]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Yehuda Sadeh [Mon, 14 Oct 2013 17:26:49 +0000 (10:26 -0700)]
radosgw-admin: add --quota-scope param to usage
and also look at it when setting quota on a bucket.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Sat, 12 Oct 2013 00:21:21 +0000 (17:21 -0700)]
rgw: switch out param to a pointer instead of reference
following code review
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Sat, 12 Oct 2013 00:08:06 +0000 (17:08 -0700)]
test: update cli test for radosgw-admin
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Sat, 12 Oct 2013 00:07:38 +0000 (17:07 -0700)]
radosgw-admin: clarify --max-size param
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 11 Oct 2013 23:56:46 +0000 (16:56 -0700)]
formatter: dump_bool dumps unquoted strings
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 11 Oct 2013 23:43:00 +0000 (16:43 -0700)]
rgw_quoa.{h,cc}: add copyright notice
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 11 Oct 2013 04:04:39 +0000 (21:04 -0700)]
radosgw-admin: limit user bucket-level quota
We only allow it if --quota-scope=bucket is specified. This is done in
order to avoid confusion with the future user level quota command.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 10 Oct 2013 21:45:46 +0000 (14:45 -0700)]
radosgw-admin: can set user's bucket quota
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 10 Oct 2013 17:50:39 +0000 (10:50 -0700)]
rgw: bucket quota threshold
Add bucket quota threshold so that when we're passed that value we
reread the bucket stats before every write and not rely on cached value.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 8 Oct 2013 21:32:37 +0000 (14:32 -0700)]
rgw: configurable bucket quota size
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 8 Oct 2013 21:05:59 +0000 (14:05 -0700)]
rgw: protect against concurrent async quota updates
Leverage the cache lru_map locking for making sure that we don't end
up with more than a single concurrent async update on the same bucket
within the same update window.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 8 Oct 2013 03:22:51 +0000 (20:22 -0700)]
rgw: async quota update
Asynchronously update bucket stats when a period passed, but bucket
stats are within the ttl window.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 8 Oct 2013 18:20:56 +0000 (11:20 -0700)]
rgw: rearrange includes
Need to have ceph_json.h first due to some boost weirdness.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 4 Oct 2013 17:36:41 +0000 (10:36 -0700)]
rgw: update cache atomically
Use new lru_map::find_and_update() to update internal cache state.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 4 Oct 2013 17:36:13 +0000 (10:36 -0700)]
lru_map: add find_and_update()
A new find_and_update() call to make atomic changes.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 23:34:40 +0000 (16:34 -0700)]
rgw: init RGWBucketStats
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 23:29:23 +0000 (16:29 -0700)]
rgw: handle negative quota as non-assigned values
This means that we can have a 0 as a quota value.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 23:09:07 +0000 (16:09 -0700)]
radosgw-admin: quota control api
Can set, disable and enable bucket quota.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 23:08:49 +0000 (16:08 -0700)]
rgw: dump quota params even if disabled
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 22:35:43 +0000 (15:35 -0700)]
rgw: bucket stats also dump quota info
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 21:50:57 +0000 (14:50 -0700)]
rgw: init quota
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 2 Oct 2013 21:37:53 +0000 (14:37 -0700)]
rgw: update quota stats when needed
Now update the quota stats when completing specific operations.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 1 Oct 2013 20:37:56 +0000 (13:37 -0700)]
rgw: also check quota before starting write
In that case we only do it if it's not chunked upload (for which we
don't have the content length).
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 1 Oct 2013 20:32:28 +0000 (13:32 -0700)]
rgw: check quota before completing write op
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 1 Oct 2013 20:32:08 +0000 (13:32 -0700)]
rgw: fix qutoa check
size is in kb
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Tue, 1 Oct 2013 18:45:03 +0000 (11:45 -0700)]
rgw: higher level quota check functionality
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 27 Sep 2013 23:43:10 +0000 (16:43 -0700)]
rgw: more quota implementation
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 27 Sep 2013 21:51:43 +0000 (14:51 -0700)]
rgw: add bucket quota config to entities
Add bucket quota fields to various entities: regionmap (for global
configuration), user info, bucket info.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 27 Sep 2013 20:15:35 +0000 (13:15 -0700)]
Formatter: add dump_bool()
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Fri, 27 Sep 2013 17:41:14 +0000 (10:41 -0700)]
rgw: more quota utility stuff
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 26 Sep 2013 20:24:48 +0000 (13:24 -0700)]
rgw: quota utility class
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Mon, 23 Sep 2013 17:53:54 +0000 (10:53 -0700)]
Merge pull request #618 from dachary/wip-jerasure-minimum-to-decode
ErasureCode: minimum_to_decode unit tests and optimization
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 23 Sep 2013 17:27:47 +0000 (10:27 -0700)]
Merge pull request #605 from dachary/wip-erasure-code-plugin-test
erasure code plugin test
Sage Weil [Mon, 23 Sep 2013 17:26:48 +0000 (10:26 -0700)]
Merge pull request #603 from dachary/wip-erasure-code-example
erasure code example cleanup
Sage Weil [Mon, 23 Sep 2013 17:14:07 +0000 (10:14 -0700)]
Merge pull request #607 from ceph/prctl-getname-test
code_env: use feature test for PR_GET_NAME support
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 23 Sep 2013 17:12:16 +0000 (10:12 -0700)]
Merge pull request #616 from ceph/automake-flags
automake: fixup am_common_flags to avoid re-definitions
Reviewed-by: Roald J. van Loon <roaldvanloon@gmail.com>
Gregory Farnum [Mon, 23 Sep 2013 16:55:32 +0000 (09:55 -0700)]
Merge pull request #614 from ceph/wip-rados-model
Prevent copy_from against yourself, and fix issues with the testing framework around it
Reviewed-by: Greg Farnum <greg@inktank.com>
Sage Weil [Mon, 23 Sep 2013 16:50:10 +0000 (09:50 -0700)]
Merge remote-tracking branch 'gh/next'
Wido den Hollander [Mon, 23 Sep 2013 11:51:26 +0000 (13:51 +0200)]
doc: When bootstrapping mon set the correct caps for client.admin
Otherwise client.admin will be privilege-less and the admin can't
access the cluster.
Loic Dachary [Sun, 22 Sep 2013 08:58:23 +0000 (10:58 +0200)]
ErasureCode: minimum_to_decode unit tests and optimization
The minimum_to_decode function simply returns the available chunks if no
recovery is necessary.
Add unit tests covering all minimum_to_decode situations:
* trying to read nothing
* read a chunk if none are available
* reading a subset of the available chunks
* read a missing chunk if there is less than k chunks available.
* reading when a chunk must be recovered
Signed-off-by: Loic Dachary <loic@dachary.org>
Noah Watkins [Sat, 21 Sep 2013 17:41:15 +0000 (10:41 -0700)]
automake: add per-target AM_CPPFLAGS
Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
Noah Watkins [Sat, 21 Sep 2013 17:18:24 +0000 (10:18 -0700)]
makefile-env: separate cppflags and cflags usage
AM_COMMON_FLAGS included both CPP and C[XX] flags. This can cause
re-definition errors like this:
make[4]: Nothing to be done for `all-am'.
CXX cls/version/cls_version_client.o
<command-line>: warning: "__STDC_FORMAT_MACROS" redefined
because the macro definition appears in both AM_CXXFLAGS and
AM_CPPFLAGS.
Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
João Eduardo Luís [Sat, 21 Sep 2013 11:40:55 +0000 (04:40 -0700)]
Merge pull request #612 from ceph/wip-6361
perfglue/heapprofiler: expect cmd name when handling command instead of 'heap <cmd>'
This was broken by the cli rework.
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Sat, 21 Sep 2013 04:06:09 +0000 (21:06 -0700)]
ceph_test_rados: fix COPY_FROM completion
Fix the copy_from operation to not remove the objects from the in_use list
until after the entire operation is complete. In particular, the racing
read was completing and removing the dest oid from the in-use list before
the copy-from completed. This keeps the model in sync with what the OSD
is actually doing.
If another new read started up, it would grab the previous value from the
model and expect to see that, but would instead see the updated value.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 21 Sep 2013 03:49:55 +0000 (20:49 -0700)]
ceph_test_rados: fix seq_num, improve error output
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 21 Sep 2013 03:44:36 +0000 (20:44 -0700)]
ceph_test_rados: limit max_in_flight to objects/2
In particular, we need to allow all in-flight ops to be COPY_FROM.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 21 Sep 2013 03:43:12 +0000 (20:43 -0700)]
ceph_test_rados: do not COPY_FROM self
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 21 Sep 2013 03:43:00 +0000 (20:43 -0700)]
osd/ReplicatedPG: handle COPY_FROM self
Return EINVAL if we try to COPY_FROM ourselves.
Signed-off-by: Sage Weil <sage@inktank.com>
Dan Mick [Fri, 20 Sep 2013 22:44:34 +0000 (15:44 -0700)]
Merge pull request #592 from ceph/wip-6287
mon/OSDMonitor: make busy creating pgs message more explicit
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Dan Mick [Fri, 20 Sep 2013 22:41:09 +0000 (15:41 -0700)]
Merge pull request #608 from dachary/wip-6357
mon: fix inverted test in osd pool create
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Dan Mick [Thu, 19 Sep 2013 23:04:16 +0000 (16:04 -0700)]
mon: fix wrong arg to "instructed to" status message
Fixes: #6293
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit
16ebb25f7cdb8e92c618a333c505c16edb16c95c )
Yehuda Sadeh [Mon, 16 Sep 2013 21:35:25 +0000 (14:35 -0700)]
rgw: destroy get_obj handle in copy_obj()
Fixes: #6176
Backport: dumpling
We take different code paths in copy_obj, make sure we close the handle
when we exit the function. Move the call to finish_get_obj() out of
copy_obj_data() as we don't create the handle there, so that should
makes code less confusing and less prone to errors.
Also, note that RGWRados::get_obj() also calls finish_get_obj(). For
everything to work in concert we need to pass a pointer to the handle
and not the handle itself. Therefore we needed to also change the call
to copy_obj_data().
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
John Wilkins [Fri, 20 Sep 2013 20:08:06 +0000 (13:08 -0700)]
Merge pull request #613 from ceph/wip-6356
note that ceph-deploy should not be called with sudo on certain situations
Alfredo Deza [Fri, 20 Sep 2013 20:02:28 +0000 (16:02 -0400)]
note that ceph-deploy should not be called with sudo on certain situations
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
Joao Eduardo Luis [Fri, 20 Sep 2013 16:06:30 +0000 (17:06 +0100)]
qa: workunits: cephtool: check if 'heap' commands are parseable
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Joao Eduardo Luis [Fri, 20 Sep 2013 16:50:27 +0000 (17:50 +0100)]
osd: OSD: add 'heap' command to known osd commands array
Must have been forgotten during the cli rework.
Backport: dumpling
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Joao Eduardo Luis [Fri, 20 Sep 2013 15:43:27 +0000 (16:43 +0100)]
mds: MDS: pass only heap profiler commands instead of the whole cmd vector
The heap profiler doesn't care, nor should it, what our command name is.
It only cares about the commands it handles.
Backport: dumpling
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Joao Eduardo Luis [Fri, 20 Sep 2013 15:41:14 +0000 (16:41 +0100)]
perfglue/heap_profiler.cc: expect args as first element on cmd vector
We used to pass 'heap' as the first element of the cmd vector when
handling commands. We haven't been doing so for a while now, so we
needed to fix this.
Not expecting 'heap' also makes sense, considering that what we need to
know when we reach this function is what command we should handle, and
we should not care what the caller calls us when handling his business.
Fixes: #6361
Backport: dumpling
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Samuel Just [Fri, 20 Sep 2013 02:51:46 +0000 (19:51 -0700)]
FileStore: only remove the omap entries if nlink == 1
Fixes: #6359
Introduced in
17c5d765d7c7573f875f6b3ba66e3b6813110a06
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Yehuda Sadeh [Thu, 12 Sep 2013 21:32:17 +0000 (14:32 -0700)]
lru_map: don't use list::size()
replace list::size() with map::size(), which should have
a constant time complexity.
Reviewed-by: Sage Weil <sage@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 12 Sep 2013 21:30:19 +0000 (14:30 -0700)]
common/lru_map: rename tokens to entries
This code was originally used in a token cache, now
as a generic infrastructure rename token fields.
Reviewed-by: Sage Weil <sage@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 12 Sep 2013 19:26:41 +0000 (12:26 -0700)]
bufferlist: don't use list::size()
Reviewed-by: Sage Weil <sage@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Fri, 20 Sep 2013 15:46:29 +0000 (08:46 -0700)]
Merge pull request #590 from ceph/wip-6284
osdc/ObjectCacher: finish contexts after dropping object reference
Tested: ubuntu@teuthology:/a/sage-2013-09-19_19:40:56-fs-wip-6284-testing-basic-plana
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 20 Sep 2013 04:03:36 +0000 (21:03 -0700)]
Makefile: add extra cls DENCODER_DEPS in the cls makefile, not rgw
They belong next to the class. Also, if rgw isn't built, ceph-dencoder
will fail to link.
Fixes: #6352
Tested-by: Haomai Wang <haomaiwang@gmail.com>
Signed-off-by: Sage Weil <sage@inktank.com>
David Zafman [Fri, 20 Sep 2013 01:06:27 +0000 (18:06 -0700)]
Merge pull request #609 from dmick/wip-6293
mon: fix wrong arg to "instructed to" status message
Reviewed-by: David Zafman <david.zafman@inktank.com>
Dan Mick [Thu, 19 Sep 2013 23:04:16 +0000 (16:04 -0700)]
mon: fix wrong arg to "instructed to" status message
Fixes: #6293
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Loic Dachary [Wed, 11 Sep 2013 20:52:20 +0000 (22:52 +0200)]
mon: fix inverted test in osd pool create
When using the properties key=value only, the test was inverted
and an attempt to obtain a substring at index string::npos throws
an exception.
Add variations of osd pool create to qa/workunits/mon/pool_ops.sh
to assert the problem has been fixed and all code paths are used.
http://tracker.ceph.com/issues/6357 fixes #6357
Signed-off-by: Loic Dachary <loic@dachary.org>
Noah Watkins [Sun, 21 Jul 2013 01:41:39 +0000 (18:41 -0700)]
code_env: use feature test for PR_GET_NAME support
Function `get_process_name` has platform specific dependencies. Check
for Linux prctl function and correct command flag.
Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
Yehuda Sadeh [Wed, 18 Sep 2013 17:37:21 +0000 (10:37 -0700)]
rgw: use bufferlist::append() instead of bufferlist::push_back()
push_back() expects char *, whereas append can append a single char.
Appending a NULL char to push_back is cast as a NULL pointer which is
bad.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Gary Lowell [Thu, 19 Sep 2013 20:01:00 +0000 (13:01 -0700)]
Merge pull request #604 from ceph/wip-buck-centos-core
Wip buck centos core
Reviewed-by: Gary Lowell <gary.lowell@inktank.com>
Sage Weil [Thu, 19 Sep 2013 19:08:45 +0000 (12:08 -0700)]
Makefile: fix unittest_arch
Also broken by makefile refactor.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 19 Sep 2013 15:47:09 +0000 (08:47 -0700)]
Makefile: fix unittest_crc32c
This got dropped during the makefile refactor.
Signed-off-by: Sage Weil <sage@inktank.com>
Loic Dachary [Thu, 19 Sep 2013 07:28:14 +0000 (09:28 +0200)]
ErasureCode: complete plugin loader unit tests
* TestErasureCodePluginExample.cc is renamed to TestErasureCodePlugin.cc
because it's not limited to the example which is really used to
support tests rather than being tested.
* Bugous plugins are added to exhibit failures and enable the unit tests
to check they are handled as expected
ErasureCodePluginFailToInitialize : the entry point returns != 0
ErasureCodePluginFailToRegister : the plugin registry is not updated
ErasureCodePluginMissingEntryPoint : the shared library has no entry
point
* It would be difficult to prove that the mutex protecting against
multiple loads actually does what it is expected to because of the
lack of thread introspection functions such as : tell me if this
thread is waiting on this mutex. A simpler approach is chosen : create
a thread that blocks forever when loading ( that's what the delay in
the example plugin is for ) and then check that the lock has indeed
been acquired. Since this mutex is merely about making sure that only
one thread at a time runs this sequence of code, it's probably enough.
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Thu, 19 Sep 2013 07:19:51 +0000 (09:19 +0200)]
ErasureCode: plugin loading in progress flag
The bool loading data member of ErasureCodePluginRegistry is
set to true when a plugin is being loaded, to provide an observable side
effect for test purposes.
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Tue, 17 Sep 2013 08:50:58 +0000 (10:50 +0200)]
ErasureCode: improve API implementation example
* minimum_to_decode and minimum_to_decode_with_cost are replaced with
meaningfull examples instead of placeholders
* encode and decode are commented and hard coded constants are replaced
by defines for readability
* run against valgrind
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Mon, 16 Sep 2013 10:12:52 +0000 (12:12 +0200)]
ErasureCode: proofread abstract API documentation
* Andreas-Joachim Peters suggests to reduce copies to the minimum. When
possible the output arguments will just point to the input
argument. This must be documented as any side effect on the input
argument may modify the output argument
* Fix typos
* Fix may/could/must/should to better reflect what's mandatory and
what's not.
* Reword the explanation of minimum_to_decode_with_cost to not suggest
an implementation. This will need to be revisited anyway, when the
semantic of the cost is defined.
* Explain chunk size constraints
Signed-off-by: Loic Dachary <loic@dachary.org>
Sage Weil [Thu, 19 Sep 2013 03:25:09 +0000 (20:25 -0700)]
os/FileStore: fix uninitialized var
Signed-off-by: Sage Weil <sage@inktank.com>
Yan, Zheng [Thu, 12 Sep 2013 02:36:39 +0000 (10:36 +0800)]
osdc/ObjectCacher: finish contexts after dropping object reference
The context to finish can be class C_Client_PutInode, which may drop
inode's last reference. So we should first drop object's reference,
then finish contexts.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Sage Weil [Wed, 18 Sep 2013 19:31:42 +0000 (12:31 -0700)]
doc/release-notes: v0.69
Signed-off-by: Sage Weil <sage@inktank.com>
Gary Lowell [Wed, 18 Sep 2013 09:11:02 +0000 (09:11 +0000)]
Merge branch 'next'
Gary Lowell [Wed, 18 Sep 2013 01:40:51 +0000 (01:40 +0000)]
v0.69
Gregory Farnum [Tue, 17 Sep 2013 18:14:26 +0000 (11:14 -0700)]
Merge pull request #586 from ceph/wip-copyfrom-big
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Wed, 11 Sep 2013 22:57:00 +0000 (15:57 -0700)]
ceph_test_rados: identify write seq_num in output
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 11 Sep 2013 17:27:49 +0000 (10:27 -0700)]
librados: test copy_from without src_version specified
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 11 Sep 2013 22:11:35 +0000 (15:11 -0700)]
osd: allow a copy_from without knowing the src_version
If we don't specify the version up front, learn the version after the first
chunk and enforce it thereafter to ensure we do not get torn content.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 10 Sep 2013 23:30:25 +0000 (16:30 -0700)]
ceph_test_rados: submit a racing read with every COPY_FROM
Verify that the racing read completes after the COPY_FROM does (i.e., is
blocked by it).
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 10 Sep 2013 23:29:48 +0000 (16:29 -0700)]
ceph_test_rados: use get_version64() throughout
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 11 Sep 2013 22:09:59 +0000 (15:09 -0700)]
osd: block requests on object during COPY_FROM
Block any request on an object (read or write) during the COPY_FROM
operation.
This could potentially be broken down into read vs write operations without
much difficulty, but blocking any op indescriminately is sufficient for
now, so let's keep it simple.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 11 Sep 2013 22:10:47 +0000 (15:10 -0700)]
osd: add infrastructure to block io on an obc
Add an is_blocked() method for the obc, and add infrastructure to block
any operations if it returns true. Clean up on_change(), and add a helper
to kick an obc when whatever condition leading to it being blocked is no
longer true.
For now, is_blocked() is always false...
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 9 Sep 2013 19:47:03 +0000 (12:47 -0700)]
osd/ReplicatedPG: factor some bits into finish_copy
This is a bit cleaner and hopefully more usable for osd-driven cache
promotion.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 5 Sep 2013 23:09:13 +0000 (16:09 -0700)]
ceph_test_rados: more readable output
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 5 Sep 2013 00:09:52 +0000 (17:09 -0700)]
osd/ReplicatedPG: stage object chunks to replicas during COPY_FROM
As we get each chunk of data during the COPY_FROM operation, write it out
to a temporary object on the replicas. When we get all the pieces, move
it into place.
Signed-off-by: Sage Weil <sage@inktank.com>
John Wilkins [Tue, 17 Sep 2013 17:39:24 +0000 (10:39 -0700)]
doc: Replaced code-block with inline literal to fix auto numbering.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins [Tue, 17 Sep 2013 17:34:47 +0000 (10:34 -0700)]
doc: Made s3gw.fcgi paths consistent.
fixes: #6182
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
Alexandre Oliva [Tue, 17 Sep 2013 07:23:32 +0000 (04:23 -0300)]
osd: compute full ratio from kb_avail
On btrfs, kb_used + kb_avail can be much smaller than total kb, and
what really matters to avoid filling up the disk is how much space is
available, not how much we've used. Thus, compute the ratio we use to
determine full or nearfull from kb_avail rather than from kb_used.
Signed-off-by: Alexandre Oliva <oliva@gnu.org>
Signed-off-by: Sage Weil <sage@inktank.com>
Loic Dachary [Tue, 17 Sep 2013 08:50:58 +0000 (10:50 +0200)]
ErasureCode: improve API implementation example
* minimum_to_decode and minimum_to_decode_with_cost are replaced with
meaningfull examples instead of placeholders
* encode and decode are commented and hard coded constants are replaced
by defines for readability
* run against valgrind
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Mon, 16 Sep 2013 10:12:52 +0000 (12:12 +0200)]
ErasureCode: proofread abstract API documentation
* Andreas-Joachim Peters suggests to reduce copies to the minimum. When
possible the output arguments will just point to the input
argument. This must be documented as any side effect on the input
argument may modify the output argument
* Fix typos
* Fix may/could/must/should to better reflect what's mandatory and
what's not.
* Reword the explanation of minimum_to_decode_with_cost to not suggest
an implementation. This will need to be revisited anyway, when the
semantic of the cost is defined.
* Explain chunk size constraints
Signed-off-by: Loic Dachary <loic@dachary.org>
Joe Buck [Mon, 16 Sep 2013 23:37:58 +0000 (16:37 -0700)]
Remove unneeded junit4 check
We refactored the java tests to not use
the ExternalResource class, so this test
is no longer necessary.
Signed-off-by: Joe Buck <jbbuck@gmail.com>
Joe Buck [Sat, 14 Sep 2013 00:41:31 +0000 (17:41 -0700)]
Removing extraneous code
The ExternalResource code was unnecessary and caused
issues on CentOS. Removing it.
Update Makefile.am to reflect the fact that
an anonymous class was removed and its
$1.class file is no longer generated.
Signed-off-by: Joe Buck <jbbuck@gmail.com>
Joe Buck [Thu, 12 Sep 2013 20:14:28 +0000 (13:14 -0700)]
Use a loop for testing jdk paths
Loop through a list of sensible default
locations for a JDK, stopping if a
workable JDK is found.
Also, add support for CentOS' default
java location.
Signed-off-by: Joe Buck <jbbuck@gmail.com>