ceph_test_objectstore: add omap iterator test

Signed-off-by: Sage Weil <sage@redhat.com>

ceph-objectstore-tool: add fsck command

Signed-off-by: Sage Weil <sage@redhat.com>

os/ObjectStore: add fsck to interface

Only bluestore and kstore implement this currently.

Signed-off-by: Sage Weil <sage@redhat.com>

os/kstore: clear onode on _do_remove

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: clear onode in _do_remove

Signed-off-by: Sage Weil <sage@redhat.com>

osd/PGBackend: fix omap digest error message

Print the *omap* digest, not *data* digest.

Signed-off-by: Sage Weil <sage@redhat.com>

os: remove {get,set}_allow_sharded_objects from interface

We've already forced everyone to upgrade through hammer, so everyone
supports this.  Just unconditionally set the feature if it is not set
(for consistency's sake).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: do not flush metadata on flush()

That's what fsync is for.  Moreover, this can lead to some squirreliness
if we trigger this from _flush_log().

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: helpful error when aio cannot init

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/FreelistManager: audit

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: implement invalidate_cache

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: add invalidate_cache

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: simplify _do_remove

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: debug msg on statfs

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: optimize _dump_onode slightly

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: enable bluestore debug options

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fsck_on_umount

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/StupidAllocator: bluestore_debug_small_allocations

Force small allocations for debugging purposes.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix _zero when previous extent partially unwritten

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: support copy-on-write clones

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: only allow clone if hash matches

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: Enode infrastructure

Enodes will track extent ref counts for any extent
that is marked shared.  There will be an enode for
any unique hash value that has any refs.  We will keep
in-memory copies of only those Enodes that are
referenced by in-memory Onodes, and only if the enode
is requested (e.g., the enode won't be loaded as a
result of an object read because we never need to
call get_enode.).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/bluestore_types: add extent FLAG_COW_{HEAD,TAIL}

Signed-off-by: Sage Weil <sage@redhat.com>

unittest_bluefs, unittest_bluestore_types

These should run during make check.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/bluestore_types: add contains(), clear(), empty() to extent_ref_map

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: wal_op_t::OP_COPY

Assume block-aligned.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: fix waiter wakeup use-after-free race

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: add bluestore_debug_no_reuse_blocks

This makes debugging a bit easier because we never use the same
extent of the disk twice, leaving useful evidence behind.

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: do Synthetic tests over larger objects

400k for objects.

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: use a few hash values for objects; clone between them

We only guarantee support for clone between objects with the same hash.

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: dump actual vs expected on read data mismatch

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: add many clone tests

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: validate full object contents after writes

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: debug enter/exit points

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: save map lookups for a few ops

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: fix locking for a few ops

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: fix clone

Copy the buffer, in case other threads modify it in place.

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: simplify object name generation

The long names don't exercise useful code paths, and having
consistent naming makes it easier to grep through logs.

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: clone non-empty objects, not empty ones

This condition was backwards.

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: clone objects with same hash

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: add some slow debug path

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: clean up comments a bit

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: note wal releases in fsck

Include these in used_blocks (they are about to be released but
not reflected in the onode).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix read bug when there is a hole

Signed-off-by: Sage Weil <sage@redhat.com>

os/kstore: fix rename

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix rename

Install a negative onode entry at the old name position.
Otherwise, a simple transaction like

 rename a -> b
 touch b

will re-read the old b onode key on the second op, and chaos will
ensue (e.g., because it'll reference the same extents from a
different object).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: remove unused OnodeMap::remove

We install negative entries instead.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: adjust debug output

5 helpful (read/write offsets)
10 more, with aio completions
20 everything
30 fire hose
40 data hexdumps

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: fix path

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: do WAL ops buffered to avoid RMW issues

We may have multiple WAL ops that do read/modify/write covering
the same blocks.  To avoid the complexity of identifying those
situations and ensuring that we, say, wait for writes to complete
before reading them back again, just make the IO buffered and let
the page cache handle that for us.

This fixes the failure of LibRadosAio.RoundTripWriteFull.

Signed-off-by: Sage Weil <sage@redhat.com>

rocksdb: debug log writes/reads

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: handle both buffered and direct+async IO

Prefer aio unless explicitly directed otherwise.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: rename bdev options

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: use BlueFS::get_usage()

...just so we log bdev utilization in the log.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: get_usage()

Return (and log) usage for all bdevs.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: do not dirty file when overwriting bytes

The rocksdb log recycle option allows us to overwrite previously
allocated space in an old log file to avoid updating the file
metadata on normal file systems.  Take advantage of that here to
by implementing what is effectively O_NOCMTIME semantics: we do
not dirty the file metadata just because mtime is updated.
Instead, we dirty the file only if we allocate new space or if
the size has to be increased.

Note that on my NVME drive a single-thread rados bench test, we
jump from 30MB/sec to 50MB/sec 128KB writes as soon as we start
recycling previous logs (about 40 second into the run).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: ignore flush when buffer is small

Rocksdb does a flush after every append, each of which is often
less than a full block.  This is very inefficient when our
_flush() will send that to disk (and block).

Avoid this most of the time by ignoring small flush requests
entirely, unless the force flag is set (e.g., by fsync).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: update freelist in individual transactions

We submit each operation's transaction individually to rocksdb,
and then since a final transction to flush them all.  However,
they may not commit atomically (all together), which means we
need to leave the individual freelist updates within each
transaction.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: better debugging on fsck alloc errors

Signed-off-by: Sage Weil <sage@redhat.com>

script/crash_bdev: simple script to inject bdev failures

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: fail mount of fsck finds errors

Signed-off-by: Sage Weil <sage@redhat.com>

os/fs/FS.h: fix aio_t::pread

Allocate aligned buffer.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: better error msg for bdev label check

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: don't create block.{db,wal} by default

Signed-off-by: Sage Weil <sage@redhat.com>

vstart.sh: less noisy debug

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: fix fsck contains vs intersects

Any overlap is an error.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: bluestore bluefs = true

Signed-off-by: Sage Weil <sage@redhat.com>

rpm, debian: package ceph-bluefs-tool

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix error path if label set fails

Reported-by: David Zafman <dzafman@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>

rocksdb: fix recycle replay

Signed-off-by: Sage Weil <sage@redhat.com>

Makefile-rocksdb.am: update

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: default to 64k min_alloc_size

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix _open_bdev() failure path

Signed-off-by: Sage Weil <sage@redhat.com>

kv/RocksDBStore: behave if options string is empty

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: clear coll_map on umount, fsck finish

Signed-off-by: Sage Weil <sage@redhat.com>

os/kstore/KStore: fix object key decode with key

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix object key decode with key

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_objectstore_test: fix warning

Signed-off-by: Sage Weil <sage@redhat.com>

os/KeyValueStore: drop kinetic #include

Signed-off-by: Sage Weil <sage@redhat.com>

os/kstore: add new KStore backend

This is based on BlueStore, but with all of the block-related code
and complexity ripped out, and a simple striping strategy added
in its place.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/bluestore_types: localize types

Prefix with bluestore_

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: add extent_ref_map_t

This will be used to refcount extents for some subset
of the store (objects with same name or hash value?).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/FreelistManager: drop unused db ref

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: record kv backend

Record kv backend at mkfs time instead of relying on current value
of config option.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: statfs

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: inject block failures

Signed-off-by: Sage Weil <sage@redhat.com>

ceph_test_objectstore: clean up synthetic collections

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: block.db support

Support a mid- to fast device that will preferentially
store the rocksdb data (and wal, if block.wal is not
present).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: less debug noise

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: all overwrites on open_for_write

rocksdb will occasionally overwrite an existing file
if it is not present/valid in the manifest.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: drop internal EnvMirror

Signed-off-by: Sage Weil <sage@redhat.com>

rocksdb: pull up to master, include EnvMirror

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: label all block devices

Label all of our block devices with a simple label
that includes the osd_uuid.  Wire this into the
ObjectStore and OSD probe mechanism.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: flush log if needed

If a file has dirty metadata (but no dirty data), we
still need to flush the log when it is flushed.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: fix replay of unlink

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: support second block.wal device

Use this device for the bluefs log.

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore: fix zero gap bug

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore: disable overlay for now

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlockDevice: restructure interface

use atomics, do not track in-flight extents or magically cope
with racing ios (that is the users responsibility).

Signed-off-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueFS: fix overwrite

Signed-off-by: Sage Weil <sage@redhat.com>