The user sstables manager will need to provide endpoint config for
sstables' storage drivers. For that it needs to get it from db::config
and keep in-sync with its updates.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
This series fixes a few issues caused by f1bbf705f9
(f1bbf705f9):
- table, compaction_manager: prevent cross shard access to owned_ranges_ptr
- Fixes#13631
- distributed_loader: distribute_reshard_jobs: pick one of the sstable shard owners
- compaction: make_partition_filter: do not assert shard ownership
- allow the filtering reader now used during resharding to process tokens owned by other shards
Closes#13635
* github.com:scylladb/scylladb:
compaction: make_partition_filter: do not assert shard ownership
distributed_loader: distribute_reshard_jobs: pick one of the sstable shard owners
table, compaction_manager: prevent cross shard access to owned_ranges_ptr
This PR introduces an experimental feature called "tablets". Tablets are
a way to distribute data in the cluster, which is an alternative to the
current vnode-based replication. Vnode-based replication strategy tries
to evenly distribute the global token space shared by all tables among
nodes and shards. With tablets, the aim is to start from a different
side. Divide resources of replica-shard into tablets, with a goal of
having a fixed target tablet size, and then assign those tablets to
serve fragments of tables (also called tablets). This will allow us to
balance the load in a more flexible manner, by moving individual tablets
around. Also, unlike with vnode ranges, tablet replicas live on a
particular shard on a given node, which will allow us to bind raft
groups to tablets. Those goals are not yet achieved with this PR, but it
lays the ground for this.
Things achieved in this PR:
- You can start a cluster and create a keyspace whose tables will use
tablet-based replication. This is done by setting `initial_tablets`
option:
```
CREATE KEYSPACE test WITH replication = {'class': 'NetworkTopologyStrategy',
'replication_factor': 3,
'initial_tablets': 8};
```
All tables created in such a keyspace will be tablet-based.
Tablet-based replication is a trait, not a separate replication
strategy. Tablets don't change the spirit of replication strategy, it
just alters the way in which data ownership is managed. In theory, we
could use it for other strategies as well like
EverywhereReplicationStrategy. Currently, only NetworkTopologyStrategy
is augmented to support tablets.
- You can create and drop tablet-based tables (no DDL language changes)
- DML / DQL work with tablet-based tables
Replicas for tablet-based tables are chosen from tablet metadata
instead of token metadata
Things which are not yet implemented:
- handling of views, indexes, CDC created on tablet-based tables
- sharding is done using the old method, it ignores the shard allocated in tablet metadata
- node operations (topology changes, repair, rebuild) are not handling tablet-based tables
- not integrated with compaction groups
- tablet allocator piggy-backs on tokens to choose replicas.
Eventually we want to allocate based on current load, not statically
Closes#13387
* github.com:scylladb/scylladb:
test: topology: Introduce test_tablets.py
raft: Introduce 'raft_server_force_snapshot' error injection
locator: network_topology_strategy: Support tablet replication
service: Introduce tablet_allocator
locator: Introduce tablet_aware_replication_strategy
locator: Extract maybe_remove_node_being_replaced()
dht: token_metadata: Introduce get_my_id()
migration_manager: Send tablet metadata as part of schema pull
storage_service: Load tablet metadata when reloading topology state
storage_service: Load tablet metadata on boot and from group0 changes
db, migration_manager: Notify about tablet metadata changes via migration_listener::on_update_tablet_metadata()
migration_notifier: Introduce before_drop_keyspace()
migration_manager: Make prepare_keyspace_drop_announcement() return a future<>
test: perf: Introduce perf-tablets
test: Introduce tablets_test
test: lib: Do not override table id in create_table()
utils, tablets: Introduce external_memory_usage()
db: tablets: Add printers
db: tablets: Add persistence layer
dht: Use last_token_of_compaction_group() in split_token_range_msb()
locator: Introduce tablet_metadata
dht: Introduce first_token()
dht: Introduce next_token()
storage_proxy: Improve trace-level logging
locator: token_metadata: Fix confusing comment on ring_range()
dht, storage_proxy: Abstract token space splitting
Revert "query_ranges_to_vnodes_generator: fix for exclusive boundaries"
db: Exclude keyspace with per-table replication in get_non_local_strategy_keyspaces_erms()
db: Introduce get_non_local_vnode_based_strategy_keyspaces()
service: storage_proxy: Avoid copying keyspace name in write handler
locator: Introduce per-table replication strategy
treewide: Use replication_strategy_ptr as a shorter name for abstract_replication_strategy::ptr_type
locator: Introduce effective_replication_map
locator: Rename effective_replication_map to vnode_effective_replication_map
locator: effective_replication_map: Abstract get_pending_endpoints()
db: Propagate feature_service to abstract_replication_strategy::validate_options()
db: config: Introduce experimental "TABLETS" feature
db: Log replication strategy for debugging purposes
db: Log full exception on error in do_parse_schema_tables()
db: keyspace: Remove non-const replication strategy getter
config: Reformat
It's meant to be used in places where currently
get_non_local_strategy_keyspaces() is used, but work only with
keyspaces which use vnode-based replication strategy.
Will be used by tablet-based replication strategies, for which
effective replication map is different per table.
Also, this patch adapts existing users of effective replication map to
use the per-table effective replication map.
For simplicity, every table has an effective replication map, even if
the erm is per keyspace. This way the client code can be uniform and
doesn't have to check whether replication strategy is per table.
Not all users of per-keyspace get_effective_replication_map() are
adapted yet to work per-table. Those algorithms will throw an
exception when invoked on a keyspace which uses per-table replication
strategy.
this is the first step to the uuid-based generation identifier. the goal is to encapsulate the generation related logic in generator, so its consumers do not have to understand the difference between the int64_t based generation and UUID v1 based generation.
this commit should not change the behavior of existing scylla. it just allows us to derive from `generation_generator` so we can have another generator which generates UUID based generation identifier.
Closes#13073
* github.com:scylladb/scylladb:
replica, test: create generation id using generator
sstables: add generation_generator
test: sstables: use generate_n for generating ids for testing
The only reason why it's there (right next to compaction_fwd.hh) is
because the database::table_truncate_state subclass needs the definition
of compaction_manager::compaction_reenabler subclass.
However, the former sub is not used outside of database.cc and can be
defined in .cc. Keeping it outside of the header allows dropping the
compaction_manager.hh from database.hh thus greatly reducing its fanout
over the code (from ~180 indirect inclusions down to ~20).
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Closes#13622
Seen after f1bbf705f9 in debug mode
distributed_loader collect_all_shared_sstables copies
compaction::owned_ranges_ptr (lw_shared_ptr<const
dht::token_range_vector>)
across shards.
Since update_sstable_cleanup_state is synchronous, it can
be passed a const refrence to the token_range_vector instead.
It is ok to access the memory read-only across shards
and since this happens on start-up, there are no special
performance requirements.
Fixes#13631
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
reuse generation_generator for generating generation identifiers for
less repeatings. also, add allow update generator to update its
lastest known generation id.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Caller already has it to call mutate_MV() on. The method in question
will need the generator in one of the next patches.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
There's a legacy safety check in view code that needs to find a base table from its schema ID. To do it it calls for global storage proxy instance. The comment says that this code can be removed once computes_column feature is known by everyone. I'm not sure if that's the case, so here's more complicated yet less incompatible way to stop using global proxy instance.
Closes#13504
* github.com:scylladb/scylladb:
view: Remove unused view_ptr reference
view: Carry backing-secondary-index bit via view builder
view: Keep backing-seconday-index bool on value_getter
table: Add const index manager sgetter
enable_optimized_twcs_queries is specific to TWCS, therefore it
belongs to TWCS, not replica::table.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closes#13489
So it can be used in the next patch that will refactor
compaction_state out of class compaction_manager.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
update_sstable_cleanup_state calls needs_cleanup and
inserts (or erases) the sstable into the respective
compaction_state.sstables_requiring_cleanup set.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
That's courtersy of 153813d3b8, which annotates Seastar smart pointer classes with Clang's consumed attributes, to help Clang to statically spot use-after-move bugs.
Closes#13386
* github.com:scylladb/scylladb:
replica: Fix use-after-move in table::make_streaming_reader
index/built_indexes_virtual_reader.hh: Fix use-after-move
db/view/build_progress_virtual_reader: Fix use-after-move
sstables: Fix use-after-move when making reader in reverse mode
Variant used by
streaming/stream_transfer_task.cc: , reader(cf.make_streaming_reader(cf.schema(), std::move(permit_), prs))
as full slice is retrieved after schema is moved (clang evaluates
left-to-right), the stream transfer task can be potentially working
on a stale slice for a particular set of partitions.
static report:
In file included from replica/dirty_memory_manager.cc:6:
replica/database.hh:706:83: error: invalid invocation of method 'operator->' on object 'schema' while it is in the 'consumed' state [-Werror,-Wconsumed]
return make_streaming_reader(std::move(schema), std::move(permit), range, schema->full_slice());
Fixes#13397.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Preparing for #10459, this series defines sstables::generation_type::int_t
as `int64_t` at the moment and use that instead of naked `int64_t` variables
so it can be changed in the future to hold e.g. a `std::variant<int64_t, sstables::generation_id>`.
sstables::new_generation was defined to generation new, unique generations.
Currently it is based on incrementing a counter, but it can be extended in the future
to manufacture UUIDs.
The unit tests are cleaned up in this series to minimize their dependency on numeric generations.
Basically, they should be used for loading sstables with hard coded generation numbers stored under `test/resource/sstables`.
For all the rest, the tests should use existing and mechanisms introduced in this series such as generation_factory, sst_factory and smart make_sstable methods in sstable_test_env and table_for_tests to generate new sstables with a unique generation, and use the abstract sst->generation() method to get their generation if needed, without resorting the the actual value it may hold.
Closes#12994
* github.com:scylladb/scylladb:
everywhere: use sstables::generation_type
test: sstable_test_env: use make_new_generation
sstable_directory::components_lister::process: fixup indentation
sstables: make highest_generation_seen return optional generation
replica: table: add make_new_generation function
replica: table: move sstable generation related functions out of line
test: sstables: use generation_type::int_t
sstables: generation_type: define int_t
The wasm engine is moved from replica::database to the query_processor.
The wasm instance cache and compilation thread runner were already there,
but now they're also initialized in the query_processor constructor.
By moving the initialization to the constructor, we can now
be certain that all wasm-related objects (wasm instance cache,
compilation thread runner, and wasm engine, which was already
passed in the constructor) are initialized when we try to use
them because we have to use the query processor to access them
anyway.
The change is also motivated by the fact that we're planning
to take Wasm UDFs out of experimental, after which they should
stop getting special treatment.
Closes#13311
* github.com:scylladb/scylladb:
wasm: move wasm initialization to query_processor constructor
wasm: return wasm instance cache as a reference instead of a pointer
wasm: move wasm engine to query_processor
The method is called by view_builder::consumer when building a view and
the consumer already has stable dependency reference on the view updates
generator.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
The latter is the place where mutate_MV is called and it needs the
view updates generator nearby.
The call-stack starts at database::do_apply(). As was described in one
of the previous patches, applying mutations that need updating views
happen late enough, so if the view updates generator is not plugged to
the database yet, it's OK to bail out with exception. If it's plugged,
it's carried over thus keeping the generator instance alive and waited
for on its stop.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
The database is low-level service and currently view update generator
implicitly depend on it via storage proxy. However, database does need
to push view updates with the help of mutate_MV helper, thus adding the
dependency loop.
This patch exploits the fact that view updates start being pushed late
enough, by that time all other service, including proxy and view update
generator, seem to be up and running. This allows a "weak dependency"
from database to view update generator, like there's one from database
to system keyspace already.
So in this patch the v.u.g. puts the shared-from-this pointer onto the
database at the time it starts. On stop it removes this pointer after
database is drained and (hopefully) all view updates are pushed.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
The wasm engine is used for compiling and executing Wasm UDFs, so
the query_processor is a more appropriate location for it than
replica::database, especially because the wasm instance cache
and the wasm alien thread runner are already there.
This patch also reduces the number of wasm engines to 1, shared by
all shards, as recommended by the wasmtime developers.
We need this so that we can have multi-partition mutations which are applied atomically. If they live on different shards, we can't guarantee atomic write to the commitlog.
Fixes: #12642Closes#13134
* github.com:scylladb/scylladb:
test_raft_upgrade: add a test for schema commit log feature
scylla_cluster.py: add start flag to server_add
ServerInfo: drop host_id
scylla_cluster.py: add config to server_add
scylla_cluster.py: add expected_error to server_start
scylla_cluster.py: ScyllaServer.start, refactor error reporting
scylla_cluster.py: fix ScyllaServer.start, reset cmd if start failed
raft: check if schema commitlog is initialized Refuse to boot if neither the schema commitlog feature nor force_schema_commit_log is set. For the upgrade procedure the user should wait until the schema commitlog feature is enabled before enabling consistent_cluster_management.
raft: move raft initialization after init_system_keyspace
database: rename before_schema_keyspace_init->maybe_init_schema_commitlog
raft: use schema commitlog for raft tables
init_system_keyspace: refactoring towards explicit load phases
We are going to move the raft tables from the first
load phase to the second. This means the second
init_system_keyspace call will load raft tables along
with the schema, making the name of this function imprecise.
We aim (#12642) to use the schema commit log
for raft tables. Now they are loaded at
the first call to init_system_keyspace in
main.cc, but the schema commitlog is only
initialized shortly before the second
call. This is important, since the schema
commitlog initialization
(database::before_schema_keyspace_init)
needs to access schema commitlog feature,
which is loaded from system.scylla_local
and therefore is only available after the
first init_system_keyspace call.
So the idea is to defer the loading of the raft tables
until the second call to init_system_keyspace,
just as it works for schema tables.
For this we need a tool to mark which tables
should be loaded in the first or second phase.
To do this, in this patch we introduce system_table_load_phase
enum. It's set in the schema_static_props for schema tables.
It replaces the system_keyspace::table_selector in the
signature of init_system_keyspace.
The call site for populate_keyspace in init_system_keyspace
was changed, table_selector.contains_keyspace was replaced with
db.local().has_keyspace. This check prevents calling
populate_keyspace(system_schema) on phase1, but allows for
populate_keyspace(system) on phase2 (to init raft tables).
On this second call some tables from system keyspace
(e.g. system.local) may have already been populated on phase1.
This check protects from double-populating them, since every
populated cf is marked as ready_for_writes.
It is possible to find no generation in an empty
table directory, and in he future, with uuid generations
it'd be possible to find no numeric generations in the
directory.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
make_new_generation generates a new generation
from an optional one.
If disengaged, it just generates a new generation
based on the shard_id. Otherwise, it generates
the next generation in sequence by adding
smp::count to the previous value, like we do today.
In the future, with uuid-based generations, the
function could be used to generate a new random
uuid based on the optional parameter.
It will be up to the caller, e.g. replica::table or
sstables manager to decide which kind of generation to
create.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
updating the highest generation happens only during
startup and creating sstables is done rarely enough
there is no reason to inline either functions.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Convert all users to use sstables::generation_type::int_t.
Further patches will continue to convert most to
using sstables::generation_type instead so we can
abstract the value type.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
And propagate it down to where it is created. This will be used to add
trace points for semaphore related events, but this will come in the
next patches.
This series cleans up unit test in preparation for PR #12994.
Helpers are added (or reused) to not rely on specific sstable generation numbers where possible (other than loading reference sstables that are committed to the repo with given generation numbers), and to generate the sstables for tests easily, taking advantage of generation management in `sstable_test_env`, `table_for_tests`, or `replica::table` itself.
Closes#13242
* github.com:scylladb/scylladb:
test: add verify_mutation helpers.
test: add make_sstable_containing memtable
test: table_for_tests: add make_sstable function
test: sstable_test_env: add make_sst_factory methods
test: sstable_compaction_test: do not rely on specific generations
tests: use make_sstable defaults as much as possible
test: sstable_test_env: add make_table_for_tests
test: sstable_datafile_test: do not rely on sepecific sstable generations
test: sstable_test_env: add reusable_sst(shared_sstable)
sstable: expose get_storage function
test: mutation_reader_test: create_sstable: do not rely on specific generations
test: mutation_reader_test: do_test_clustering_order_merger_sstable_set: rely on test_envsstable generation
test: mutation_reader_test: combined_mutation_reader_test: define a local sst_factory function
test: mutation_reader_test: do not use tmpdir
test: use big format by default
test: sstable_compaction_test: use highest sstable version by default
test: test_env: make_db_config: set cfg host_id
test: sstable_datafile_test: fixup indentation
test: sstable_datafile_test: various tests: do_with_async
test: sstable_3_x_test: validate_read, sstable_assertions: get shared_sstable
test: sstable_3_x_test: compare_sstables: get shared_sstable
test: sstable_3_x_test: write_sstables: return shared_sstable
test: sstable_3_x_test: write, compare, validate_sstables: use env.tempdir
test: sstable_3_x_test: compacted_sstable_reader: do not reopen compacted_sst
test: lib: test_services: delete now unused stop_and_keep_alive
test: sstable_compaction_test: use deferred_stop to stop table_for_tests
test: sstable_compaction_test: compound_sstable_set_incremental_selector_test: do_with_async
test: sstable_compaction_test: sstable_needs_cleanup_test: do_with_async
test: sstable_compaction_test: leveled_05: fixup indentation
test: sstable_compaction_test: leveled_05: do_with_async
test: sstable_compaction_test: compact_02: do_with_async
test: sstable_compaction_test: compact_sstables: simplify variable allocation
test: sstable_compaction_test: compact_sstables: reindent
test: sstable_compaction_test: compact_sstables: use thread
test: sstable_compaction_test: sstable_rewrite: simplify variable allocation
test: sstable_compaction_test: sstable_rewrite: fixup indentation
test: sstable_compaction_test: sstable_rewrite: do_with_async
test: sstable_compaction_test: compact: fixup indentation
test: sstable_compaction_test: compact: complete conversion to async thread
test: sstable_compaction_test: compaction_manager_basic_test: rename generations to idx
table_for_tests uses a sstables manager to generate sstables
and gets the new generation from
table.calculate_generation_for_new_table().
The version to use is either the highest supported or
an ad-hoc version passed to make_sstable.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Tables need to know which storage their sstables need to be located at,
so class table needs to have itw reference of the storage options. The
thing can be inherited from the keyspace metadata.
Tests sometimes create table without keyspace at hand. For those use
default-initialized storage options (which is local storage).
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
There was some logic to call mark_as_populate at
the appropriate places, but the _populated field
and the ensure_populated function were
not used by anyone.
they are part of the CQL type system, and are "closer" to types.
let's move them into "types" directory.
the building systems are updated accordingly.
the source files referencing `types.hh` were updated using following
command:
```
find . -name "*.{cc,hh}" -exec sed -i 's/\"types.hh\"/\"types\/types.hh\"/' {} +
```
the source files under sstables include "types.hh", which is
indeed the one located under "sstables", so include "sstables/types.hh"
instea, so it's more explicit.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closes#12926
There's the distribtued_loader::populate_column_family() helper that manages sstables on their way towards table on boot. The method naturally belongs the the table_population_metadata -- a helper class that in fact prepares the ground for the method in question.
This PR moves the method into metadata class and removes whole lot of extra alias-references and private-fields exporting methods from it. Also it keeps start_subdir and populate_c._f. logic close to each other and relaxes several excessive checks from them.
Closes#12847
* github.com:scylladb/scylladb:
distributed_loader: Rename table_population_metadata
distributed_loader: Dont check for directory presense twice
distributed_loader: Move populate calls into metadata.start()
distributed_loader: Remove local aliases and exporters
distributed_loader: Move populate_column_family() into population meta
It used to be just metadata by providing the meta for population, now it
does the population by itself, so rename it.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Schema related files are moved there. This excludes schema files that
also interact with mutations, because the mutation module depends on
the schema. Those files will have to go into a separate module.
Closes#12858
Patch 55a8421e3d fixed an inefficiency when rebuilding
statistics with many compaction groups, but it incorrectly removed
the update for newly added SSTables. This patch restores it.
When a new SSTable is added to any of the groups, the stats are
incrementally updated (as before). On compaction completion,
statistics are still rebuilt by simply iterating through each
group, which keeps track of its own stats.
Unit tests are added to guarantee the stats are correct both after
compaction completion and memtable flush.
Fixes#12808.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closes#12834
Whenever any compaction group has its SSTable set updated, table's
rebuild_statistics() is called and it inefficiently iterates through
SSTable set of all compaction groups.
Now each compaction group keeps track of its statistics, such that
table's rebuild_statistics() only need to sum them up.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
There's another one that accepts explicit basedir first argument and
that's used by the rest of the code.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Closes#12643
Currently there are four helpers, this patch makes it just two and one
of them becomes private the table thus making the API small and neat
(and easy to patch further).
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Add a new virtual table `system.raft_state` that shows the currently
operating Raft configuration for each present group. The schema is the
same as `system.raft_snapshot_config` (the latter shows the config from
the last snapshot). In the future we plan to add more columns to this
table, showing more information (like the current leader and term),
hence the generic name.
Adding the table requires some plumbing of
`sharded<raft_group_registry>&` through function parameters to make it
accessible from `register_virtual_tables`, but it's mostly
straightforward.
Also added some APIs to `raft_group_registry` to list all groups and
find a given group (returning `nullptr` if one isn't found, not throwing
an exception).
Inferring shard from generation is long gone. We still use it in
some scripts, but that's no longer needed in Scylla, when loading
the SSTables, and it also conflicts with ongoing work of UUID-based
generations.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closes#12476
