IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
The current failure to handle short writes on writev fops leaves
us open to file corruption. A short write on a user request is
ignored and leaves replicas in an inconsistent state. A short write
during a self-heal is ignored and incorrectly marks the files as
consistent if the heal completes.
Modify user writev handling to return the best case return value
from each of the replicas. Short writes that occur relative to this
value are marked as failed and will require a heal. Modify
self-heal to set an error on a short write and abort the heal.
BUG: 853690
Change-Id: I18b30f58702326249230eeebb361b29e40b535f5
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-on: http://review.gluster.org/4150
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
A commonly faced problem among glusterfs users is: after a fresh
installation of glusterfs in a virtual machine, the VM image is
cloned to make multiple instances of the server. This breaks
glusterd because right after glusterfs installation on the first
boot glusterd would have created the node UUID and this gets
inherited into the clone. The result is wierd behavior at the time
of peer probe where glusterd does not (yet) deal with UUID
collisions in a user friendly way.
To handle it gluster peer reset command is implemented which upon
execution changes the uuid of local glusterd.
Change-Id: If207dd2ad93ab94ef1a3253f409c21c442975f87
BUG: 811493
Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
Reviewed-on: http://review.gluster.org/3637
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anand Avati <avati@redhat.com>
- epoll on RPCSVC_EVENT_ACCEPT would add corresponding xprt
onto the xprt_list. Concurrently, synctask thread (volume op)
would call into glusterd_fetchspec_notify which iterates on
the xprt_list. Added a mutex to protect such a racy access of
the list.
Change-Id: Idc51b4bdb1c814dfab7790e1c899d6977f7640f2
BUG: 878873
Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com>
Reviewed-on: http://review.gluster.org/4241
Reviewed-by: Raghavendra G <raghavendra@gluster.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
with the option, the idea is all client-side caching will be disabled,
where as on server side process, the fd will be treated as a regular
fd, thus helping the performance better.
"gluster volume set <VOLNAME> remote-dio enable" would set
this option in client protocol volumes.
Change-Id: Id2255a167137f8fee20849513e3011274dc829b4
Signed-off-by: Amar Tumballi <amarts@redhat.com>
BUG: 845213
Reviewed-on: http://review.gluster.org/4206
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
The functional issue is described by the subject line. This patch also
addresses several efficiency/structure issues, such as...
* Calling dict_set_ptr once for each txn type, instead of once overall.
* Calling afr_index_for_transaction_type once per iteration instead of
once per call (or better yet zero since the conversion is unnecessary).
* Implementation of inner functions in a different file than their one
caller, creating a spurious header-file dependency.
Change-Id: I29e0df906a820533b66b9ced73e015dfe77267d2
BUG: 865825
Signed-off-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-on: http://review.gluster.org/4070
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Reviewed-by: Anand Avati <avati@redhat.com>
Problem:
Whenever gluster volume heal vol full command is executed, the entries
stored in the circual buffer for sh->healed are added in the dictionary
in the _crawl_post_sh_action function irrespective of whether actual self heal
(due to non-zero values in chage log) takes place or not.
Fix:
Value of key (actual-sh-done) will be set to 1 whenever self heal takes place
due to non-zero change log values and if for some FOP self heal daemon finds
that no self heal required after examining the pending matrix, the value will
be 0.
Change-Id: I11fd0b9ee76759af17c5bca6bfafbaf66bcaacbc
BUG: 863068
Signed-off-by: Venkatesh Somyajula <vsomyaju@redhat.com>
Reviewed-on: http://review.gluster.org/4181
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
The 'least-rate-limit' io-threads translator option enables
throttling of least priority operations. This is initially intended
as a debug/diagnostic tool for users who might experience
overloaded servers via background activity (i.e., self-heal).
least-rate-limit defines the maximum number of least priority
operations the io-threads translator will dequeue in one second.
If the specified rate limit is met, the worker threads sleep for
the minimal amount of time before the next least priority operation
becomes available (or until a new request arrives).
The requests/second metric is generic and relative to a variety of
factors involved with a background operation (server, storage,
etc.). The most recent measured rate ("cached least rate") is added
to the io-threads state dump content (kill -USR1) to serve as a
reference point to throttle background activity under particular
conditions.
Change-Id: I80f2282992137d57b1becaa5c6ae3858c066862a
BUG: 853680
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-on: http://review.gluster.org/4119
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anand Avati <avati@redhat.com>
Problem:
If the extended attribute (trusted.glusterfs.volume-id) of a brick is
absent and <gluster volume start volume-name> command is executed then
curretly volume-id from the volume file will be set as an extended attribute of
the brick and volume will get started.
But if setup is such that brick is used as a mount point and before
executing the <gluster volume start volume-name> command, nothing is mounted on
the brick then all the file operations will take place at the brick but actual
intention of the brick is to be used as mount point only.
FIX:
Do not start the volume if extended attribute (trusted.glusterfs.volume-id)
is set absent.
Change-Id: Id2462d87d6087e97e0b8831512fdbc3595f7078b
BUG: 860297
Signed-off-by: Venkatesh Somyajulu <vsomyaju@redhat.com>
Reviewed-on: http://review.gluster.org/4202
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
Problem:
When posix-aio is enabled to perform aio fd is set with O_DIRECT
whenever possible in read, writev fops. Rchecksum does not take
this into account. If either offset/size/memory-buf passed to
pread in rchecksum fop is not aligned, pread fails with EINVAL.
Fix:
Before doing pread necessary O_DIRECT manipulation is done when
aio is enabled. Memory buffer passed to pread is now page-aligned.
Test:
1) Create replica volume with aio enabled.
2) dd if=/dev/urandom of=a bs=1M count=1
3) kill one of the bricks in the replica pair
4) dd if=/dev/urandom of=a bs=1M count=1
5) bring back the brick. Self-heal succeeds after the change.
The test above checks both rchecksum, writev fops that were
changed in this patch.
Change-Id: I186099a2854d4864c5b48086ab7bc5f1a7b27313
BUG: 866459
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
Reviewed-on: http://review.gluster.org/4134
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
First issue is that _EXPECT wasn't handling args with spaces correctly. This
was fixed by enclosing the entire expected-value expression in double quotes
with a simple backslash escape for the end-of-line dollar sign.
Second issue is that we were throwing away potentially useful debug
information. Fixed by using egrep's -q option to suppress output, and
eliminating redirections so that we can see any stderr output related to a
script error.
Change-Id: Ide3f49558dcece55bd90cad50b1ffc572592f11c
Signed-off-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-on: http://review.gluster.org/4126
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
Tested-by: Vijay Bellur <vbellur@redhat.com>
Recent changes in fuse-resolver had broken the use case of having
a trivial configuration with only storage/posix translator. Fix
it and include a regression test to avoid breakage in the future.
Change-Id: I85cbcac4b5c2d3517dd03fd3dc6d07610697981b
BUG: 868478
Signed-off-by: Anand Avati <avati@redhat.com>
Reviewed-on: http://review.gluster.org/4114
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
The internal mount API had no access to the generic
mountflags used by mount(2).
Thus the "ro" mount option that needs to be passed down to mount(2) as
as a mountflag was incorrectly mangled into the fuse-specific mount
parameter string (cf. http://review.gluster.com/655).
This commit fixes the internal API and the "ro" issue. It also adds a
check for the "rw" and "ro" mount options in tests/basic/mount.t.
Thanks to Csaba Henk (csaba@) for suggestions and proposing an updated
patch.
Change-Id: I7f7bf49ae44d148f5c16f10736a0e412fb8f5e67
BUG: 853895
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Reviewed-on: http://review.gluster.org/4163
Reviewed-by: Csaba Henk <csaba@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
When a user sets nfs.enable-ino32 on,
the root inode, which must be 1, is jumbled during
hashing of the gfid into 32 bits.
This patch avoids doing that, and returns inode of 1
for the root inode.
Change-Id: Ib65f2660998a95e1059be32bc298485c5cbe52df
BUG: 864222
Signed-off-by: Rajesh Amaravathi <rajesh@redhat.com>
Reviewed-on: http://review.gluster.org/4154
Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
Add anonymous member to fd_t and use it instead of over-loading pid for
geo-rep and self heal
Change-Id: I4d6b29a044a8ed4b8f69ff6e3f35ee227739b2af
Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
BUG: 874272
Reviewed-on: http://review.gluster.org/4185
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
This commit resolves the parent inode in nfs_inode_loc_fill
if the inode has a resolved path.
Change-Id: If407c91c246b0b9f3349cedae0baec8bd6831d10
BUG: 872923
Signed-off-by: Rajesh Amaravathi <rajesh@redhat.com>
Reviewed-on: http://review.gluster.org/4157
Reviewed-by: Krishna Srinivas <krishna.zresearch@gmail.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
use event-history framework for saving and dumping (on necessity)
important xlator specific information.
Tests:
Included the regression testcase.
Change-Id: I6c0532e9ffe0b624286cdc4d2637b1bd2c0579e0
BUG: 858215
Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
Signed-off-by: root <root@thinkpad.(none)>
Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
Reviewed-on: http://review.gluster.org/3925
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
PROBLEM:
The first colon in the limit string is used to separate the path and the
limit-value. The ':' in the path was the source of problem.
FIX:
Search for the last colon and separate based on the same.
Added regression test.
TEST:
Create and set quota on directories with names containing ':' in start, middle
and end.
Change-Id: I363c8ad4cbfd02c23fc73974bef9aa8bc362d29c
BUG: 848251
Signed-off-by: Varun Shastry <vshastry@redhat.com>
Reviewed-on: http://review.gluster.org/4137
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
in a fix to avoid race between stopping the brick and deleting
it in remove-brick (15396f490d)
we moved delete of the brick inside stop volume, which was
totally inside the 'if (volinfo->status == STARTED)' section.
thus it made remove-brick of stopped/created volume as a failure.
Signed-off-by: Amar Tumballi <amarts@redhat.com>
BUG: 867252
Change-Id: Ie251e59a0b7ddb6965d8d48fb2a3bdb3bd11653d
Reviewed-on: http://review.gluster.org/4127
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: Jeff Darcy <jdarcy@redhat.com>
Framework for writing test cases to be submitted with patches.
This framework and the test cases get exercised by Jenkins in
the pre-commit regression test. Jenkins is configured to give
a +1 verified vote only if the regression test passes without
failures (which includes test cases added/changed by the patch
being tested)
Every patch should include a test case (either extensions/changes
to existing test cases or add new ones, as appropriate). The test
case should be part of the same commit so that both code and
test case get reviewed together.
Test cases added are cumulative. Every new patch gets
tested against its own test case and every test case previously
added.
A lot of new commits in the near future will be pure test cases
(with no code change) which will get added in "catch up" mode.
The tool used for implementing test cases is 'prove', and the
framework itself is modeled similar to the POSIX compliance
filesystem test suite.
Under the top level directory, a new directory named 'tests/'
is added. This contains top level classifier directories and
framework files/scripts.
Functionality tests should be created under a classifier directory
below 'tests/'. For e.g:
tests/basic/mount.t
tests/performance/write-behind.t
Bugs which get fixed should include a test case script named
by the bug id, so that we are guaranteed any new change will
not bring the issue back. For e.g:
tests/bugs/bug-123456.t
Triggering of regression tests in Jenkins is manual at this point
as we do not want the entire test suite to run against every
revision of a patch while it is still in the review/resubmit cycle.
Signed-off-by: Anand Avati <avati@redhat.com>
Change-Id: I8078244619135ccaba38e068925f8ca85141055a
BUG: 764966
Signed-off-by: Anand Avati <avati@redhat.com>
Reviewed-on: http://review.gluster.org/4101
Tested-by: Gluster Build System <jenkins@build.gluster.com>
