Currently, the tablet repair scheduler repairs all replicas of a tablet. It does not support hosts or DCs selection. It should be enough for most cases. However, users might still want to limit the repair to certain hosts or DCs in production. https://github.com/scylladb/scylladb/pull/21985 added the preparation work to add the config options for the selection. This patch adds the hosts or DCs selection support.
Fixes https://github.com/scylladb/scylladb/issues/22417
New feature. No backport is needed.
Closesscylladb/scylladb#22621
* github.com:scylladb/scylladb:
test: add test to check dcs and hosts repair filter
test: add repair dc selection to test_tablet_metadata_persistence
repair: Introduce Host and DC filter support
docs: locator: update the docs and formatter of tablet_task_info
This commit eliminates unused boost header includes from the tree.
Removing these unnecessary includes reduces dependencies on the
external Boost.Adapters library, leading to faster compile times
and a slightly cleaner codebase.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#22857
Currently, the tablet repair scheduler repairs all replicas of a tablet.
It does not support hosts or DCs selection. It should be enough for most
cases. However, users might still want to limit the repair to certain
hosts or DCs in production. #21985 added the preparation work to add the
config options for the selection. This patch adds the hosts or DCs
selection support.
Fixes#22417
Set true to wait for the repair to complete. Set false to skip waiting
for the repair to complete. When the option is not provided, it defaults
to false.
It is useful for management tool that wants the api to be async.
Fixes#22418Closesscylladb/scylladb#22436
Use host_id in a children list of a task in task manager to indicate
a node on which the child was created.
Move TASKS_CHILDREN_REQUEST to IDL. Send it by host_id.
Fixes: https://github.com/scylladb/scylladb/issues/22284.
Ip to host_id transition; backport isn't needed.
Closesscylladb/scylladb#22487
* github.com:scylladb/scylladb:
tasks: drop task_manager::config::broadcast_address as it's unused
tasks: replace ip with host_id in task_identity
api: task_manager: pass gossiper to api::set_task_manager
tasks: keep host_id in task_manager
tasks: move tasks_get_children to IDL
In column_family.cc and storage_service.cc there exist a bunch of helpers that parse and/or validate ks/cf names, and different endpoints use different combinations of those, duplicating the functionality of each other and generating some mess. This PR cleans the endpoints from column_family.cc that parse and validate fully qualified table name (the '$ks:$cf' string).
A visible "improvement" is that `validate_table()` helper usage in the api/ directory is narrowed down to storage_service.cc file only (with the intent to remove that helper completely), and the aforementioned `for_tables_on_all_shards()` helper becomes shorter and tiny bit faster, because it doesn't perform some re-lookups of tables, that had been performed by validation sanity checks before it.
There's more to be done in those helpers, this PR wraps only one part of this mess.
Below is the list of endpoints this PR affects and the tests that validate the changes:
|endpoint|test|
|-|-|
|column_family/autocompaction|rest_api/test_column_family::test_column_family_auto_compaction_table|
|column_family/tombstone_gc|rest_api/test_column_family::test_column_family_tombstone_gc_api|
|column_family/compaction_strategy|rest_api/test_column_family/test_column_family_compaction_strategy|
|compaction_manager/stop_keyspace_compaction/|rest_api/test_compaction_manager::{test_compaction_manager_stop_keyspace_compaction,test_compaction_manager_stop_keyspace_compaction_tables}|
Closesscylladb/scylladb#21533
* github.com:scylladb/scylladb:
api: Hide parse_tables() helper
api: Use parse_table_infos() in stop_keyspace_compaction handler
api: Re-use parse_table_info() in column_family API
api: Make get_uuid() return table_info (and rename)
api: Remove keyspace argument from for_table_on_all_shards()
api: Switch for_table_on_all_shards() to use table_info-s
api: Hide validate_table() helper
api: Tables vector is never empty now in for_table_on_all_shards()
api: Move vectors of tables, not copy
api: Add table validation to set_compaction_strategy_class endpoint
api: Use get_uuid() to validate_table() in column family API
api: Use parse_table_infos() in column family API
task_stats contains short info about a task. To get a list of task_stats
in the module, one needs to request /task_manager/list_module_tasks/{module}.
To make identification and navigation between tasks easier, extend
task_stats to contain shard, start_time, and end_time.
Closesscylladb/scylladb#22351
This was originally an attempt to reduce the compile time of this
translation unit, but apparently it doesn't work. Still, it has
the effect of converting stack traces that say "set_storage_service"
and refer to some lambda to stack traces that refer to the operation
being performed, so it's a net positive.
To faciliate the change, we introduce new functions rest_bind(),
similar to (and in fact wrapping) std::bind_front(), that capture
references like the lambdas did originally. We can't use
std::bind_front directly since the call to
seastar::httpd::path_description::set() cannot be disambiguated
after the function is obscured by the template returned by
std::bind_front. The new function rest_bind() has constraints
to understand which overload is in use.
Closesscylladb/scylladb#22526
Currently, when the status of a task is queried and the task is already finished,
it gets unregistered. Getting the status shouldn't be a one-time operation.
Stop removing the task after its status is queried. Adjust tests not to rely
on this behavior. Add task_manager/drain API and nodetool tasks drain
command to remove finished tasks in the module.
Fixes: https://github.com/scylladb/scylladb/issues/21388.
It's a fix to task_manager API, should be backported to all branches
Closesscylladb/scylladb#22310
* github.com:scylladb/scylladb:
api: task_manager: do not unregister tasks on get_status
api: task_manager: add /task_manager/drain
In the following patches, get_status won't be unregistering finished
tasks. However, tests need a functionality to drop a task, so that
they could manipulate only with the tasks for operations that were
invoked by these tests.
Add /task_manager/drain/{module} to unregister all finished tasks
from the module. Add respective nodetool command.
Previously, during backup, SSTable components are preserved in the
snapshot directory even after being uploaded. This leads to redundant
uploads in case of failed backups or restarts, wasting time and
resources (S3 API calls).
This change
- adds an optional query parameter named "move_files" to
"/storage_service/backup" API. if it is set to "true", SSTable
components are removed once they are backed up to object storage.
- conditionally removes SSTable components from the snapshot directory once
they are successfully uploaded to the target location. This prevents
re-uploading the same files and reduces disk usage.
This change only "Refs" #20655, because, we can move further optimize
the backup process, consider:
- Sending HEAD requests to S3 to check for existing files before uploading.
- Implementing support for resuming partially uploaded files.
Fixes#21799
Refs #20655
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
In this change, tablet_virtual_task starts supporting tablet
resize (i.e. split and merge).
Users can see running resize tasks - finished tasks are not
presented with the task manager API.
A new task state "suspended" is added. If a resize was revoked,
it will appear to users as suspended. We assume that the resize was revoked
when the tablet number didn't change.
Fixes: #21366.
Fixes: #21367.
No backport, new feature
Closesscylladb/scylladb#21891
* github.com:scylladb/scylladb:
test: boost: check resize_task_info in tablet_test.cc
test: add tests to check revoked resize virtual tasks
test: add tests to check the list of resize virtual tasks
test: add tests to check spilt and merge virtual tasks status
test: test_tablet_tasks: generalize functions
replica: service: add split virtual task's children
replica: service: pass parent info down to storage_group::split
tasks: children of virtual tasks aren't internal by default
tasks: initialize shard in task_info ctor
service: extend tablet_virtual_task::abort
service: retrun status_helper struct from tablet_virtual_task::get_status_helper
service: extend tablet_virtual_task::wait
tasks: add suspended task state
service: extend tablet_virtual_task::get_status
service: extend tablet_virtual_task::contains
service: extend tablet_virtual_task::get_stats
service: add service::task_manager_module::get_nodes
tasks: add task_manager::get_nodes
tasks: drop noexcept from module::get_nodes
replica: service: add resize_task_info static column to system.tablets
locator: extend tablet_task_info to cover resize tasks
Now that all topology related code uses host ids there is not point to
maintain ip to id (and back) mappings in the token metadata. After the
patch the mapping will be maintained in the gossiper only. The rest of
the system will use host ids and in rare cases where translation is
needed (mostly for UX compatibility reasons) the translation will be
done using gossiper.
Fixes: scylladb/scylla#21777
* 'gleb/drop-ip-from-tm-v3' of github.com:scylladb/scylla-dev: (57 commits)
hint manager: do not translate ip to id in case hint manager is stopped already
locator: token_metadata: drop update_host_id() function that does nothing now
locator: topology: drop indexing by ips
repair: drop unneeded code
storage_service: use host_id to look for a node in on_alive handler
storage_proxy: translate ips to ids in forward array using gossiper
locator: topology: remove unused functions
storage_service: check for outdated ip in on_change notification in the peers table
storage_proxy: translate id to ip using address map in tablets's describe_ring code instead of taking one from the topology
topology coordinator: change connection dropping code to work on host ids
cql3: report host id instead of ip in error during SELECT FROM MUTATION_FRAGMENTS query
locator: drop unused function from tablet_effective_replication_map
api: view_build_statuses: do not use IP from the topology, but translate id to ip using address map instead
locator: token_metadata: remove unused ip based functions
locator: network_topology_strategy: use host_id based function to check number of endpoints in dcs
gossiper: drop get_unreachable_token_owners functions
storage_service: use gossiper to map ip to id in node_ops operations
storage_service: fix indentation after the last patch
storage_service: drop loops from node ops replace_prepare handling since there can be only one replacing node
token_metadata: drop no longer used functions
...
these unused includes were identifier by clang-include-cleaner. after
auditing these source files, all of the reports have been confirmed.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#22201
The API /storage_service/truncate/{ks} returns an unimplemented
error when invoked. As we already have a CQL command,
`TRUNCATE TABLE ks.cf` that causes the table to be truncated on all
nodes, the API can be dropped. Due to the error, it is unused.
Fixes https://github.com/scylladb/scylladb/issues/10520
No backport is required. A small cleanup of not working API.
Closesscylladb/scylladb#22258
One of its caller is in the RESTful API which gets ips from the user, so
we convert ips to ids inside the API handler using gossiper before
calling the function. We need to deprecate ip based API and move to host
id based.
In Scylla there are two options that control IO bandwidth limit -- the /storage_service/(compaction|stream)_throughput REST API endpoints. The endpoints are partially implemented and have no counterparts in the nodetool.
This set implements the missing bits and adds tests for new functionality.
Closesscylladb/scylladb#21877
* github.com:scylladb/scylladb:
nodetool: Implement [gs]etstreamthroughput commands
nodetool: Implement [gs]etcompationthroughput commands
test: Add validation of how IO-updating endpoints work
api: Implement /storage_service/(stream|compaction)_throughput endpoints
api: Disqualify const config reference
api: Implement /storage_service/stream_throughput endpoint
api: Move stream throughput set/get endpoints from storage service block
api: Move set_compaction_throughput_mb_per_sec to config block
util: Include fmt/ranges.h in config_file.hh
It now parses only table names from its "cf" argument. Parsing
table_infos has two benefits -- it makes it possible to hide
parse_tables() thus keeping less validation code around, and the
subsequent db.find_column_family() call can avoid re-lookup of table
uuid by its ks:table pair.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Several places call parse_fully_qualified_cf_name() and get_uuid()
helpers one after another. Previous patch introduced the
parse_table_info() one that wraps both.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
The method gets "fully qualified" table name, which is 'ks:cf' string
and returns back the resolved table_id value. Some callers will benefit
from knowing the parsed 'cf' part of it (see next patch).
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
This argument is needed to find table by ks:cf prair. The "table" part
is taken from the vector of table_info-s, but table_info-s have table_id
value onboard, and the table can be found by this id. So keyspace is not
needed any longer.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
All callers of it already have one. Next patch will make even more use
of those passed table_info-s.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
It's no longer used outside of api/storage_service.cc. It's not yet
possible to remove it completely, but it's better not to encourage
others to use it outside of its current .cc file.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Callers of this method provide vectors of two kinds:
- explicitly single-entry one from endpoints that work on single table
- vector returned by parse_table_infos()
The latter helper, if it gets empty list of tables from user, populates
its return value with all tables from the given keyspace.
The removed check became obsolete after recent changes. Prior to those,
the 2nd case provided vector from another helper called parse_tables(),
which could return empty result.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
The set_tables_...() helper called here accept vector by value, so the
existing code copies it. It's better to move, all the more so next
changes will make this place pass vectors with more data onboard.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
This handler doesn't check if the requested table exists. If it doesn't
it will throw later anyway, but most of other endpoints that work with
tables check table early. This early check allows throwing bad-param
exception on missing table, not internal-server-error one.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
This helper returns uuid, but also "Validates" the table exists by
calling db.find_uuid() and throwing bad_param exception on error.
This change will allow making for_table_on_all_shards() smaller a bit
later.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
The one is the same as parse_tables(), but returns back name:id pairs.
This change will allow making for_table_on_all_shards() smaller a bit
later, as well as removing the parse_tables() code eventually.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Previously, we created a vector<utils_json::histogram> and returned it
by copying into a future. Since histogram is a JSON representation of
ihistogram, it can be heavyweight, making the vector copy overhead
significant.
Now we move the vector into the returned future instead of copying it,
eliminating the deep copy overhead. The APIs backed by this function
are marked deprecated, so this performance improvement is not that
important.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#22004
Previously, `api/service_levels.hh` includes `api/api.hh` for
accessing symbols like `api/http_context`. but these symbols are
already available in a "smaller" header -- `api/api_init.hh`. so,
in order to improve the build efficiency, let's include smaller
headers in favor of "larger" ones.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#22178
Introduces two endpoints with operations specific to service levels:
- switch_tenants: updates the scheduling group of all connections to be
aligned with the service level specific to the logged in user. This is
mostly legacy API, as with service levels on raft this is done
automatically.
- count_connections: for each user and for each scheduling group, counts
how many connections are assigned to that user and scheduling group.
This API is used in tests.
The main purpose of this change is to enhance the restore from object storage usage.
Currently, restore uses the load-and-stream facility. When triggered, the restoring task opens the provided list of sstables directory from the remote bucket and then feeds the list of sstables to load_and_stream() method. The method, in turn, iterates over this list, reads mutations and for each mutation decides where to send one by checking the replication map (it's pretty much the same for both vnodes and tablets, but for tablets that are "fully contained" by a range there's the plan to stream faster).
As described above, restore is governed by a single node and this single node reads all sstables from the object store, which can be very slow. This PR allows speeding things up. For that, the load-and-stream code is equipped with the "scope" filter which limits where mutations can be streamed to. There are four options for that -- all, dc, rack and node. The "all" is how things work currently, "dc" and "rack" filter out target nodes that don't belong to this node's dc/rack respectively. The "node" scope only streams mutations to local node.
With the "node" scope it's possible to make all nodes in the cluster load mutations that belong to them in parallel, without re-sending them to peers. The last patch in this PR is the test that shows how it can be possible.
Closesscylladb/scylladb#21169
* github.com:scylladb/scylladb:
test: Add scope-streaming test (for restore from backup)
api: New "scope" API param to load-and-stream calls
sstables_loader: Propagate scope from API down
sstables_loader: Filter tablets based on scope
streamer: Disable scoped streaming of primary replica only
sstables_loader: Introduce streaming scope
sstables_loader: Wrap get_endpoints()
Somehow system."IndexInfo" table and column_family/built_indexes REST
API endpoint declare an index "built" at slightly different times:
The former a virtual table which declares an index completely built
when it appears on the system.built_views table.
The latter uses different data -- it takes the list of indexes in
the schema and eliminates indexes which are still listed in the
system.scylla_views_builds_in_progress table.
The mentioned system. tables are updated at different times, so API
notices the change a bit later. It's worth improving the consistency
of these two APIs by making the REST API endpoint piggy-back the
load_built_views() instead of load_view_build_progress(). With that
change the filtering of indexes should be negated.
Fixes#21587
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
There are two of those -- the POST /storage_service/keyspace that loads
and streams new sstables from /upload and POST /storage_service/restore
that does the same, but gets sstables from object store.
The new optional parameter allow users to tun the streaming phase
behavior. The test/pylib client part is also updated here.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Semi-mechanical change that adds newly introduced "scope" parameter to
all the functions between API methods and the low-level streamer object.
No real functional changes. API methods set it to "all" to keep existing
behavior.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Both values are in fact db::config named values. They are observed by,
respectively, compaction manager and stream manager: when changed, the
observer kicks corresponding sched group's update_io_bandwidth() method.
Despite being referenced by managers, there's no way to update those
values anyhow other than updating config's named values themselves.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Some endpoints in config block will need to actually _update_ values on
config (see next patches why), and const reference stands on the way.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
In order to get stream throughput, the API will need stream_manager.
In order to set stream throughput, the API will need db::config to
update the corresponding named value on it.
Said that, move the endpoints to relevant blocks.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
In order to update compaction throughput API would need to update the
db::config value, so the endpoint in question should sit in the block
that has db::config at hand.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
