>:/DIS /mnt
3) Touch some file and write some data on file
4) Add another brick along with replica 2
gluster volume add-brick DIS replica 2 <IP>:/dist2/brick2
5) Before apply the patch file size is 0 bytes in mount point.
BUG: 1365455
Change-Id: Ief0ccbf98ea21b53d0e27edef177db6cabb3397f
Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
Reviewed-on: http://review.gluster.org/15118
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-by: Anuradha Talur <atalur@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Right now glusterd won't come up if vols directory contains an invalid entry.
Instead of doing that with this change a message will be logged and then skip
that entry
Change-Id: I665b5c35291b059cf054622da0eec4db44ec5f68
BUG: 1318591
Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
Reviewed-on: http://review.gluster.org/13764
Reviewed-by: Prashanth Pai <ppai@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
At the time of start of a volume, it is checked if
.glusterfs/unlink exist or not. If it does, move it
to landfill and recreate unlink directory. If a volume
is mounted and we write data on it till we face ENOSPC,
restart of that volume fails as it will not be able to
create unlink dir. mkdir will fail with ENOSPC.
This will not allow volume to restart.
Solution:
If .glusterfs/unlink directory exist, don't move it to
landfill. Delete all the entries inside it.
Change-Id: Icde3fb36012f2f01aeb119a2da042f761203c11f
BUG: 1360679
Signed-off-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-on: http://review.gluster.org/15030
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of rebalance stop, its always better to wait for rebalance to complete
as the former doesn't have any purpose.
Change-Id: Ia1bc2a34d937a0a96543bebd257dcda619f12474
BUG: 1363948
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-on: http://review.gluster.org/15085
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit 3c04a91 removed setting typeStr to NULL if num_allocs is set to 0, this
has caused this regression. Code has been put back like earlier and to avoid
statedump printing all the NULL values check is modified to see skip the records
if num_allocs is 0 instead of total_allocs
Change-Id: Ib8bcc2fba908e88cf52b641c3f6bcba74f5e667c
BUG: 1359190
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-on: http://review.gluster.org/14987
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: N Balachandran <nbalacha@redhat.com>
Reviewed-by: Prashanth Pai <ppai@redhat.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
GlusterD as of now was blindly assuming that the brick port which was already
allocated would be available to be reused and that assumption is absolutely
wrong.
Solution : On first attempt, we thought GlusterD should check if the already
allocated brick ports are free, if not allocate new port and pass it to the
daemon. But with that approach there is a possibility that if PMAP_SIGNOUT is
missed out, the stale port will be given back to the clients where connection
will keep on failing. Now given the port allocation always start from base_port,
if everytime a new port has to be allocated for the daemons, the port range will
still be under control. So this fix tries to clean up old port using
pmap_registry_remove () if any and then goes for pmap_registry_alloc ()
Change-Id: If54a055d01ab0cbc06589dc1191d8fc52eb2c84f
BUG: 1221623
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-on: http://review.gluster.org/15005
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Avra Sengupta <asengupt@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Continuation of http://review.gluster.org/#/c/14985.
Also renamed tests/bugs/disperse to tests/bugs/ec for a better
correlation to tests/basic/ec and xlators/cluster/ec
Change-Id: I662b3477c12af8a0b94597769e8f00f354b1168c
BUG: 1332054
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/15006
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This will be unnecessary, and mostly in the way, as real fairness
guarantees are implemented.
Change-Id: Ic61ec1c9e9add58385f1a4eafcfe2cc554ceefc8
BUG: 1360402
Signed-off-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-on: http://review.gluster.org/14989
Smoke: Gluster Build System <jenkins@build.gluster.org>
Tested-by: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We were previously reading the pidfile, and displaying
the pid even if snapd daemon is not running. Now to fix
it, we re-assign pid value to -1, if snapd is offline.
Change-Id: I4baff8d489fe9380061c52aea006db90fa421cd7
BUG: 1358244
Signed-off-by: Avra Sengupta <asengupt@redhat.com>
Reviewed-on: http://review.gluster.org/14981
Tested-by: Vijay Bellur <vbellur@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Thanks to Krutika for a cleaner way to track inode refs in
afr_set_split_brain_choice().
Change-Id: I2d968d05b815ad764b7e3f8aa9ad95a792b3c1df
BUG: 1355604
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/14895
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Moving ./tests/bugs/snapshot/bug-1316437.t to bad test, while mulling
over the pros and cons of the fix. Will update the bug, as we go.
Sending this patch to unblock master.
Change-Id: Ia863312913686b4fa0ee0b63da13aedc0439a835
BUG: 1359717
Signed-off-by: Avra Sengupta <asengupt@redhat.com>
Reviewed-on: http://review.gluster.org/15001
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: N Balachandran <nbalacha@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
EXPECT_WITHIN takes regular expression to match the count,
so even when there are say 10 entries to heal, it would
think that the heal is complete. Fixed checking
pending heal count with correct regex.
Thanks to Xavi for finding this problem.
Change-Id: Ic593d22468b2b586bfca864962ffa0eda96b1d1f
BUG: 1332054
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/14985
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: I32bfec4af91348d96dc3e81a9d5c9cad599f821b
Bug: 1358594
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/14748
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra Talur <rtalur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On a faster machine the ps check was returning two pids,
including the glusterfsd process's pid, right after that,
process forked. Hence removing that ps, as for the scope
of this test, verifying the snapd pid from the status
command itself is enough.
Change-Id: I8bd8fc4ea406d96e3a47f952cfe44560b615dbe6
BUG: 1358195
Signed-off-by: Avra Sengupta <asengupt@redhat.com>
Reviewed-on: http://review.gluster.org/14963
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Rajesh Joseph <rjoseph@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
md-cache currently updates its stat in cbks of selected fops.
The default cache time is 1 second, if this is increasd to reap the
benefits of caching, we may end up with stale cache for long time,
as there is no logic yet to notify md-cache of backend changes by
another client.
Solution:
Use the existing upcall mechanism to invalidate the cache.
For this feature to work, "features.cache-invalidation" volume
option should be enabled.
This patch as is doesn't improve any performance, the benifit of the
patch is that it provides coherency for stat cache, hence the cache
timeout can be quite longer which in turn can improve the performance.
Change-Id: I2dbb0afa7b5e4a5a248f910188e0918e02f18692
BUG: 1211863
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/12951
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In the notify function in trash xlator, a lookup is performed
to obtain path of old trash directory. The result usually contains
path without '/' at the end. The trash xlator maintains expects
'/' at the end for the values such as 'old trash dir' and
'new trash dir'. Otherwise certian checks in the code will fail.
Change-Id: I89e02e4b249314fb6536297f959865feee182c83
BUG: 1357397
Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
Reviewed-on: http://review.gluster.org/14938
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Anoop C S <anoopcs@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts commit 0086a55bb7de1ef5dc7a24583f5fc2b560e835fd.
As part of Richard's patch for lock-revocation feature this bug
is completely fixed (I think at least ;-) ). So bringing these
back so that we will find out if there are anymore things we need
to address in this code path.
BUG: 1350867
Change-Id: If1440fc83b376576ae1a77b1156188a6bf53fe3a
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/14817
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
With a start and stop rebalance, the stop command may fail
as by that time the rebalance process may not come up.
Using the rebalance status commmand to ensure that the rebalance
process is up before stoping rebalance.
Change-Id: I3d5123cd5dfabde2720428455b257d11b980ce21
BUG: 1354372
Signed-off-by: Sakshi Bansal <sabansal@redhat.com>
Reviewed-on: http://review.gluster.org/14885
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Upon glusterd restart if it is observered that the server quorum
isn't met anymore due to changes to the "server-quorum-ratio"
global option, the bricks should be stopped if they are running.
Also if glusterd has been restarted, and if server quorum is not
applicable for a volume, do not restart the bricks corresponding
to the volume to make sure that bricks that have been brought
down purposely, say for maintenance, are not brought up. This
commit moves this check that was previously inside
"glusterd_spawn_daemons" to "glusterd_restart_bricks" instead.
Change-Id: I0a44a2e7cad0739ed7d56d2d67ab58058716de6b
BUG: 1345727
Signed-off-by: Samikshan Bairagya <samikshan@gmail.com>
Reviewed-on: http://review.gluster.org/14758
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Allow glusterd to spawn the daemons at the time of initialization when peer
count is less than 2. This is required if user wants to set up a two node
cluster with out server side quorum and want the bricks to come up on a node
where the other node is down, however the behaviour will be overriden when
server side quorum is enabled.
Change-Id: I21118e996655822467eaf329f638eb9a8bf8b7d5
BUG: 1352277
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-on: http://review.gluster.org/14848
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: I0a95f4897440c5bf6f54612d9c232e015c8bf983
BUG: 1211863
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/14824
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Prashanth Pai <ppai@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: I8404b864a405411e3af2fbee46ca20330e656045
BUG: 1351021
Signed-off-by: Sakshi Bansal <sabansal@redhat.com>
Reviewed-on: http://review.gluster.org/14827
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: I6ded40a1b1cff5c72e5b61fd353db3d8c688efd8
BUG: 1225718
Signed-off-by: Sakshi <sabansal@redhat.com>
Reviewed-on: http://review.gluster.org/10956
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: For a read on a file in metadata split-brain:
1.lookup_done resets event_generation to zero.
2. readv is issued, goes to inode refresh due to mismatching event_gen.
3. After refresh is successful, we update event_generation, data and
metdata readable.
3. We then call afr_read_txn_refresh_done() which in turn calls
afr_inode_get_readable() but doesn't check for EIO. So afr_readv_wind
is called with local->readable (which is populated with data_readable),
thus winding the read to a brick.
4. Also, further parallel reads that come directly go to the wind path
because there is no inode_refresh needed.
Fix:
1.For any afr_read_txn(), readable must be an intersection of data and metadata
readable.
2.Check for EIO in afr_read_txn_refresh_done().
Change-Id: I22dd221fdfaf96d7aced2f474e28ed1337d69f0e
BUG: 1305031
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/13389
Reviewed-by: Ashish Pandey <aspandey@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The opRet field was being assigned to 0 in the XML output when a
gluster volume info --xml call is made on a non-existent volume.
This change assigns a value of -1 to opRet for volume info calls
for non-existent volumes. Other fields like opErrno and opErrstr
are also assigned relevant values
Change-Id: I3920c602328f74252c87bb521f5a43d4bdc7d44d
BUG: 1321836
Signed-off-by: Samikshan Bairagya <samikshan@gmail.com>
Reviewed-on: http://review.gluster.org/13843
Smoke: Gluster Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: darshan n <dnarayan@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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.
Change-Id: I9de86555f66cc94a05649ac863b40ed3426ffd4b
BUG: 1343038
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-on: http://review.gluster.org/14656
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previous commit to fix the bug, where io-stat-dump was overwriting
the dump file when the client and a brick was on the same host,
failed to consider the existing behaviour where io-stats can
help generate closely correlated set of stats across clients
and bricks, by triggering the dump using the same command.
This was introduced in commit:
0facb11220aea20a6573b656785922219c9650cf
Further, by limiting the first io-stat to unwind the dump request,
there is no way to trigger other io-stat xlators in the stack to
dump their stat information.
This bug hence is being fixed by this commit keeping the
following in mind,
- We need to trigger io-stat-dump for all instances in the
graph when this attr is set
- We need to write the output to different files, so that
they do not overwrite each others data
- We need to prevent this xattr from being set on the path
that is used to trigger the io-stat-dump information
Change-Id: I31ec380f0d85e10313a9d7b977da0e1ec74638a6
BUG: 1322825
Signed-off-by: Shyam <srangana@redhat.com>
Reviewed-on: http://review.gluster.org/14552
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Ravishankar N <ravishankar@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
|