| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
This will let us pass in an hobject_t::hash directly (for example) without
rehashing a string.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| |
|
|
|
|
|
|
|
|
| |
This field will track the most recent eversion recorded in an archived
bloom_filter. Any events after this point in the pg log have not yet been
archived and should therefore be included in the in-memory bloom filter.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
|
|
|
| |
Add pool properties to control what type of bloom filter the OSD should track
for each PG.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fun facts:
- fpp = false positive probability
- fpp is a function of insert count only
- at .1% fpp, we pay about 2 bytes per insert
- at 1-2% fpp, we pay about 1 byte per insert
- at 15% fpp, we pay about .5 bytes per insert
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
|
|
|
|
| |
We pass the ctor our target behavior and calculate parameters based on
that. Avoid storing the target behavior, and make that calc a static
method. And add a new ctor that takes the parameters explicitly.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
|
|
| |
3 space tabs.. blech. Let's not change it now, though.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
|
|
| |
Also broken by makefile refactor.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
|
|
| |
This got dropped during the makefile refactor.
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|
|
|
| |
Signed-off-by: Sage Weil <sage@inktank.com>
|
|\ |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| | |
When using replica log, if the log pool doesn't exist all operations are
going to fail. Try to create it if doesn't exist.
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Fixes: #6175
Backport: dumpling
We get a buffer off the remote gateway which might
not be NULL terminated. The JSON parser needs the
buffer to be NULL terminated even though we provide
a buffer length as it calls strlen().
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Fixes: #6286
Use an external counter instead of calling list::size()
Reviewed-by: Sage Weil <sage@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Fixes: #6268
When doing aio write of objects (either regular or multipart parts) we
need to drain pending aio requests. Otherwise if gateway goes down then
object might end up corrupted.
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Fixes: #6214
When getting a failed read from client when putting an object
we returned the wrong value (always 0), which in the chunked-
upload case ended up in assuming that the write was done
successfully.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
|
| |\
| | |
| | |
| | |
| | | |
qa: workunits: mon: crush_ops: test 'ceph osd crush move'
Reviewed-by: Sage Weil <sage@inktank.com>
|
| |/
| |
| |
| | |
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
|
| |\
| | |
| | |
| | |
| | | |
mon: MonCommands: expect a CephString as 1st arg for 'osd crush move'
Reviewed-by: Sage Weil <sage@inktank.com>
|
| |/
| |
| |
| |
| |
| | |
Fixes: #6230
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
|
|\ \
| | |
| | |
| | | |
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
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>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
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>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
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>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
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>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This is a bit cleaner and hopefully more usable for osd-driven cache
promotion.
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
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>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
If the repop has no version set, skip the updates to last_update and
last_update_{applied,ondisk} and last_complete_ondisk.
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Allow us to mark when we start and stop using a temporary object in a
sub_op. If we start to use it, make sure the collection exists on the
replica.
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | | |
These are cruft from the old parallelexec mode and can be removed.
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | | |
This is unused and references otherwise unused variables we can remove!
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This is similar to a collection_add + collection_move sequence in that we
apply the same replay guards. The difference is that we roll it up into
a single operation, change the filename, and make the omap content carry
over by calling DBObjectMap->clone (as there is no rename function or
collection awareness in the DBObjectMap).
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | | |
Add method to move an object between collections *and* change its name.
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | | |
For now, the only caller leaves the name the same.
Signed-off-by: Sage Weil <sage@inktank.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
|
| | |
| | |
| | |
| | |
| | |
| | | |
fixes: #6182
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
|