Compare commits

...

40 Commits

Author SHA1 Message Date
Lei, HUANG
97e9b97a57 feat/bulk-support-flow-batch:
Update `greptime-proto` Dependency

 - Updated the `greptime-proto` dependency to a new revision in both `Cargo.lock` and `Cargo.toml`.
   - `Cargo.lock`: Changed the source revision from `f0913f179ee1d2ce428f8b85a9ea12b5f69ad636` to `17971523673f4fbc982510d3c9d6647ff642e16f`.
   - `Cargo.toml`: Updated the `greptime-proto` git revision to `17971523673f4fbc982510d3c9d6647ff642e16f`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:28:12 +00:00
Lei, HUANG
7fc74e2928 feat/bulk-support-flow-batch:
## Update `handle_mark_window_dirty` Method in `flownode_impl.rs`

 - Replaced `unimplemented!()` with `unreachable!()` in the `handle_mark_window_dirty` method for both `FlowDualEngine` and `StreamingEngine` implementations in `flownode_impl.rs`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:22:25 +00:00
Lei, HUANG
77f20ede7a feat/bulk-support-flow-batch:
**Refactor and Enhance Timestamp Handling in `bulk_insert.rs`**

 - **Refactored Timestamp Extraction**: Moved timestamp extraction logic to a new method `maybe_update_flow_dirty_window` to improve code readability and maintainability.
 - **Enhanced Flow Update Logic**: Updated the flow dirty window update mechanism to conditionally notify flownodes only if they are configured, using `table_info` and `record_batch`.
 - **Imports Adjusted**: Updated imports to reflect changes in table metadata handling, replacing `TableId` with `TableInfoRef`.

 Files affected:
 - `src/operator/src/bulk_insert.rs`

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:22:24 +00:00
Lei, HUANG
ced018fce0 feat/bulk-support-flow-batch:
### Update Metrics in Batching Mode Engine

 - **Modified Metrics**: Replaced `METRIC_FLOW_BATCHING_ENGINE_BULK_MARK_TIME_WINDOW_RANGE` with `METRIC_FLOW_BATCHING_ENGINE_BULK_MARK_TIME_WINDOW` to track the count of time windows instead of their range.
   - Files affected: `engine.rs`, `metrics.rs`
 - **New Method**: Added `len()` method to `DirtyTimeWindows` to return the number of dirty windows.
   - File affected: `state.rs`

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:22:24 +00:00
Lei, HUANG
41dacff283 feat/bulk-support-flow-batch:
**Refactor Timestamp Handling and Update Dependencies**

 - **Dependency Update**: Updated `greptime-proto` dependency in `Cargo.lock` and `Cargo.toml` to a new revision.
 - **Batching Engine Refactor**: Modified `src/flow/src/batching_mode/engine.rs` to replace `dirty_time_ranges` with `timestamps` for improved timestamp handling.
 - **Bulk Insert Refactor**: Updated `src/operator/src/bulk_insert.rs` to refactor timestamp extraction and handling. Replaced `compute_timestamp_range` with `extract_timestamps` and adjusted related logic to handle timestamps directly.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:22:22 +00:00
discord9
94a14b6da7 metrics: more useful metrics batching mode 2025-06-16 06:22:06 +00:00
discord9
6ad3a32cb2 feat: metrics 2025-06-16 06:22:06 +00:00
discord9
ac00314578 feat: mark dirty time window 2025-06-16 06:22:05 +00:00
Lei, HUANG
2f08bee08f feat/bulk-support-flow-batch:
### Add Dirty Window Handling in Flow Module

 - **Updated `greptime-proto` Dependency**: Updated the `greptime-proto` dependency to a new revision in `Cargo.lock` and `Cargo.toml`.
 - **Flow Module Enhancements**:
   - Added `DirtyWindowRequest` handling in `flow.rs`, `node_manager.rs`, `test_util.rs`, `flownode_impl.rs`, and `server.rs`.
   - Implemented `handle_mark_window_dirty` function to manage dirty time windows.
 - **Bulk Insert Enhancements**:
   - Modified `bulk_insert.rs` to notify flownodes about dirty time windows using `update_flow_dirty_window`.
 - **Removed Unused Imports**: Cleaned up unused imports in `greptime_handler.rs`, `grpc.rs`, and `mod.rs`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:22:03 +00:00
Lei, HUANG
8ebb31cdcd feat/bulk-support-flow-batch:
### Refactor and Enhance Timestamp Handling in gRPC and Bulk Insert

 - **Refactor Table Handling**:
   - Updated `put_record_batch` method to use `TableRef` instead of `TableId` in `grpc.rs`, `greptime_handler.rs`, and `grpc.rs`.
   - Modified `handle_bulk_insert` to accept `TableRef` and extract `TableId` internally in `bulk_insert.rs`.

 - **Enhance Timestamp Processing**:
   - Added `compute_timestamp_range` function to calculate timestamp range in `bulk_insert.rs`.
   - Introduced error handling for invalid time index types in `error.rs`.

 - **Test Adjustments**:
   - Updated `DummyInstance` implementation in `tests/mod.rs` to align with new method signatures.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-16 06:21:32 +00:00
localhost
f4f8d65a39 fix: event api content type only check type and subtype (#6317)
* fix: event api content type only check type and subtype

Signed-off-by: paomian <xpaomian@gmail.com>

* chore: make clippy happy

Signed-off-by: paomian <xpaomian@gmail.com>

---------

Signed-off-by: paomian <xpaomian@gmail.com>
2025-06-13 18:50:05 +00:00
Lei, HUANG
b31990e881 chore: add connection info to QueryContext (#6319)
chore/add-conn-info-to-query-ctx:
 ### Add Connection Information to Query Context

 - **`src/frontend/src/instance.rs`**: Updated to use `query_ctx.conn_info().to_string()` for connection information instead of a placeholder string.
 - **`src/session/src/context.rs`**: Introduced `conn_info` field in `QueryContext` and added a method `conn_info()` to retrieve it. Updated `QueryContextBuilder` to handle `conn_info`.
 - **`src/session/src/lib.rs`**: Modified `Session` to include `conn_info` in the query context building process.

 These changes enhance the query context by incorporating connection information, allowing for more detailed session management.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-13 18:42:13 +00:00
Lei, HUANG
6da633e70d feat: support killing process (#6309)
* feat/kill-process:
 ### Add Cancellation Support and Enhance Process Management

 - **Cancellation Handle Implementation**: Introduced `CancellationHandle` in `cancellation_handle.rs` to facilitate cancellation of futures and streams.
 - **Process Management Enhancements**:
   - Updated `ProcessManager` in `process_manager.rs` to support cancellable processes using `CancellableProcess`.
   - Added `kill_process` method for terminating processes.
 - **Stream Wrapper Update**:
   - Replaced `StreamWrapper` with `CancellableStreamWrapper` in `stream_wrapper.rs` and `instance.rs` to handle stream cancellation.
 - **Error Handling**:
   - Added `StreamCancelled` error variant in `error.rs` to handle stream cancellation scenarios.
 - **gRPC Handler Update**:
   - Added `kill_process` gRPC method in `frontend_grpc_handler.rs` to allow external process termination.
 - **Dependency Updates**:
   - Updated `Cargo.lock` and `Cargo.toml` to include `common-base` and `tokio-util`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 **Enhancements and Bug Fixes**

 - **Dependency Update**: Updated `greptime-proto` dependency in `Cargo.lock` and `Cargo.toml` to a new revision.
 - **Error Handling Improvements**:
   - Modified error variants in `src/catalog/src/error.rs` and `src/common/frontend/src/error.rs` to improve error messages and handling.
   - Added `FrontendNotFound` error variant for better error specificity.
 - **Process Management Enhancements**:
   - Updated `ProcessManager` in `src/catalog/src/process_manager.rs` to include `kill_process` functionality with server address validation.
   - Enhanced `FrontendClient` trait in `src/common/frontend/src/selector.rs` to support `kill_process` requests.
 - **gRPC Handler Update**:
   - Refactored `FrontendGrpcHandler` in `src/servers/src/grpc/frontend_grpc_handler.rs` to handle `kill_process` requests asynchronously and return process status.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 ### Add Kill Process Functionality

 - **`Cargo.lock`, `Cargo.toml`**: Added `common-frontend` as a dependency.
 - **`server.rs`, `builder.rs`, `instance.rs`**: Updated `FrontendInvoker` and `FrontendBuilder` to support process management.
 - **`error.rs`**: Introduced `InvalidProcessId` error for handling invalid process IDs.
 - **`statement.rs`, `kill.rs`**: Implemented `execute_kill` method in `StatementExecutor` to handle the `KILL` statement.
 - **`parser.rs`, `statement.rs`**: Updated SQL parser to recognize and parse the `KILL` statement.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 ## Add Cancellation Support to Query Execution

 - **`process_manager.rs`**: Updated `CancellationHandle` initialization to use `default()` method.
 - **`cancellation_handle.rs`**: Implemented `Debug` trait for `CancellationHandle` and added `Cancellation` and `CancellableFuture` structs to support cancellable futures.
 - **`error.rs`**: Introduced `Cancelled` error variant to handle query cancellations.
 - **`instance.rs`**: Integrated `CancellableFuture` to manage query execution with cancellation support.
 - **`stream_wrapper.rs`**: Modified `CancellableStreamWrapper` to use the new `waker()` method for cancellation handling.
 - **`statement.rs`**: Added `#[allow(clippy::too_many_arguments)]` to `StatementExecutor::new` to suppress clippy warnings.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 - **Add `MetaClientMissing` Error**: Introduced a new error variant `MetaClientMissing` in `error.rs` to handle missing meta client scenarios.
 - **Refactor Cancellation Handling**: Merged `cancellation_handle.rs` into `cancellation.rs` and updated related logic in `process_manager.rs`, `instance.rs`, and `stream_wrapper.rs`.
 - **Enhance Process Management**: Improved process management logic in `process_manager.rs` to handle process cancellation more effectively.
 - **Update Tests**: Added and updated tests in `cancellation.rs` and `stream_wrapper.rs` to cover new cancellation logic and error handling.
 - **Cargo.toml Update**: Adjusted workspace settings in `Cargo.toml` for `common-frontend`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 - **Add Tests for Process Management**: Introduced multiple async tests in `process_manager.rs` to verify query registration, deregistration, cancellation, and process killing functionalities.
 - **Update Error Message in SQL Parser**: Modified the expected error message in `parser.rs` to clarify the expected token as a "process id string literal".

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 ### Add Process Count Metrics to Catalog

 - **`metrics.rs`**: Introduced a new metric `PROCESS_LIST_COUNT` to track the count of running processes per catalog using `IntGaugeVec`.
 - **`process_manager.rs`**: Updated `CancellableProcess` to increment and decrement `PROCESS_LIST_COUNT` upon creation and destruction, respectively. Added a `Drop` implementation for `CancellableProcess` to handle metric updates.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 ### Fix process removal logic in `process_manager.rs`

 - Corrected the condition for removing an entry from the catalog in `ProcessManager` by using `o.get()` instead of `o.get_mut()`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 - **Error Handling Improvements**:
   - Updated status codes for `Error::FrontendNotFound` and `Error::MetaClientMissing` to `StatusCode::Unexpected` in `src/catalog/src/error.rs`.
   - Changed `InvokeFrontend` error display message and status code in `src/common/frontend/src/error.rs`.
   - Added `ProcessManagerMissing` error in `src/operator/src/error.rs` and updated its handling in `src/operator/src/statement/kill.rs`.

 - **Process Management Enhancements**:
   - Added documentation for `ProcessManager` and `register_query` in `src/catalog/src/process_manager.rs`.
   - Modified `kill_process` response handling in `src/servers/src/grpc/frontend_grpc_handler.rs`.

 - **Cancellation Logic Update**:
   - Improved cancellation logic in `src/common/base/src/cancellation.rs` to use `compare_exchange` for atomic operations.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 ### Add Process Kill Count Metric and Refactor Cancellation Handle

 - **Metrics Update**: Added a new metric `PROCESS_KILL_COUNT` in `metrics.rs` to track the count of completed kill process requests per catalog.
 - **Refactor Cancellation Handle**: Renamed `cancellation_handler` to `cancellation_handle` across multiple files for consistency:
   - `process_manager.rs`
   - `instance.rs`
   - `stream_wrapper.rs`
 - **Process Management**: Updated process management logic in `process_manager.rs` to increment the `PROCESS_KILL_COUNT` metric upon successful process termination.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 Update metric description in `metrics.rs`

 - Changed the description of `PROCESS_KILL_COUNT` to reflect the count of killed processes instead of running processes in `metrics.rs`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* feat/kill-process:
 Update `greptime-proto` Dependency and Fix Response Field

 - **Updated Dependency**: Changed the `greptime-proto` Git revision in `Cargo.lock` and `Cargo.toml` to `f0913f1`.
 - **Code Fix**: Modified `frontend_grpc_handler.rs` to correct the response field from `found` to `success` in `KillProcessResponse`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

---------

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-13 13:30:25 +00:00
zyy17
9633e794c7 fix: always use linux path style in windows platform unit tests (#6314)
Signed-off-by: zyy17 <zyylsxm@gmail.com>
2025-06-13 07:15:53 +00:00
Yingwen
eaf1e1198f refactor: Extract mito codec part into a new crate (#6307)
* chore: add a new crate mito-codec

Signed-off-by: evenyag <realevenyag@gmail.com>

* feat: port necessary mods for primary key codec

Signed-off-by: evenyag <realevenyag@gmail.com>

* refactor: use codec utils in mito-codec

Signed-off-by: evenyag <realevenyag@gmail.com>

* refactor: remove unused mods

Signed-off-by: evenyag <realevenyag@gmail.com>

* style: fix clippy

Signed-off-by: evenyag <realevenyag@gmail.com>

* refactor: remove Partition::is_partition_column()

Signed-off-by: evenyag <realevenyag@gmail.com>

* refactor: remove duplicated test utils

Signed-off-by: evenyag <realevenyag@gmail.com>

* chore: remove unused comment

Signed-off-by: evenyag <realevenyag@gmail.com>

* fix: fix is_partition_column check

Signed-off-by: evenyag <realevenyag@gmail.com>

---------

Signed-off-by: evenyag <realevenyag@gmail.com>
2025-06-13 07:14:29 +00:00
ZonaHe
505bf25505 feat: update dashboard to v0.9.3 (#6311)
Co-authored-by: ZonaHex <ZonaHex@users.noreply.github.com>
2025-06-13 07:13:12 +00:00
Ning Sun
f1b29ece3c feat: process id for session, query context and postgres (#6301)
* feat: process id for session, query context and postgres

Signed-off-by: Ning Sun <sunning@greptime.com>

* feat: add sql functions to retrieve connection/process id

Signed-off-by: Ning Sun <sunning@greptime.com>

---------

Signed-off-by: Ning Sun <sunning@greptime.com>
2025-06-12 16:53:57 +00:00
discord9
74df12e8c0 fix: check for zero parallelism (#6310)
* fix: check for zero parallelism

Signed-off-by: discord9 <discord9@163.com>

* chore: silently use default value

Signed-off-by: discord9 <discord9@163.com>

---------

Signed-off-by: discord9 <discord9@163.com>
2025-06-12 15:58:59 +00:00
discord9
be6a5d2da8 feat: parallelism hint in grpc (#6306)
* feat: parallelism hint in grpc

Signed-off-by: discord9 <discord9@163.com>

* chore: per review

Signed-off-by: discord9 <discord9@163.com>

* chore: comment

Signed-off-by: discord9 <discord9@163.com>

* chore:docs

Signed-off-by: discord9 <discord9@163.com>

---------

Signed-off-by: discord9 <discord9@163.com>
2025-06-12 10:12:45 +00:00
Ruihang Xia
7468a8ab2a feat: organize EXPLAIN ANALYZE VERBOSE's output in JSON format (#6308)
Signed-off-by: Ruihang Xia <waynestxia@gmail.com>
2025-06-12 09:55:53 +00:00
Lei, HUANG
5bb0466ff2 feat: introduce file group in compaction (#6261)
* fix/file-group-in-compaction:
 ### Enhance Compaction Logic with File Grouping

 - **`run.rs`**: Introduced `FileGroup` struct to manage groups of `FileHandle` objects, allowing for more efficient compaction operations. Updated `Ranged` and `Item` trait implementations to work with `FileGroup`.
 - **`test_util.rs`**: Added `new_file_handle_with_sequence` function to support file handles with sequence numbers, enhancing test utilities.
 - **`twcs.rs`**: Modified `TwcsPicker` to utilize `FileGroup` for managing files within windows, improving compaction logic. Updated `Window` struct to use `HashMap` for storing `FileGroup` objects.
 - **`version_util.rs`**: Updated version control utilities to handle sequence numbers in file metadata, aligning with new compaction logic.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* fix/file-group-in-compaction:
 ### Add Test for File Group Assignment in TWCS

 - **Enhancements in `twcs.rs`:**
   - Added a new test `test_assign_file_groups_to_windows` to verify the correct assignment of file groups to windows.
   - Enhanced `test_assign_compacting_to_windows` with a new case to ensure files with overlapping time ranges and the same sequence are treated as one `FileGroup`.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* fix/file-group-in-compaction:
 **Enhance Compaction Task Documentation and Initialization**

 - **`run.rs`**: Added documentation for `FileGroup` to clarify its role in representing a group of files created by the same compaction task.
 - **`twcs.rs`**: Introduced comments in the `Window` struct to explain the mapping of file sequences to file groups, indicating files created from the same compaction task. Simplified the initialization of the `files` hashmap using `HashMap::from`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

---------

Signed-off-by: Lei, HUANG <lhuang@greptime.com>
Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
2025-06-12 09:33:40 +00:00
Ruihang Xia
f6db419afd feat: support using expressions as literal in PromQL (#6297)
Signed-off-by: Ruihang Xia <waynestxia@gmail.com>
2025-06-12 08:18:10 +00:00
Lei, HUANG
05b708ed2e feat: implement process manager and information_schema.process_list (#5865)
* ### Add Process List Management

 - **Error Handling Enhancements**:

* refactor: Update test IP addresses to include ports in ProcessKey

* feat/show-process-list:
 Refactor Process Management in Meta Module

 - Introduced `ProcessManager` for handling process registration and deregistration.
 - Added methods for managing and querying process states, including `register_query`, `deregister_query`, and `list_all_processes`.
 - Removed redundant process management code from the query module.
 - Updated error handling to reflect changes in process management.
 - Enhanced test coverage for process management functionalities.

* chore: rebase main

* add information schema process list table

* integrate process list table to system catalog

* build ProcessManager on frontend and standalone mode

* feat/show-process-list:
 **Add Process Management Enhancements**

 - **`manager.rs`**: Introduced `process_manager` to `SystemCatalog` and `KvBackendCatalogManager` for improved process handling.
 - **`information_schema.rs`**: Updated table insertion logic to conditionally include `PROCESS_LIST`.
 - **`frontend.rs`, `standalone.rs`**: Enhanced `StartCommand` to clone `process_manager` for better resource management.
 - **`instance.rs`, `builder.rs`**: Integrated `ProcessManager` into `Instance` and `FrontendBuilder` to manage query

* feat/show-process-list:
 ### Add Process Listing and Error Handling Enhancements

 - **Error Handling**: Introduced a new error variant `ListProcess` in `error.rs` to handle failures when listing running processes.
 - **Process List Implementation**: Enhanced `InformationSchemaProcessList` in `process_list.rs` to track running queries, including defining column names and implementing the `make_process_list` function to build the process list.
 - **Frontend Builder**: Added a `#[allow(clippy::too_many_arguments)]` attribute in `builder.rs` to suppress Clippy warnings for the `FrontendBuilder::new` function.

 These changes improve error handling and process tracking capabilities within the system.

* feat/show-process-list:
 Refactor imports in `process_list.rs`

 - Updated import paths for `Predicates` and `InformationTable` in `process_list.rs` to align with the new module structure.

* feat/show-process-list:
 Refactor process list generation in `process_list.rs`

 - Simplified the process list generation by removing intermediate row storage and directly building vectors.
 - Updated `process_to_row` function to use a mutable vector for current row data, improving memory efficiency.
 - Removed `rows_to_record_batch` function, integrating its logic directly into the main loop for streamlined processing.

* wip: move ProcessManager to catalog crate

* feat/show-process-list:
 - **Refactor Row Construction**: Updated row construction in multiple files to use references for `Value` objects, improving memory efficiency. Affected files include:
   - `cluster_info.rs`
   - `columns.rs`
   - `flows.rs`
   - `key_column_usage.rs`
   - `partitions.rs`
   - `procedure_info.rs`
   - `process_list.rs`
   - `region_peers.rs`
   - `region_statistics.rs`
   - `schemata.rs`
   - `table_constraints.rs`
   - `tables.rs`
   - `views.rs`
   - `pg_class.rs`
   - `pg_database.rs`
   - `pg_namespace.rs`
 - **Remove Unused Code**: Deleted unused functions and error variants related to process management in `process_list.rs` and `error.rs`.
 - **Predicate Evaluation Update**: Modified predicate evaluation functions in `predicate.rs` to work with references, enhancing performance.

* feat/show-process-list:
 ### Implement Process Management Enhancements

 - **Error Handling Enhancements**:
   - Added new error variants `BumpSequence`, `StartReportTask`, `ReportProcess`, and `BuildProcessManager` in `error.rs` to improve error handling for process management tasks.
   - Updated `ErrorExt` implementations to handle new error types.

 - **Process Manager Improvements**:
   - Introduced `ProcessManager` enhancements in `process_manager.rs` to manage process states using `ProcessWithState` and `ProcessState` enums.
   - Implemented periodic task `ReportTask` to report running queries to the KV backend.
   - Modified `register_query` and `deregister_query` methods to use the new state management system.

 - **Testing and Validation**:
   - Updated tests in `process_manager.rs` to validate new process management logic.
   - Replaced `dump` method with `list_all_processes` for listing processes.

 - **Integration with Frontend and Standalone**:
   - Updated `frontend.rs` and `standalone.rs` to handle `ProcessManager` initialization errors using `BuildProcessManager` error variant.

 - **Schema Adjustments**:
   - Modified `process_list.rs` in `system_schema/information_schema` to use the updated process listing method.

 - **Key-Value Conversion**:
   - Added `TryFrom` implementation for converting `Process` to `KeyValue` in `process_list.rs`.

* chore: remove register

* fix: sqlness tests

* merge main

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 - **Update `greptime-proto` Dependency**: Updated the `greptime-proto` dependency in `Cargo.lock` and `Cargo.toml` to a new revision.
 - **Refactor `ProcessManager`**: Simplified the `ProcessManager` implementation by removing the use of `KvBackendRef` and `SequenceRef`, and replaced them with `AtomicU64` and `RwLock` for managing process IDs and catalogs in `process_manager.rs`.
 - **Remove Process List Metadata**: Deleted the `process_list.rs` file and removed related metadata key definitions in `key.rs`.
 - **Update Process List Logic**: Modified the process list logic in `process_list.rs` to use the new `ProcessManager` structure.
 - **Adjust Frontend and Standalone Start Commands**: Updated `frontend.rs` and `standalone.rs` to use the new `ProcessManager` constructor.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 - **Update `greptime-proto` Dependency**: Updated the `greptime-proto` dependency version in `Cargo.lock` and `Cargo.toml` to a new commit hash.
 - **Refactor Error Handling**: Removed unused error variants and added a new `ParseProcessId` error in `src/catalog/src/error.rs`.
 - **Enhance Process Management**: Introduced `DisplayProcessId` struct for better process ID representation and parsing in `src/catalog/src/process_manager.rs`.
 - **Revise Process List Schema**: Updated the schema and logic for process listing in `src/catalog/src/system_schema/information_schema/process_list.rs` to include new fields like `client` and `frontend`.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Commit Message

 **Enhancements and Refactoring**

 - **Process Management:**
   - Refactored `ProcessManager` to list local processes with an optional catalog filter in `process_manager.rs`.
   - Updated related tests in `process_manager.rs` and `process_list.rs`.

 - **Client Enhancements:**
   - Added `frontend_client` method in `client.rs` to support gRPC communication with the frontend.

 - **Error Handling:**
   - Extended error handling in `error.rs` to include gRPC and Meta errors.

 - **Frontend Module:**
   - Introduced `selector.rs` for frontend client selection and process listing.
   - Updated `Cargo.toml` to include new dependencies and dev-dependencies.

 - **gRPC Server:**
   - Integrated `FrontendServer` in `builder.rs` for enhanced gRPC server capabilities.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Commit Message

 **Refactor Process Management and Frontend Integration**

 - **Add `common-frontend` Dependency**:
   - Updated `Cargo.lock`, `Cargo.toml` files to include `common-frontend` as a dependency.

 - **Refactor Process Management**:
   - Moved `ProcessManager` trait and `DisplayProcessId` struct to `common-frontend`.
   - Updated `process_manager.rs` to use `MetaProcessManager` and `ProcessManagerRef`.
   - Removed `ParseProcessId` error variant from `error.rs` in `catalog` and `frontend`.

 - **Frontend gRPC Service**:
   - Added `frontend_grpc_handler.rs` to handle gRPC requests for frontend processes.
   - Updated `grpc.rs` and `builder.rs` to integrate `FrontendGrpcHandler`.

 - **Update Tests**:
   - Modified tests in `process_manager.rs` to align with new `ProcessManager` implementation.

 - **Remove Unused Code**:
   - Removed `DisplayProcessId` and related parsing logic from `process_manager.rs`.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Add `MetaClientRef` to `MetaProcessManager` and Update Instantiation

 - **Files Modified**:
   - `src/catalog/src/process_manager.rs`
   - `src/cmd/src/frontend.rs`
   - `src/cmd/src/standalone.rs`

 - **Key Changes**:
   - Added `MetaClientRef` as an optional parameter to the `MetaProcessManager::new` method.
   - Updated instantiation of `MetaProcessManager` to include `MetaClientRef` where applicable.

 ### Update `ProcessManagerRef` Usage

 - **Files Modified**:
   - `src/catalog/src/kvbackend/manager.rs`
   - `src/catalog/src/system_schema/information_schema.rs`
   - `src/catalog/src/system_schema/information_schema/process_list.rs`
   - `src/frontend/src/instance.rs`
   - `src/frontend/src/instance/builder.rs`

 - **Key Changes**:
   - Ensured consistent usage of `ProcessManagerRef` across various modules.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ## Refactor Process Management

 - **Unified Process Manager**:
   - Replaced `MetaProcessManager` with `ProcessManager` across the codebase.
   - Updated `ProcessManager` to use `Arc` for shared references and introduced a `Ticket` struct for query registration and deregistration.
   - Affected files: `manager.rs`, `process_manager.rs`, `frontend.rs`, `standalone.rs`, `frontend_grpc_handler.rs`, `instance.rs`, `builder.rs`, `cluster.rs`, `standalone.rs`.

 - **Stream Wrapper Implementation**:
   - Added `StreamWrapper` to handle record batch streams with process management.
   - Affected file: `stream_wrapper.rs`.

 - **Test Adjustments**:
   - Updated tests to align with the new `ProcessManager` implementation.
   - Affected file: `tests-integration/src/cluster.rs`, `tests-integration/src/standalone.rs`.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Add Error Handling and Process Management

 - **Error Handling Enhancements**:
   - Added new error variants `ListProcess` and `CreateChannel` in `error.rs` to handle specific gRPC service invocation failures.
   - Updated error handling in `selector.rs` to use the new error variants for better context and error propagation.

 - **Process Management Integration**:
   - Introduced `process_manager` method in `instance.rs` to access the process manager.
   - Integrated `FrontendGrpcHandler` with process management in `server.rs` to handle gRPC requests related to process management.

 - **gRPC Server Enhancements**:
   - Made `frontend_grpc_handler` public in `grpc.rs` to allow external access and integration with other modules.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 Update `greptime-proto` dependency and enhance process management

 - **Dependency Update**: Updated `greptime-proto` in `Cargo.lock` and `Cargo.toml` to a new revision.
 - **Process Management**:
   - Modified `process_manager.rs` to include catalog filtering in `list_process`.
   - Updated `frontend_grpc_handler.rs` to handle catalog filtering in `list_process` requests.
 - **System Schema**: Added a TODO comment in `process_list.rs` for future user catalog filtering implementation.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 - **Update Workspace Dependencies**:
   - Modified `Cargo.toml` files in `src/catalog`, `src/common/frontend`, and `src/servers` to adjust workspace dependencies.

 - **Refactor `ProcessManager` Logic**:
   - Updated `process_manager.rs` to simplify the condition in the `select` method.

 - **Remove Unused Error Variants**:
   - Deleted `BuildProcessManager` error variant from `error.rs` in `src/cmd`.
   - Removed `InvalidProcessKey` error variant from `error.rs` in `src/common/meta`.

 - **Add License Header**:
   - Added Apache License header to `stream_wrapper.rs` in `src/frontend`.

 - **Update Test Results**:
   - Adjusted expected results in `information_schema.result` to reflect changes in the schema.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Add Error Handling for Process Listing

 - **`src/catalog/src/error.rs`**: Introduced a new error variant `ListProcess` to handle failures in listing frontend nodes.
 - **`src/catalog/src/process_manager.rs`**: Updated `local_processes` and `list_all_processes` methods to return the new error type, adding context for error handling.
 - **`src/catalog/src/system_schema/information_schema/process_list.rs`**: Modified `make_process_list` to propagate errors using the new error handling mechanism.
 - **`src/servers/src/grpc/frontend_grpc_handler.rs`**: Enhanced error handling in the `list_process` method to log errors and return appropriate gRPC status codes.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 Update `greptime-proto` Dependency and Remove `frontend_client` Method

 - **Cargo.lock** and **Cargo.toml**: Updated the `greptime-proto` dependency to a new revision (`5f6119ac7952878d39dcde0343c4bf828d18ffc8`).
 - **src/client/src/client.rs**: Removed the `frontend_client` method from the `Client` implementation.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Add Query Registration with Pre-Generated ID

 - **`process_manager.rs`**: Introduced `register_query_with_id` method to allow registering queries with a pre-generated ID. This includes creating a `ProcessInfo` instance and inserting it into the catalog. Added `next_id` method to generate the next process ID.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Update Process List Retrieval Method

 - **File**: `process_list.rs`
   - Updated the method for retrieving process lists from `local_processes` to `list_all_processes` to support asynchronous operations.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* feat/show-process-list:
 ### Update error handling in `error.rs`

 - Refined status code handling for `CreateChannel` error by delegating to `source.status_code()`.
 - Separated `ListProcess` and `CreateChannel` error handling for clarity.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

---------

Signed-off-by: Lei, HUANG <lhuang@greptime.com>
2025-06-12 06:55:22 +00:00
Yiran
f4c3950f57 fix: doc links (#6304)
Signed-off-by: Yiran <cuiyiran3@gmail.com>
2025-06-12 03:18:26 +00:00
liyang
88c4409df4 ci: use the new meta backendStorage etcd structure (#6303)
Signed-off-by: liyang <daviderli614@gmail.com>
2025-06-12 03:17:32 +00:00
localhost
c10b8f8474 chore: add failover cache for pipeline table (#6284)
* chore: add second level cache for pipeline table

* chore: change pipeline failover cache name

* chore: add counter metrics for get pipeline operate
2025-06-12 03:15:02 +00:00
shuiyisong
041b683a8d refactor: remove PipelineMap and use Value instead (#6278)
* refactor: remove pipeline_map and use value instead

* chore: remove unused comments

* chore: move error to illegal state
2025-06-11 17:02:32 +00:00
Weny Xu
03bb6e4f28 feat(cli): add metadata get commands (#6299)
* refactor(cli): restructure cli modules and commands

Signed-off-by: WenyXu <wenymedia@gmail.com>

* feat(cli): add metadata get commands

Signed-off-by: WenyXu <wenymedia@gmail.com>

* feat(cli): enhance table metadata query capabilities

Signed-off-by: WenyXu <wenymedia@gmail.com>

* refactor: minor refactor

Signed-off-by: WenyXu <wenymedia@gmail.com>

---------

Signed-off-by: WenyXu <wenymedia@gmail.com>
2025-06-11 16:33:36 +00:00
discord9
09e5a6580f chore: silence clippy (#6298)
Signed-off-by: discord9 <discord9@163.com>
2025-06-11 14:32:41 +00:00
Lei, HUANG
f9f905ae14 fix: config docs (#6294)
fix/config-docs:
 Update `config.md` to specify default compression mode

 - Added default value `none` for `grpc.flight_compression` in both frontend and datanode sections of `config/config.md`.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>
2025-06-11 07:31:42 +00:00
Lei, HUANG
1d53dd26ae chore: add option for arrow flight compression mode (#6283)
* chore/enable-flight-encoder:
 ### Add Flight Compression Support

 - **Configuration Updates**:
   - Added `grpc.flight_compression` option to `config/config.md`, `config/datanode.example.toml`, and `config/frontend.example.toml` to specify compression modes for Arrow IPC service.

 - **Code Enhancements**:
   - Updated `FlightEncoder` in `src/common/grpc/src/flight.rs` to support compression modes.
   - Modified `RegionServer` and `DatanodeBuilder` in `src/datanode/src/datanode.rs` and `src/datanode/src/region_server.rs` to handle `FlightCompression`.
   - Integrated `FlightCompression` in `src/servers/src/grpc.rs` and `src/servers/src/grpc/flight.rs` to manage compression settings.

 - **Testing and Integration**:
   - Updated test utilities and integration tests in `tests-integration/src/grpc/flight.rs` and `tests-integration/src/test_util.rs` to include `FlightCompression`.

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>

* chore/enable-flight-encoder:
 ### Enable Compression in FlightClient

 - **`client.rs`**: Updated `make_flight_client` to accept `send_compression` and `accept_compression` parameters, enabling Zstd compression for sending and receiving messages.
 - **`client_manager.rs`**: Modified `datanode` method to pass compression settings from `ChannelConfig` to `RegionRequester`.
 - **`database.rs`**: Adjusted calls to `make_flight_client` to include compression parameters.
 - **`region.rs`**: Updated `RegionRequester` to store and utilize compression settings.
 - **`frontend.rs`**: Configured `ChannelConfig` to enable compression based on options.
 - **`channel_manager.rs`**: Added `send_compression` and `accept_compression` fields to `ChannelConfig` with default values and updated tests accordingly.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

* chore/enable-flight-encoder:
 ### Update Compression Defaults and Documentation

 - **Configuration Files**: Updated `datanode.example.toml` and `frontend.example.toml` to include a default setting comment for `flight_compression`, specifying it defaults to `none`.
 - **gRPC Server Code**: Modified `grpc.rs` to set `None` as the default for `FlightCompression` instead of `ArrowIpc`.

Signed-off-by: Lei, HUANG <lhuang@greptime.com>

---------

Signed-off-by: Lei, HUANG <mrsatangel@gmail.com>
Signed-off-by: Lei, HUANG <lhuang@greptime.com>
2025-06-11 06:54:10 +00:00
localhost
01796c9cc0 chore: org cli sub command (#6265)
* chore: org cli sub command

* chore: make clippy happy

* chore: fix info command not support absolute path

* chore: fix cli test

* Apply suggestions from code review

Co-authored-by: Weny Xu <wenymedia@gmail.com>

* chore: reorganizing the cli tool

* chore: fix limit issue

* chore: add some doc for cli

* chore: format code

---------

Co-authored-by: Weny Xu <wenymedia@gmail.com>
2025-06-11 03:34:56 +00:00
liyang
9469a8f8f2 ci: add signature information when updating downstream repository (#6282)
Signed-off-by: liyang <daviderli614@gmail.com>
2025-06-10 17:18:29 +00:00
Ruihang Xia
2fabe346a1 fix: null value handling on PromQL's join (#6289)
* fix: null value handling on PromQL's join

Signed-off-by: Ruihang Xia <waynestxia@gmail.com>

* update sqlness comment

Signed-off-by: Ruihang Xia <waynestxia@gmail.com>

---------

Signed-off-by: Ruihang Xia <waynestxia@gmail.com>
2025-06-10 13:24:09 +00:00
Zhenchi
c26138963e refactor: unify function registry (Part 1) (#6262)
* refactor: unify function registry (Part 1)

Signed-off-by: Zhenchi <zhongzc_arch@outlook.com>

* refactor: simplify via register_scalar

Signed-off-by: Zhenchi <zhongzc_arch@outlook.com>

---------

Signed-off-by: Zhenchi <zhongzc_arch@outlook.com>
2025-06-10 10:11:06 +00:00
jeremyhi
12648f388a feat: refactor grpc options of metasrv (#6275)
* feat: refactor grpc options of metasrv

* fix: unit test

* feat: config update
2025-06-10 01:35:57 +00:00
Weny Xu
2979aa048e fix(meta): enhance postgres election client with timeouts and reconnection (#6276)
* fix(meta): enhance postgres election client with timeouts and reconnection

Signed-off-by: WenyXu <wenymedia@gmail.com>

* chore: apply suggestions from CR

Signed-off-by: WenyXu <wenymedia@gmail.com>

* chore: apply suggestions from CR

Signed-off-by: WenyXu <wenymedia@gmail.com>

---------

Signed-off-by: WenyXu <wenymedia@gmail.com>
2025-06-09 13:51:13 +00:00
Weny Xu
74222c3070 chore: improve CI debugging and resource configuration (#6274)
* chore: describe pods on CI failure

Signed-off-by: WenyXu <wenymedia@gmail.com>

* chore: increase memory limit for main pod template from 2Gi to 3Gi

Signed-off-by: WenyXu <wenymedia@gmail.com>

---------

Signed-off-by: WenyXu <wenymedia@gmail.com>
2025-06-09 11:40:32 +00:00
dennis zhuang
0311db3089 fix: export metrics settings in sample config (#6170)
* fix: export metrics settings in sample config

Signed-off-by: Dennis Zhuang <killme2008@gmail.com>

* chore: forgot update docs

Signed-off-by: Dennis Zhuang <killme2008@gmail.com>

* fix: test loading example configs

Signed-off-by: Dennis Zhuang <killme2008@gmail.com>

---------

Signed-off-by: Dennis Zhuang <killme2008@gmail.com>
2025-06-09 10:53:01 +00:00
zyy17
e434294a0c refactor: support to get trace id with time range (#6236)
* refactor: add time range for getting trace id

* test: add unit test
2025-06-09 08:53:20 +00:00
286 changed files with 6384 additions and 2118 deletions

View File

@@ -10,13 +10,13 @@ inputs:
meta-replicas:
default: 2
description: "Number of Metasrv replicas"
image-registry:
image-registry:
default: "docker.io"
description: "Image registry"
image-repository:
image-repository:
default: "greptime/greptimedb"
description: "Image repository"
image-tag:
image-tag:
default: "latest"
description: 'Image tag'
etcd-endpoints:
@@ -32,12 +32,12 @@ runs:
steps:
- name: Install GreptimeDB operator
uses: nick-fields/retry@v3
with:
with:
timeout_minutes: 3
max_attempts: 3
shell: bash
command: |
helm repo add greptime https://greptimeteam.github.io/helm-charts/
helm repo add greptime https://greptimeteam.github.io/helm-charts/
helm repo update
helm upgrade \
--install \
@@ -48,10 +48,10 @@ runs:
--wait-for-jobs
- name: Install GreptimeDB cluster
shell: bash
run: |
run: |
helm upgrade \
--install my-greptimedb \
--set meta.etcdEndpoints=${{ inputs.etcd-endpoints }} \
--set meta.backendStorage.etcd.endpoints=${{ inputs.etcd-endpoints }} \
--set meta.enableRegionFailover=${{ inputs.enable-region-failover }} \
--set image.registry=${{ inputs.image-registry }} \
--set image.repository=${{ inputs.image-repository }} \
@@ -59,7 +59,7 @@ runs:
--set base.podTemplate.main.resources.requests.cpu=50m \
--set base.podTemplate.main.resources.requests.memory=256Mi \
--set base.podTemplate.main.resources.limits.cpu=2000m \
--set base.podTemplate.main.resources.limits.memory=2Gi \
--set base.podTemplate.main.resources.limits.memory=3Gi \
--set frontend.replicas=${{ inputs.frontend-replicas }} \
--set datanode.replicas=${{ inputs.datanode-replicas }} \
--set meta.replicas=${{ inputs.meta-replicas }} \
@@ -72,7 +72,7 @@ runs:
- name: Wait for GreptimeDB
shell: bash
run: |
while true; do
while true; do
PHASE=$(kubectl -n my-greptimedb get gtc my-greptimedb -o jsonpath='{.status.clusterPhase}')
if [ "$PHASE" == "Running" ]; then
echo "Cluster is ready"
@@ -86,10 +86,10 @@ runs:
- name: Print GreptimeDB info
if: always()
shell: bash
run: |
run: |
kubectl get all --show-labels -n my-greptimedb
- name: Describe Nodes
if: always()
shell: bash
run: |
run: |
kubectl describe nodes

View File

@@ -68,7 +68,7 @@ function deploy_greptimedb_cluster() {
helm install "$cluster_name" greptime/greptimedb-cluster \
--set image.tag="$GREPTIMEDB_IMAGE_TAG" \
--set meta.etcdEndpoints="etcd.$install_namespace:2379" \
--set meta.backendStorage.etcd.endpoints="etcd.$install_namespace:2379" \
-n "$install_namespace"
# Wait for greptimedb cluster to be ready.
@@ -103,7 +103,7 @@ function deploy_greptimedb_cluster_with_s3_storage() {
helm install "$cluster_name" greptime/greptimedb-cluster -n "$install_namespace" \
--set image.tag="$GREPTIMEDB_IMAGE_TAG" \
--set meta.etcdEndpoints="etcd.$install_namespace:2379" \
--set meta.backendStorage.etcd.endpoints="etcd.$install_namespace:2379" \
--set storage.s3.bucket="$AWS_CI_TEST_BUCKET" \
--set storage.s3.region="$AWS_REGION" \
--set storage.s3.root="$DATA_ROOT" \

View File

@@ -30,7 +30,7 @@ update_helm_charts_version() {
# Commit the changes.
git add .
git commit -m "chore: Update GreptimeDB version to ${VERSION}"
git commit -s -m "chore: Update GreptimeDB version to ${VERSION}"
git push origin $BRANCH_NAME
# Create a Pull Request.

View File

@@ -26,7 +26,7 @@ update_homebrew_greptime_version() {
# Commit the changes.
git add .
git commit -m "chore: Update GreptimeDB version to ${VERSION}"
git commit -s -m "chore: Update GreptimeDB version to ${VERSION}"
git push origin $BRANCH_NAME
# Create a Pull Request.

View File

@@ -250,6 +250,11 @@ jobs:
name: unstable-fuzz-logs
path: /tmp/unstable-greptime/
retention-days: 3
- name: Describe pods
if: failure()
shell: bash
run: |
kubectl describe pod -n my-greptimedb
build-greptime-ci:
if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
@@ -405,6 +410,11 @@ jobs:
shell: bash
run: |
kubectl describe nodes
- name: Describe pod
if: failure()
shell: bash
run: |
kubectl describe pod -n my-greptimedb
- name: Export kind logs
if: failure()
shell: bash
@@ -554,6 +564,11 @@ jobs:
shell: bash
run: |
kubectl describe nodes
- name: Describe pods
if: failure()
shell: bash
run: |
kubectl describe pod -n my-greptimedb
- name: Export kind logs
if: failure()
shell: bash

112
Cargo.lock generated
View File

@@ -1621,8 +1621,10 @@ dependencies = [
"cache",
"catalog",
"chrono",
"common-base",
"common-catalog",
"common-error",
"common-frontend",
"common-macro",
"common-meta",
"common-procedure",
@@ -2293,8 +2295,14 @@ version = "0.15.0"
dependencies = [
"async-trait",
"common-error",
"common-grpc",
"common-macro",
"common-meta",
"greptime-proto",
"meta-client",
"snafu 0.8.5",
"tokio",
"tonic 0.12.3",
]
[[package]]
@@ -3252,7 +3260,7 @@ dependencies = [
[[package]]
name = "datafusion"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"arrow-array 54.2.1",
@@ -3303,7 +3311,7 @@ dependencies = [
[[package]]
name = "datafusion-catalog"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"async-trait",
@@ -3323,7 +3331,7 @@ dependencies = [
[[package]]
name = "datafusion-catalog-listing"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"arrow-schema 54.3.1",
@@ -3346,7 +3354,7 @@ dependencies = [
[[package]]
name = "datafusion-common"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"ahash 0.8.11",
"arrow 54.2.1",
@@ -3371,7 +3379,7 @@ dependencies = [
[[package]]
name = "datafusion-common-runtime"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"log",
"tokio",
@@ -3380,12 +3388,12 @@ dependencies = [
[[package]]
name = "datafusion-doc"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
[[package]]
name = "datafusion-execution"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"dashmap",
@@ -3403,7 +3411,7 @@ dependencies = [
[[package]]
name = "datafusion-expr"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"chrono",
@@ -3423,7 +3431,7 @@ dependencies = [
[[package]]
name = "datafusion-expr-common"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"datafusion-common",
@@ -3434,7 +3442,7 @@ dependencies = [
[[package]]
name = "datafusion-functions"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"arrow-buffer 54.3.1",
@@ -3463,7 +3471,7 @@ dependencies = [
[[package]]
name = "datafusion-functions-aggregate"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"ahash 0.8.11",
"arrow 54.2.1",
@@ -3484,7 +3492,7 @@ dependencies = [
[[package]]
name = "datafusion-functions-aggregate-common"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"ahash 0.8.11",
"arrow 54.2.1",
@@ -3496,7 +3504,7 @@ dependencies = [
[[package]]
name = "datafusion-functions-nested"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"arrow-array 54.2.1",
@@ -3518,7 +3526,7 @@ dependencies = [
[[package]]
name = "datafusion-functions-table"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"async-trait",
@@ -3533,7 +3541,7 @@ dependencies = [
[[package]]
name = "datafusion-functions-window"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"datafusion-common",
"datafusion-doc",
@@ -3549,7 +3557,7 @@ dependencies = [
[[package]]
name = "datafusion-functions-window-common"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"datafusion-common",
"datafusion-physical-expr-common",
@@ -3558,7 +3566,7 @@ dependencies = [
[[package]]
name = "datafusion-macros"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"datafusion-expr",
"quote",
@@ -3568,7 +3576,7 @@ dependencies = [
[[package]]
name = "datafusion-optimizer"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"chrono",
@@ -3586,7 +3594,7 @@ dependencies = [
[[package]]
name = "datafusion-physical-expr"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"ahash 0.8.11",
"arrow 54.2.1",
@@ -3609,7 +3617,7 @@ dependencies = [
[[package]]
name = "datafusion-physical-expr-common"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"ahash 0.8.11",
"arrow 54.2.1",
@@ -3622,7 +3630,7 @@ dependencies = [
[[package]]
name = "datafusion-physical-optimizer"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"arrow-schema 54.3.1",
@@ -3643,7 +3651,7 @@ dependencies = [
[[package]]
name = "datafusion-physical-plan"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"ahash 0.8.11",
"arrow 54.2.1",
@@ -3673,7 +3681,7 @@ dependencies = [
[[package]]
name = "datafusion-sql"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"arrow 54.2.1",
"arrow-array 54.2.1",
@@ -3691,7 +3699,7 @@ dependencies = [
[[package]]
name = "datafusion-substrait"
version = "45.0.0"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=12c0381babd52c681043957e9d6ee083a03f7646#12c0381babd52c681043957e9d6ee083a03f7646"
dependencies = [
"async-recursion",
"async-trait",
@@ -4511,9 +4519,9 @@ dependencies = [
[[package]]
name = "flate2"
version = "1.1.1"
version = "1.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7ced92e76e966ca2fd84c8f7aa01a4aea65b0eb6648d72f7c8f3e2764a67fece"
checksum = "4a3d7db9596fecd151c5f638c0ee5d5bd487b6e0ea232e5dc96d5250f6f94b1d"
dependencies = [
"crc32fast",
"libz-rs-sys",
@@ -4701,6 +4709,7 @@ dependencies = [
"common-config",
"common-datasource",
"common-error",
"common-frontend",
"common-function",
"common-grpc",
"common-macro",
@@ -4747,6 +4756,7 @@ dependencies = [
"substrait 0.15.0",
"table",
"tokio",
"tokio-util",
"toml 0.8.19",
"tonic 0.12.3",
"tower 0.5.2",
@@ -5133,7 +5143,7 @@ dependencies = [
[[package]]
name = "greptime-proto"
version = "0.1.0"
source = "git+https://github.com/GreptimeTeam/greptime-proto.git?rev=454c52634c3bac27de10bf0d85d5533eed1cf03f#454c52634c3bac27de10bf0d85d5533eed1cf03f"
source = "git+https://github.com/GreptimeTeam/greptime-proto.git?rev=17971523673f4fbc982510d3c9d6647ff642e16f#17971523673f4fbc982510d3c9d6647ff642e16f"
dependencies = [
"prost 0.13.5",
"serde",
@@ -5146,9 +5156,9 @@ dependencies = [
[[package]]
name = "grok"
version = "2.0.0"
version = "2.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "273797968160270573071022613fc4aa28b91fe68f3eef6c96a1b2a1947ddfbd"
checksum = "6c52724b609896f661a3f4641dd3a44dc602958ef615857c12d00756b4e9355b"
dependencies = [
"glob",
"onig",
@@ -6716,9 +6726,9 @@ dependencies = [
[[package]]
name = "libz-rs-sys"
version = "0.5.0"
version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6489ca9bd760fe9642d7644e827b0c9add07df89857b0416ee15c1cc1a3b8c5a"
checksum = "172a788537a2221661b480fee8dc5f96c580eb34fa88764d3205dc356c7e4221"
dependencies = [
"zlib-rs",
]
@@ -7240,6 +7250,7 @@ dependencies = [
"humantime-serde",
"itertools 0.14.0",
"lazy_static",
"mito-codec",
"mito2",
"mur3",
"object-store",
@@ -7305,6 +7316,29 @@ dependencies = [
"windows-sys 0.52.0",
]
[[package]]
name = "mito-codec"
version = "0.15.0"
dependencies = [
"api",
"bytes",
"common-base",
"common-decimal",
"common-error",
"common-macro",
"common-recordbatch",
"common-telemetry",
"common-time",
"datafusion-common",
"datafusion-expr",
"datatypes",
"memcomparable",
"paste",
"serde",
"snafu 0.8.5",
"store-api",
]
[[package]]
name = "mito2"
version = "0.15.0"
@@ -7347,6 +7381,7 @@ dependencies = [
"lazy_static",
"log-store",
"memcomparable",
"mito-codec",
"moka",
"object-store",
"parquet",
@@ -8386,6 +8421,7 @@ dependencies = [
"common-catalog",
"common-datasource",
"common-error",
"common-frontend",
"common-function",
"common-grpc",
"common-grpc-expr",
@@ -8885,9 +8921,8 @@ dependencies = [
[[package]]
name = "pgwire"
version = "0.30.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ec79ee18e6cafde8698885646780b967ecc905120798b8359dd0da64f9688e89"
version = "0.30.2"
source = "git+https://github.com/sunng87/pgwire?rev=127573d997228cfb70c7699881c568eae8131270#127573d997228cfb70c7699881c568eae8131270"
dependencies = [
"async-trait",
"bytes",
@@ -9664,9 +9699,9 @@ dependencies = [
[[package]]
name = "psl"
version = "2.1.112"
version = "2.1.119"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1c6b4c497a0c6bfb466f75167c728b1a861b0cdc39de9c35b877208a270a9590"
checksum = "d0e49aa528239f2ca13ad87387977c208e59c3fb8c437609f95f1b3898ec6ef1"
dependencies = [
"psl-types",
]
@@ -11135,6 +11170,7 @@ dependencies = [
"common-catalog",
"common-config",
"common-error",
"common-frontend",
"common-grpc",
"common-macro",
"common-mem-prof",
@@ -14701,9 +14737,9 @@ dependencies = [
[[package]]
name = "zlib-rs"
version = "0.5.0"
version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "868b928d7949e09af2f6086dfc1e01936064cc7a819253bce650d4e2a2d63ba8"
checksum = "626bd9fa9734751fc50d6060752170984d7053f5a39061f524cda68023d4db8a"
[[package]]
name = "zstd"

View File

@@ -49,6 +49,7 @@ members = [
"src/meta-client",
"src/meta-srv",
"src/metric-engine",
"src/mito-codec",
"src/mito2",
"src/object-store",
"src/operator",
@@ -116,15 +117,15 @@ clap = { version = "4.4", features = ["derive"] }
config = "0.13.0"
crossbeam-utils = "0.8"
dashmap = "6.1"
datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
deadpool = "0.12"
deadpool-postgres = "0.14"
derive_builder = "0.20"
@@ -133,7 +134,7 @@ etcd-client = "0.14"
fst = "0.4.7"
futures = "0.3"
futures-util = "0.3"
greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "454c52634c3bac27de10bf0d85d5533eed1cf03f" }
greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "17971523673f4fbc982510d3c9d6647ff642e16f" }
hex = "0.4"
http = "1"
humantime = "2.1"
@@ -274,6 +275,7 @@ log-store = { path = "src/log-store" }
meta-client = { path = "src/meta-client" }
meta-srv = { path = "src/meta-srv" }
metric-engine = { path = "src/metric-engine" }
mito-codec = { path = "src/mito-codec" }
mito2 = { path = "src/mito2" }
object-store = { path = "src/object-store" }
operator = { path = "src/operator" }

View File

@@ -195,13 +195,13 @@
| `slow_query.record_type` | String | Unset | The record type of slow queries. It can be `system_table` or `log`. |
| `slow_query.threshold` | String | Unset | The threshold of slow query. |
| `slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics` | -- | -- | The standalone can export its metrics and send to Prometheus compatible service (e.g. `greptimedb`) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
| `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommended to collect metrics generated by itself<br/>You must create the database before enabling it. |
| `export_metrics.self_import.db` | String | Unset | -- |
| `export_metrics.remote_write` | -- | -- | -- |
| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
| `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -232,6 +232,7 @@
| `grpc.bind_addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
| `grpc.server_addr` | String | `127.0.0.1:4001` | The address advertised to the metasrv, and used for connections from outside the host.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `grpc.bind_addr`. |
| `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
| `grpc.flight_compression` | String | `arrow_ipc` | Compression mode for frontend side Arrow IPC service. Available options:<br/>- `none`: disable all compression<br/>- `transport`: only enable gRPC transport compression (zstd)<br/>- `arrow_ipc`: only enable Arrow IPC compression (lz4)<br/>- `all`: enable all compression.<br/>Default to `none` |
| `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
| `grpc.tls.mode` | String | `disable` | TLS mode. |
| `grpc.tls.cert_path` | String | Unset | Certificate file path. |
@@ -298,13 +299,11 @@
| `slow_query.threshold` | String | `30s` | The threshold of slow query. It can be human readable time string, for example: `10s`, `100ms`, `1s`. |
| `slow_query.sample_ratio` | Float | `1.0` | The sampling ratio of slow query log. The value should be in the range of (0, 1]. For example, `0.1` means 10% of the slow queries will be logged and `1.0` means all slow queries will be logged. |
| `slow_query.ttl` | String | `30d` | The TTL of the `slow_queries` system table. Default is `30d` when `record_type` is `system_table`. |
| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics` | -- | -- | The frontend can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
| `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommend to collect metrics generated by itself<br/>You must create the database before enabling it. |
| `export_metrics.self_import.db` | String | Unset | -- |
| `export_metrics.remote_write` | -- | -- | -- |
| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
| `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -315,11 +314,9 @@
| Key | Type | Default | Descriptions |
| --- | -----| ------- | ----------- |
| `data_home` | String | `./greptimedb_data` | The working home directory. |
| `bind_addr` | String | `127.0.0.1:3002` | The bind address of metasrv. |
| `server_addr` | String | `127.0.0.1:3002` | The communication server address for the frontend and datanode to connect to metasrv.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `bind_addr`. |
| `store_addrs` | Array | -- | Store server address default to etcd store.<br/>For postgres store, the format is:<br/>"password=password dbname=postgres user=postgres host=localhost port=5432"<br/>For etcd store, the format is:<br/>"127.0.0.1:2379" |
| `store_key_prefix` | String | `""` | If it's not empty, the metasrv will store all data with this key prefix. |
| `backend` | String | `etcd_store` | The datastore for meta server.<br/>Available values:<br/>- `etcd_store` (default value)<br/>- `memory_store`<br/>- `postgres_store` |
| `backend` | String | `etcd_store` | The datastore for meta server.<br/>Available values:<br/>- `etcd_store` (default value)<br/>- `memory_store`<br/>- `postgres_store`<br/>- `mysql_store` |
| `meta_table_name` | String | `greptime_metakv` | Table name in RDS to store metadata. Effect when using a RDS kvbackend.<br/>**Only used when backend is `postgres_store`.** |
| `meta_election_lock_id` | Integer | `1` | Advisory lock id in PostgreSQL for election. Effect when using PostgreSQL as kvbackend<br/>Only used when backend is `postgres_store`. |
| `selector` | String | `round_robin` | Datanode selector type.<br/>- `round_robin` (default value)<br/>- `lease_based`<br/>- `load_based`<br/>For details, please see "https://docs.greptime.com/developer-guide/metasrv/selector". |
@@ -331,6 +328,12 @@
| `runtime` | -- | -- | The runtime options. |
| `runtime.global_rt_size` | Integer | `8` | The number of threads to execute the runtime for global read operations. |
| `runtime.compact_rt_size` | Integer | `4` | The number of threads to execute the runtime for global write operations. |
| `grpc` | -- | -- | The gRPC server options. |
| `grpc.bind_addr` | String | `127.0.0.1:3002` | The address to bind the gRPC server. |
| `grpc.server_addr` | String | `127.0.0.1:3002` | The communication server address for the frontend and datanode to connect to metasrv.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `bind_addr`. |
| `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
| `grpc.max_recv_message_size` | String | `512MB` | The maximum receive message size for gRPC server. |
| `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
| `http` | -- | -- | The HTTP server options. |
| `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
| `http.timeout` | String | `0s` | HTTP request timeout. Set to 0 to disable timeout. |
@@ -372,13 +375,11 @@
| `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
| `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
| `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics` | -- | -- | The metasrv can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
| `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommend to collect metrics generated by itself<br/>You must create the database before enabling it. |
| `export_metrics.self_import.db` | String | Unset | -- |
| `export_metrics.remote_write` | -- | -- | -- |
| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
| `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -404,6 +405,7 @@
| `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
| `grpc.max_recv_message_size` | String | `512MB` | The maximum receive message size for gRPC server. |
| `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
| `grpc.flight_compression` | String | `arrow_ipc` | Compression mode for datanode side Arrow IPC service. Available options:<br/>- `none`: disable all compression<br/>- `transport`: only enable gRPC transport compression (zstd)<br/>- `arrow_ipc`: only enable Arrow IPC compression (lz4)<br/>- `all`: enable all compression.<br/>Default to `none` |
| `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
| `grpc.tls.mode` | String | `disable` | TLS mode. |
| `grpc.tls.cert_path` | String | Unset | Certificate file path. |
@@ -536,13 +538,11 @@
| `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
| `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
| `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
| `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
| `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommend to collect metrics generated by itself<br/>You must create the database before enabling it. |
| `export_metrics.self_import.db` | String | Unset | -- |
| `export_metrics.remote_write` | -- | -- | -- |
| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
| `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
| `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |

View File

@@ -44,6 +44,13 @@ runtime_size = 8
max_recv_message_size = "512MB"
## The maximum send message size for gRPC server.
max_send_message_size = "512MB"
## Compression mode for datanode side Arrow IPC service. Available options:
## - `none`: disable all compression
## - `transport`: only enable gRPC transport compression (zstd)
## - `arrow_ipc`: only enable Arrow IPC compression (lz4)
## - `all`: enable all compression.
## Default to `none`
flight_compression = "arrow_ipc"
## gRPC server TLS options, see `mysql.tls` section.
[grpc.tls]
@@ -635,24 +642,16 @@ max_log_files = 720
[logging.tracing_sample_ratio]
default_ratio = 1.0
## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
## The datanode can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.
## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
[export_metrics]
## whether enable export metrics.
enable = false
## The interval of export metrics.
write_interval = "30s"
## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
## You must create the database before enabling it.
[export_metrics.self_import]
## @toml2docs:none-default
db = "greptime_metrics"
[export_metrics.remote_write]
## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
url = ""
## HTTP headers of Prometheus remote-write carry.

View File

@@ -54,6 +54,13 @@ bind_addr = "127.0.0.1:4001"
server_addr = "127.0.0.1:4001"
## The number of server worker threads.
runtime_size = 8
## Compression mode for frontend side Arrow IPC service. Available options:
## - `none`: disable all compression
## - `transport`: only enable gRPC transport compression (zstd)
## - `arrow_ipc`: only enable Arrow IPC compression (lz4)
## - `all`: enable all compression.
## Default to `none`
flight_compression = "arrow_ipc"
## gRPC server TLS options, see `mysql.tls` section.
[grpc.tls]
@@ -247,24 +254,16 @@ sample_ratio = 1.0
## The TTL of the `slow_queries` system table. Default is `30d` when `record_type` is `system_table`.
ttl = "30d"
## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
## The frontend can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.
## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
[export_metrics]
## whether enable export metrics.
enable = false
## The interval of export metrics.
write_interval = "30s"
## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
## You must create the database before enabling it.
[export_metrics.self_import]
## @toml2docs:none-default
db = "greptime_metrics"
[export_metrics.remote_write]
## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
url = ""
## HTTP headers of Prometheus remote-write carry.

View File

@@ -1,14 +1,6 @@
## The working home directory.
data_home = "./greptimedb_data"
## The bind address of metasrv.
bind_addr = "127.0.0.1:3002"
## The communication server address for the frontend and datanode to connect to metasrv.
## If left empty or unset, the server will automatically use the IP address of the first network interface
## on the host, with the same port number as the one specified in `bind_addr`.
server_addr = "127.0.0.1:3002"
## Store server address default to etcd store.
## For postgres store, the format is:
## "password=password dbname=postgres user=postgres host=localhost port=5432"
@@ -24,6 +16,7 @@ store_key_prefix = ""
## - `etcd_store` (default value)
## - `memory_store`
## - `postgres_store`
## - `mysql_store`
backend = "etcd_store"
## Table name in RDS to store metadata. Effect when using a RDS kvbackend.
@@ -67,6 +60,21 @@ node_max_idle_time = "24hours"
## The number of threads to execute the runtime for global write operations.
#+ compact_rt_size = 4
## The gRPC server options.
[grpc]
## The address to bind the gRPC server.
bind_addr = "127.0.0.1:3002"
## The communication server address for the frontend and datanode to connect to metasrv.
## If left empty or unset, the server will automatically use the IP address of the first network interface
## on the host, with the same port number as the one specified in `bind_addr`.
server_addr = "127.0.0.1:3002"
## The number of server worker threads.
runtime_size = 8
## The maximum receive message size for gRPC server.
max_recv_message_size = "512MB"
## The maximum send message size for gRPC server.
max_send_message_size = "512MB"
## The HTTP server options.
[http]
## The address to bind the HTTP server.
@@ -229,24 +237,16 @@ max_log_files = 720
[logging.tracing_sample_ratio]
default_ratio = 1.0
## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
## The metasrv can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.
## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
[export_metrics]
## whether enable export metrics.
enable = false
## The interval of export metrics.
write_interval = "30s"
## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
## You must create the database before enabling it.
[export_metrics.self_import]
## @toml2docs:none-default
db = "greptime_metrics"
[export_metrics.remote_write]
## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
url = ""
## HTTP headers of Prometheus remote-write carry.

View File

@@ -750,13 +750,11 @@ default_ratio = 1.0
## @toml2docs:none-default
#+ sample_ratio = 1.0
## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
## The standalone can export its metrics and send to Prometheus compatible service (e.g. `greptimedb`) from remote-write API.
## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
[export_metrics]
## whether enable export metrics.
enable = false
## The interval of export metrics.
write_interval = "30s"
@@ -767,7 +765,7 @@ write_interval = "30s"
db = "greptime_metrics"
[export_metrics.remote_write]
## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
url = ""
## HTTP headers of Prometheus remote-write carry.

View File

@@ -9,7 +9,7 @@ We highly recommend using the self-monitoring feature provided by [GreptimeDB Op
- **Metrics Dashboards**
- `dashboards/metrics/cluster/dashboard.json`: The Grafana dashboard for the GreptimeDB cluster. Read the [dashboard.md](./dashboards/metrics/cluster/dashboard.md) for more details.
- `dashboards/metrics/standalone/dashboard.json`: The Grafana dashboard for the standalone GreptimeDB instance. **It's generated from the `cluster/dashboard.json` by removing the instance filter through the `make dashboards` command**. Read the [dashboard.md](./dashboards/metrics/standalone/dashboard.md) for more details.
- **Logs Dashboard**
@@ -83,7 +83,7 @@ If you use the [Helm Chart](https://github.com/GreptimeTeam/helm-charts) to depl
- `monitoring.enabled=true`: Deploys a standalone GreptimeDB instance dedicated to monitoring the cluster;
- `grafana.enabled=true`: Deploys Grafana and automatically imports the monitoring dashboard;
The standalone GreptimeDB instance will collect metrics from your cluster, and the dashboard will be available in the Grafana UI. For detailed deployment instructions, please refer to our [Kubernetes deployment guide](https://docs.greptime.com/nightly/user-guide/deployments/deploy-on-kubernetes/getting-started).
The standalone GreptimeDB instance will collect metrics from your cluster, and the dashboard will be available in the Grafana UI. For detailed deployment instructions, please refer to our [Kubernetes deployment guide](https://docs.greptime.com/user-guide/deployments-administration/deploy-on-kubernetes/getting-started).
### Self-host Prometheus and import dashboards manually

View File

@@ -17,8 +17,10 @@ arrow-schema.workspace = true
async-stream.workspace = true
async-trait.workspace = true
bytes.workspace = true
common-base.workspace = true
common-catalog.workspace = true
common-error.workspace = true
common-frontend.workspace = true
common-macro.workspace = true
common-meta.workspace = true
common-procedure.workspace = true

View File

@@ -277,6 +277,26 @@ pub enum Error {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Failed to invoke frontend services"))]
InvokeFrontend {
source: common_frontend::error::Error,
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Meta client is not provided"))]
MetaClientMissing {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Failed to find frontend node: {}", addr))]
FrontendNotFound {
addr: String,
#[snafu(implicit)]
location: Location,
},
}
impl Error {
@@ -345,6 +365,10 @@ impl ErrorExt for Error {
Error::GetViewCache { source, .. } | Error::GetTableCache { source, .. } => {
source.status_code()
}
Error::InvokeFrontend { source, .. } => source.status_code(),
Error::FrontendNotFound { .. } | Error::MetaClientMissing { .. } => {
StatusCode::Unexpected
}
}
}

View File

@@ -51,6 +51,7 @@ use crate::error::{
};
use crate::information_schema::{InformationExtensionRef, InformationSchemaProvider};
use crate::kvbackend::TableCacheRef;
use crate::process_manager::ProcessManagerRef;
use crate::system_schema::pg_catalog::PGCatalogProvider;
use crate::system_schema::SystemSchemaProvider;
use crate::CatalogManager;
@@ -84,6 +85,7 @@ impl KvBackendCatalogManager {
backend: KvBackendRef,
cache_registry: LayeredCacheRegistryRef,
procedure_manager: Option<ProcedureManagerRef>,
process_manager: Option<ProcessManagerRef>,
) -> Arc<Self> {
Arc::new_cyclic(|me| Self {
information_extension,
@@ -102,12 +104,14 @@ impl KvBackendCatalogManager {
DEFAULT_CATALOG_NAME.to_string(),
me.clone(),
Arc::new(FlowMetadataManager::new(backend.clone())),
process_manager.clone(),
)),
pg_catalog_provider: Arc::new(PGCatalogProvider::new(
DEFAULT_CATALOG_NAME.to_string(),
me.clone(),
)),
backend,
process_manager,
},
cache_registry,
procedure_manager,
@@ -419,6 +423,7 @@ struct SystemCatalog {
information_schema_provider: Arc<InformationSchemaProvider>,
pg_catalog_provider: Arc<PGCatalogProvider>,
backend: KvBackendRef,
process_manager: Option<ProcessManagerRef>,
}
impl SystemCatalog {
@@ -486,6 +491,7 @@ impl SystemCatalog {
catalog.to_string(),
self.catalog_manager.clone(),
Arc::new(FlowMetadataManager::new(self.backend.clone())),
self.process_manager.clone(),
))
});
information_schema_provider.table(table_name)

View File

@@ -40,6 +40,7 @@ pub mod information_schema {
pub use crate::system_schema::information_schema::*;
}
pub mod process_manager;
pub mod table_source;
#[async_trait::async_trait]

View File

@@ -356,6 +356,7 @@ impl MemoryCatalogManager {
catalog,
Arc::downgrade(self) as Weak<dyn CatalogManager>,
Arc::new(FlowMetadataManager::new(Arc::new(MemoryKvBackend::new()))),
None, // we don't need ProcessManager on regions server.
);
let information_schema = information_schema_provider.tables().clone();

View File

@@ -34,4 +34,20 @@ lazy_static! {
register_histogram!("greptime_catalog_kv_get", "catalog kv get").unwrap();
pub static ref METRIC_CATALOG_KV_BATCH_GET: Histogram =
register_histogram!("greptime_catalog_kv_batch_get", "catalog kv batch get").unwrap();
/// Count of running process in each catalog.
pub static ref PROCESS_LIST_COUNT: IntGaugeVec = register_int_gauge_vec!(
"greptime_process_list_count",
"Running process count per catalog",
&["catalog"]
)
.unwrap();
/// Count of killed process in each catalog.
pub static ref PROCESS_KILL_COUNT: IntCounterVec = register_int_counter_vec!(
"greptime_process_kill_count",
"Completed kill process requests count",
&["catalog"]
)
.unwrap();
}

View File

@@ -0,0 +1,488 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use std::collections::hash_map::Entry;
use std::collections::HashMap;
use std::fmt::{Debug, Formatter};
use std::sync::atomic::{AtomicU64, Ordering};
use std::sync::{Arc, RwLock};
use api::v1::frontend::{KillProcessRequest, ListProcessRequest, ProcessInfo};
use common_base::cancellation::CancellationHandle;
use common_frontend::selector::{FrontendSelector, MetaClientSelector};
use common_telemetry::{debug, info};
use common_time::util::current_time_millis;
use meta_client::MetaClientRef;
use snafu::{ensure, OptionExt, ResultExt};
use crate::error;
use crate::metrics::{PROCESS_KILL_COUNT, PROCESS_LIST_COUNT};
pub type ProcessManagerRef = Arc<ProcessManager>;
/// Query process manager.
pub struct ProcessManager {
/// Local frontend server address,
server_addr: String,
/// Next process id for local queries.
next_id: AtomicU64,
/// Running process per catalog.
catalogs: RwLock<HashMap<String, HashMap<u64, CancellableProcess>>>,
/// Frontend selector to locate frontend nodes.
frontend_selector: Option<MetaClientSelector>,
}
impl ProcessManager {
/// Create a [ProcessManager] instance with server address and kv client.
pub fn new(server_addr: String, meta_client: Option<MetaClientRef>) -> Self {
let frontend_selector = meta_client.map(MetaClientSelector::new);
Self {
server_addr,
next_id: Default::default(),
catalogs: Default::default(),
frontend_selector,
}
}
}
impl ProcessManager {
/// Registers a submitted query. Use the provided id if present.
#[must_use]
pub fn register_query(
self: &Arc<Self>,
catalog: String,
schemas: Vec<String>,
query: String,
client: String,
id: Option<u64>,
) -> Ticket {
let id = id.unwrap_or_else(|| self.next_id.fetch_add(1, Ordering::Relaxed));
let process = ProcessInfo {
id,
catalog: catalog.clone(),
schemas,
query,
start_timestamp: current_time_millis(),
client,
frontend: self.server_addr.clone(),
};
let cancellation_handle = Arc::new(CancellationHandle::default());
let cancellable_process = CancellableProcess::new(cancellation_handle.clone(), process);
self.catalogs
.write()
.unwrap()
.entry(catalog.clone())
.or_default()
.insert(id, cancellable_process);
Ticket {
catalog,
manager: self.clone(),
id,
cancellation_handle,
}
}
/// Generates the next process id.
pub fn next_id(&self) -> u64 {
self.next_id.fetch_add(1, Ordering::Relaxed)
}
/// De-register a query from process list.
pub fn deregister_query(&self, catalog: String, id: u64) {
if let Entry::Occupied(mut o) = self.catalogs.write().unwrap().entry(catalog) {
let process = o.get_mut().remove(&id);
debug!("Deregister process: {:?}", process);
if o.get().is_empty() {
o.remove();
}
}
}
/// List local running processes in given catalog.
pub fn local_processes(&self, catalog: Option<&str>) -> error::Result<Vec<ProcessInfo>> {
let catalogs = self.catalogs.read().unwrap();
let result = if let Some(catalog) = catalog {
if let Some(catalogs) = catalogs.get(catalog) {
catalogs.values().map(|p| p.process.clone()).collect()
} else {
vec![]
}
} else {
catalogs
.values()
.flat_map(|v| v.values().map(|p| p.process.clone()))
.collect()
};
Ok(result)
}
pub async fn list_all_processes(
&self,
catalog: Option<&str>,
) -> error::Result<Vec<ProcessInfo>> {
let mut processes = vec![];
if let Some(remote_frontend_selector) = self.frontend_selector.as_ref() {
let frontends = remote_frontend_selector
.select(|node| node.peer.addr != self.server_addr)
.await
.context(error::InvokeFrontendSnafu)?;
for mut f in frontends {
processes.extend(
f.list_process(ListProcessRequest {
catalog: catalog.unwrap_or_default().to_string(),
})
.await
.context(error::InvokeFrontendSnafu)?
.processes,
);
}
}
processes.extend(self.local_processes(catalog)?);
Ok(processes)
}
/// Kills query with provided catalog and id.
pub async fn kill_process(
&self,
server_addr: String,
catalog: String,
id: u64,
) -> error::Result<bool> {
if server_addr == self.server_addr {
if let Some(catalogs) = self.catalogs.write().unwrap().get_mut(&catalog) {
if let Some(process) = catalogs.remove(&id) {
process.handle.cancel();
info!(
"Killed process, catalog: {}, id: {:?}",
process.process.catalog, process.process.id
);
PROCESS_KILL_COUNT.with_label_values(&[&catalog]).inc();
Ok(true)
} else {
debug!("Failed to kill process, id not found: {}", id);
Ok(false)
}
} else {
debug!("Failed to kill process, catalog not found: {}", catalog);
Ok(false)
}
} else {
let mut nodes = self
.frontend_selector
.as_ref()
.context(error::MetaClientMissingSnafu)?
.select(|node| node.peer.addr == server_addr)
.await
.context(error::InvokeFrontendSnafu)?;
ensure!(
!nodes.is_empty(),
error::FrontendNotFoundSnafu { addr: server_addr }
);
let request = KillProcessRequest {
server_addr,
catalog,
process_id: id,
};
nodes[0]
.kill_process(request)
.await
.context(error::InvokeFrontendSnafu)?;
Ok(true)
}
}
}
pub struct Ticket {
pub(crate) catalog: String,
pub(crate) manager: ProcessManagerRef,
pub(crate) id: u64,
pub cancellation_handle: Arc<CancellationHandle>,
}
impl Drop for Ticket {
fn drop(&mut self) {
self.manager
.deregister_query(std::mem::take(&mut self.catalog), self.id);
}
}
struct CancellableProcess {
handle: Arc<CancellationHandle>,
process: ProcessInfo,
}
impl Drop for CancellableProcess {
fn drop(&mut self) {
PROCESS_LIST_COUNT
.with_label_values(&[&self.process.catalog])
.dec();
}
}
impl CancellableProcess {
fn new(handle: Arc<CancellationHandle>, process: ProcessInfo) -> Self {
PROCESS_LIST_COUNT
.with_label_values(&[&process.catalog])
.inc();
Self { handle, process }
}
}
impl Debug for CancellableProcess {
fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
f.debug_struct("CancellableProcess")
.field("cancelled", &self.handle.is_cancelled())
.field("process", &self.process)
.finish()
}
}
#[cfg(test)]
mod tests {
use std::sync::Arc;
use crate::process_manager::ProcessManager;
#[tokio::test]
async fn test_register_query() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let ticket = process_manager.clone().register_query(
"public".to_string(),
vec!["test".to_string()],
"SELECT * FROM table".to_string(),
"".to_string(),
None,
);
let running_processes = process_manager.local_processes(None).unwrap();
assert_eq!(running_processes.len(), 1);
assert_eq!(&running_processes[0].frontend, "127.0.0.1:8000");
assert_eq!(running_processes[0].id, ticket.id);
assert_eq!(&running_processes[0].query, "SELECT * FROM table");
drop(ticket);
assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
}
#[tokio::test]
async fn test_register_query_with_custom_id() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let custom_id = 12345;
let ticket = process_manager.clone().register_query(
"public".to_string(),
vec!["test".to_string()],
"SELECT * FROM table".to_string(),
"client1".to_string(),
Some(custom_id),
);
assert_eq!(ticket.id, custom_id);
let running_processes = process_manager.local_processes(None).unwrap();
assert_eq!(running_processes.len(), 1);
assert_eq!(running_processes[0].id, custom_id);
assert_eq!(&running_processes[0].client, "client1");
}
#[tokio::test]
async fn test_multiple_queries_same_catalog() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let ticket1 = process_manager.clone().register_query(
"public".to_string(),
vec!["schema1".to_string()],
"SELECT * FROM table1".to_string(),
"client1".to_string(),
None,
);
let ticket2 = process_manager.clone().register_query(
"public".to_string(),
vec!["schema2".to_string()],
"SELECT * FROM table2".to_string(),
"client2".to_string(),
None,
);
let running_processes = process_manager.local_processes(Some("public")).unwrap();
assert_eq!(running_processes.len(), 2);
// Verify both processes are present
let ids: Vec<u64> = running_processes.iter().map(|p| p.id).collect();
assert!(ids.contains(&ticket1.id));
assert!(ids.contains(&ticket2.id));
}
#[tokio::test]
async fn test_multiple_catalogs() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let _ticket1 = process_manager.clone().register_query(
"catalog1".to_string(),
vec!["schema1".to_string()],
"SELECT * FROM table1".to_string(),
"client1".to_string(),
None,
);
let _ticket2 = process_manager.clone().register_query(
"catalog2".to_string(),
vec!["schema2".to_string()],
"SELECT * FROM table2".to_string(),
"client2".to_string(),
None,
);
// Test listing processes for specific catalog
let catalog1_processes = process_manager.local_processes(Some("catalog1")).unwrap();
assert_eq!(catalog1_processes.len(), 1);
assert_eq!(&catalog1_processes[0].catalog, "catalog1");
let catalog2_processes = process_manager.local_processes(Some("catalog2")).unwrap();
assert_eq!(catalog2_processes.len(), 1);
assert_eq!(&catalog2_processes[0].catalog, "catalog2");
// Test listing all processes
let all_processes = process_manager.local_processes(None).unwrap();
assert_eq!(all_processes.len(), 2);
}
#[tokio::test]
async fn test_deregister_query() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let ticket = process_manager.clone().register_query(
"public".to_string(),
vec!["test".to_string()],
"SELECT * FROM table".to_string(),
"client1".to_string(),
None,
);
assert_eq!(process_manager.local_processes(None).unwrap().len(), 1);
process_manager.deregister_query("public".to_string(), ticket.id);
assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
}
#[tokio::test]
async fn test_cancellation_handle() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let ticket = process_manager.clone().register_query(
"public".to_string(),
vec!["test".to_string()],
"SELECT * FROM table".to_string(),
"client1".to_string(),
None,
);
assert!(!ticket.cancellation_handle.is_cancelled());
ticket.cancellation_handle.cancel();
assert!(ticket.cancellation_handle.is_cancelled());
}
#[tokio::test]
async fn test_kill_local_process() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let ticket = process_manager.clone().register_query(
"public".to_string(),
vec!["test".to_string()],
"SELECT * FROM table".to_string(),
"client1".to_string(),
None,
);
assert!(!ticket.cancellation_handle.is_cancelled());
let killed = process_manager
.kill_process(
"127.0.0.1:8000".to_string(),
"public".to_string(),
ticket.id,
)
.await
.unwrap();
assert!(killed);
assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
}
#[tokio::test]
async fn test_kill_nonexistent_process() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let killed = process_manager
.kill_process("127.0.0.1:8000".to_string(), "public".to_string(), 999)
.await
.unwrap();
assert!(!killed);
}
#[tokio::test]
async fn test_kill_process_nonexistent_catalog() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let killed = process_manager
.kill_process("127.0.0.1:8000".to_string(), "nonexistent".to_string(), 1)
.await
.unwrap();
assert!(!killed);
}
#[tokio::test]
async fn test_process_info_fields() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
let _ticket = process_manager.clone().register_query(
"test_catalog".to_string(),
vec!["schema1".to_string(), "schema2".to_string()],
"SELECT COUNT(*) FROM users WHERE age > 18".to_string(),
"test_client".to_string(),
Some(42),
);
let processes = process_manager.local_processes(None).unwrap();
assert_eq!(processes.len(), 1);
let process = &processes[0];
assert_eq!(process.id, 42);
assert_eq!(&process.catalog, "test_catalog");
assert_eq!(process.schemas, vec!["schema1", "schema2"]);
assert_eq!(&process.query, "SELECT COUNT(*) FROM users WHERE age > 18");
assert_eq!(&process.client, "test_client");
assert_eq!(&process.frontend, "127.0.0.1:8000");
assert!(process.start_timestamp > 0);
}
#[tokio::test]
async fn test_ticket_drop_deregisters_process() {
let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
{
let _ticket = process_manager.clone().register_query(
"public".to_string(),
vec!["test".to_string()],
"SELECT * FROM table".to_string(),
"client1".to_string(),
None,
);
// Process should be registered
assert_eq!(process_manager.local_processes(None).unwrap().len(), 1);
} // ticket goes out of scope here
// Process should be automatically deregistered
assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
}
}

View File

@@ -19,6 +19,7 @@ mod information_memory_table;
pub mod key_column_usage;
mod partitions;
mod procedure_info;
mod process_list;
pub mod region_peers;
mod region_statistics;
mod runtime_metrics;
@@ -42,6 +43,7 @@ use common_recordbatch::SendableRecordBatchStream;
use datatypes::schema::SchemaRef;
use lazy_static::lazy_static;
use paste::paste;
use process_list::InformationSchemaProcessList;
use store_api::storage::{ScanRequest, TableId};
use table::metadata::TableType;
use table::TableRef;
@@ -50,6 +52,7 @@ use views::InformationSchemaViews;
use self::columns::InformationSchemaColumns;
use crate::error::{Error, Result};
use crate::process_manager::ProcessManagerRef;
use crate::system_schema::information_schema::cluster_info::InformationSchemaClusterInfo;
use crate::system_schema::information_schema::flows::InformationSchemaFlows;
use crate::system_schema::information_schema::information_memory_table::get_schema_columns;
@@ -113,6 +116,7 @@ macro_rules! setup_memory_table {
pub struct InformationSchemaProvider {
catalog_name: String,
catalog_manager: Weak<dyn CatalogManager>,
process_manager: Option<ProcessManagerRef>,
flow_metadata_manager: Arc<FlowMetadataManager>,
tables: HashMap<String, TableRef>,
}
@@ -207,6 +211,10 @@ impl SystemSchemaProviderInner for InformationSchemaProvider {
self.catalog_manager.clone(),
),
) as _),
PROCESS_LIST => self
.process_manager
.as_ref()
.map(|p| Arc::new(InformationSchemaProcessList::new(p.clone())) as _),
_ => None,
}
}
@@ -217,11 +225,13 @@ impl InformationSchemaProvider {
catalog_name: String,
catalog_manager: Weak<dyn CatalogManager>,
flow_metadata_manager: Arc<FlowMetadataManager>,
process_manager: Option<ProcessManagerRef>,
) -> Self {
let mut provider = Self {
catalog_name,
catalog_manager,
flow_metadata_manager,
process_manager,
tables: HashMap::new(),
};
@@ -277,6 +287,9 @@ impl InformationSchemaProvider {
self.build_table(TABLE_CONSTRAINTS).unwrap(),
);
tables.insert(FLOWS.to_string(), self.build_table(FLOWS).unwrap());
if let Some(process_list) = self.build_table(PROCESS_LIST) {
tables.insert(PROCESS_LIST.to_string(), process_list);
}
// Add memory tables
for name in MEMORY_TABLES.iter() {
tables.insert((*name).to_string(), self.build_table(name).expect(name));

View File

@@ -0,0 +1,189 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
use common_catalog::consts::INFORMATION_SCHEMA_PROCESS_LIST_TABLE_ID;
use common_error::ext::BoxedError;
use common_frontend::DisplayProcessId;
use common_recordbatch::adapter::RecordBatchStreamAdapter;
use common_recordbatch::{RecordBatch, SendableRecordBatchStream};
use common_time::util::current_time_millis;
use common_time::{Duration, Timestamp};
use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
use datatypes::prelude::ConcreteDataType as CDT;
use datatypes::scalars::ScalarVectorBuilder;
use datatypes::schema::{ColumnSchema, Schema, SchemaRef};
use datatypes::value::Value;
use datatypes::vectors::{
DurationMillisecondVectorBuilder, StringVectorBuilder, TimestampMillisecondVectorBuilder,
VectorRef,
};
use snafu::ResultExt;
use store_api::storage::{ScanRequest, TableId};
use crate::error::{self, InternalSnafu};
use crate::information_schema::Predicates;
use crate::process_manager::ProcessManagerRef;
use crate::system_schema::information_schema::InformationTable;
/// Column names of `information_schema.process_list`
const ID: &str = "id";
const CATALOG: &str = "catalog";
const SCHEMAS: &str = "schemas";
const QUERY: &str = "query";
const CLIENT: &str = "client";
const FRONTEND: &str = "frontend";
const START_TIMESTAMP: &str = "start_timestamp";
const ELAPSED_TIME: &str = "elapsed_time";
/// `information_schema.process_list` table implementation that tracks running
/// queries in current cluster.
pub struct InformationSchemaProcessList {
schema: SchemaRef,
process_manager: ProcessManagerRef,
}
impl InformationSchemaProcessList {
pub fn new(process_manager: ProcessManagerRef) -> Self {
Self {
schema: Self::schema(),
process_manager,
}
}
fn schema() -> SchemaRef {
Arc::new(Schema::new(vec![
ColumnSchema::new(ID, CDT::string_datatype(), false),
ColumnSchema::new(CATALOG, CDT::string_datatype(), false),
ColumnSchema::new(SCHEMAS, CDT::string_datatype(), false),
ColumnSchema::new(QUERY, CDT::string_datatype(), false),
ColumnSchema::new(CLIENT, CDT::string_datatype(), false),
ColumnSchema::new(FRONTEND, CDT::string_datatype(), false),
ColumnSchema::new(
START_TIMESTAMP,
CDT::timestamp_millisecond_datatype(),
false,
),
ColumnSchema::new(ELAPSED_TIME, CDT::duration_millisecond_datatype(), false),
]))
}
}
impl InformationTable for InformationSchemaProcessList {
fn table_id(&self) -> TableId {
INFORMATION_SCHEMA_PROCESS_LIST_TABLE_ID
}
fn table_name(&self) -> &'static str {
"process_list"
}
fn schema(&self) -> SchemaRef {
self.schema.clone()
}
fn to_stream(&self, request: ScanRequest) -> error::Result<SendableRecordBatchStream> {
let process_manager = self.process_manager.clone();
let stream = Box::pin(DfRecordBatchStreamAdapter::new(
self.schema.arrow_schema().clone(),
futures::stream::once(async move {
make_process_list(process_manager, request)
.await
.map(RecordBatch::into_df_record_batch)
.map_err(|e| datafusion::error::DataFusionError::External(Box::new(e)))
}),
));
Ok(Box::pin(
RecordBatchStreamAdapter::try_new(stream)
.map_err(BoxedError::new)
.context(InternalSnafu)?,
))
}
}
/// Build running process list.
async fn make_process_list(
process_manager: ProcessManagerRef,
request: ScanRequest,
) -> error::Result<RecordBatch> {
let predicates = Predicates::from_scan_request(&Some(request));
let current_time = current_time_millis();
// todo(hl): find a way to extract user catalog to filter queries from other users.
let queries = process_manager.list_all_processes(None).await?;
let mut id_builder = StringVectorBuilder::with_capacity(queries.len());
let mut catalog_builder = StringVectorBuilder::with_capacity(queries.len());
let mut schemas_builder = StringVectorBuilder::with_capacity(queries.len());
let mut query_builder = StringVectorBuilder::with_capacity(queries.len());
let mut client_builder = StringVectorBuilder::with_capacity(queries.len());
let mut frontend_builder = StringVectorBuilder::with_capacity(queries.len());
let mut start_time_builder = TimestampMillisecondVectorBuilder::with_capacity(queries.len());
let mut elapsed_time_builder = DurationMillisecondVectorBuilder::with_capacity(queries.len());
for process in queries {
let display_id = DisplayProcessId {
server_addr: process.frontend.to_string(),
id: process.id,
}
.to_string();
let schemas = process.schemas.join(",");
let id = Value::from(display_id);
let catalog = Value::from(process.catalog);
let schemas = Value::from(schemas);
let query = Value::from(process.query);
let client = Value::from(process.client);
let frontend = Value::from(process.frontend);
let start_timestamp = Value::from(Timestamp::new_millisecond(process.start_timestamp));
let elapsed_time = Value::from(Duration::new_millisecond(
current_time - process.start_timestamp,
));
let row = [
(ID, &id),
(CATALOG, &catalog),
(SCHEMAS, &schemas),
(QUERY, &query),
(CLIENT, &client),
(FRONTEND, &frontend),
(START_TIMESTAMP, &start_timestamp),
(ELAPSED_TIME, &elapsed_time),
];
if predicates.eval(&row) {
id_builder.push(id.as_string().as_deref());
catalog_builder.push(catalog.as_string().as_deref());
schemas_builder.push(schemas.as_string().as_deref());
query_builder.push(query.as_string().as_deref());
client_builder.push(client.as_string().as_deref());
frontend_builder.push(frontend.as_string().as_deref());
start_time_builder.push(start_timestamp.as_timestamp().map(|t| t.value().into()));
elapsed_time_builder.push(elapsed_time.as_duration().map(|d| d.value().into()));
}
}
RecordBatch::new(
InformationSchemaProcessList::schema(),
vec![
Arc::new(id_builder.finish()) as VectorRef,
Arc::new(catalog_builder.finish()) as VectorRef,
Arc::new(schemas_builder.finish()) as VectorRef,
Arc::new(query_builder.finish()) as VectorRef,
Arc::new(client_builder.finish()) as VectorRef,
Arc::new(frontend_builder.finish()) as VectorRef,
Arc::new(start_time_builder.finish()) as VectorRef,
Arc::new(elapsed_time_builder.finish()) as VectorRef,
],
)
.context(error::CreateRecordBatchSnafu)
}

View File

@@ -47,3 +47,4 @@ pub const VIEWS: &str = "views";
pub const FLOWS: &str = "flows";
pub const PROCEDURE_INFO: &str = "procedure_info";
pub const REGION_STATISTICS: &str = "region_statistics";
pub const PROCESS_LIST: &str = "process_list";

View File

@@ -328,6 +328,7 @@ mod tests {
backend.clone(),
layered_cache_registry,
None,
None,
);
let table_metadata_manager = TableMetadataManager::new(backend);
let mut view_info = common_meta::key::test_utils::new_test_table_info(1024, vec![]);

View File

@@ -58,6 +58,7 @@ where
info!("{desc}, average operation cost: {cost:.2} ms");
}
/// Command to benchmark table metadata operations.
#[derive(Debug, Default, Parser)]
pub struct BenchTableMetadataCommand {
#[clap(long)]

39
src/cli/src/data.rs Normal file
View File

@@ -0,0 +1,39 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
mod export;
mod import;
use clap::Subcommand;
use common_error::ext::BoxedError;
use crate::data::export::ExportCommand;
use crate::data::import::ImportCommand;
use crate::Tool;
/// Command for data operations including exporting data from and importing data into GreptimeDB.
#[derive(Subcommand)]
pub enum DataCommand {
Export(ExportCommand),
Import(ImportCommand),
}
impl DataCommand {
pub async fn build(&self) -> std::result::Result<Box<dyn Tool>, BoxedError> {
match self {
DataCommand::Export(cmd) => cmd.build().await,
DataCommand::Import(cmd) => cmd.build().await,
}
}
}

View File

@@ -50,6 +50,7 @@ enum ExportTarget {
All,
}
/// Command for exporting data from the GreptimeDB.
#[derive(Debug, Default, Parser)]
pub struct ExportCommand {
/// Server address to connect

View File

@@ -40,6 +40,7 @@ enum ImportTarget {
All,
}
/// Command to import data from a directory into a GreptimeDB instance.
#[derive(Debug, Default, Parser)]
pub struct ImportCommand {
/// Server address to connect

View File

@@ -30,6 +30,7 @@ pub enum Error {
location: Location,
msg: String,
},
#[snafu(display("Failed to create default catalog and schema"))]
InitMetadata {
#[snafu(implicit)]
@@ -228,19 +229,41 @@ pub enum Error {
#[snafu(source)]
error: ObjectStoreError,
},
#[snafu(display("S3 config need be set"))]
S3ConfigNotSet {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Output directory not set"))]
OutputDirNotSet {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("KV backend not set: {}", backend))]
KvBackendNotSet {
backend: String,
#[snafu(display("Empty store addresses"))]
EmptyStoreAddrs {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Unsupported memory backend"))]
UnsupportedMemoryBackend {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("File path invalid: {}", msg))]
InvalidFilePath {
msg: String,
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Invalid arguments: {}", msg))]
InvalidArguments {
msg: String,
#[snafu(implicit)]
location: Location,
},
@@ -262,6 +285,9 @@ impl ErrorExt for Error {
| Error::ConnectEtcd { .. }
| Error::CreateDir { .. }
| Error::EmptyResult { .. }
| Error::InvalidFilePath { .. }
| Error::UnsupportedMemoryBackend { .. }
| Error::InvalidArguments { .. }
| Error::ParseProxyOpts { .. } => StatusCode::InvalidArguments,
Error::StartProcedureManager { source, .. }
@@ -282,7 +308,7 @@ impl ErrorExt for Error {
Error::OpenDal { .. } => StatusCode::Internal,
Error::S3ConfigNotSet { .. }
| Error::OutputDirNotSet { .. }
| Error::KvBackendNotSet { .. } => StatusCode::InvalidArguments,
| Error::EmptyStoreAddrs { .. } => StatusCode::InvalidArguments,
Error::BuildRuntime { source, .. } => source.status_code(),

View File

@@ -13,11 +13,10 @@
// limitations under the License.
mod bench;
mod data;
mod database;
pub mod error;
mod export;
mod import;
mod meta_snapshot;
mod metadata;
use async_trait::async_trait;
use clap::Parser;
@@ -26,9 +25,8 @@ pub use database::DatabaseClient;
use error::Result;
pub use crate::bench::BenchTableMetadataCommand;
pub use crate::export::ExportCommand;
pub use crate::import::ImportCommand;
pub use crate::meta_snapshot::{MetaRestoreCommand, MetaSnapshotCommand};
pub use crate::data::DataCommand;
pub use crate::metadata::MetadataCommand;
#[async_trait]
pub trait Tool: Send + Sync {

42
src/cli/src/metadata.rs Normal file
View File

@@ -0,0 +1,42 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
mod common;
mod control;
mod snapshot;
use clap::Subcommand;
use common_error::ext::BoxedError;
use crate::metadata::control::ControlCommand;
use crate::metadata::snapshot::SnapshotCommand;
use crate::Tool;
/// Command for managing metadata operations, including saving metadata snapshots and restoring metadata from snapshots.
#[derive(Subcommand)]
pub enum MetadataCommand {
#[clap(subcommand)]
Snapshot(SnapshotCommand),
#[clap(subcommand)]
Control(ControlCommand),
}
impl MetadataCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
match self {
MetadataCommand::Snapshot(cmd) => cmd.build().await,
MetadataCommand::Control(cmd) => cmd.build().await,
}
}
}

View File

@@ -0,0 +1,116 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
use clap::Parser;
use common_error::ext::BoxedError;
use common_meta::kv_backend::chroot::ChrootKvBackend;
use common_meta::kv_backend::etcd::EtcdStore;
use common_meta::kv_backend::KvBackendRef;
use meta_srv::bootstrap::create_etcd_client;
use meta_srv::metasrv::BackendImpl;
use crate::error::{EmptyStoreAddrsSnafu, UnsupportedMemoryBackendSnafu};
#[derive(Debug, Default, Parser)]
pub(crate) struct StoreConfig {
/// The endpoint of store. one of etcd, postgres or mysql.
///
/// For postgres store, the format is:
/// "password=password dbname=postgres user=postgres host=localhost port=5432"
///
/// For etcd store, the format is:
/// "127.0.0.1:2379"
///
/// For mysql store, the format is:
/// "mysql://user:password@ip:port/dbname"
#[clap(long, alias = "store-addr", value_delimiter = ',', num_args = 1..)]
store_addrs: Vec<String>,
/// The maximum number of operations in a transaction. Only used when using [etcd-store].
#[clap(long, default_value = "128")]
max_txn_ops: usize,
/// The metadata store backend.
#[clap(long, value_enum, default_value = "etcd-store")]
backend: BackendImpl,
/// The key prefix of the metadata store.
#[clap(long, default_value = "")]
store_key_prefix: String,
/// The table name in RDS to store metadata. Only used when using [postgres-store] or [mysql-store].
#[cfg(any(feature = "pg_kvbackend", feature = "mysql_kvbackend"))]
#[clap(long, default_value = common_meta::kv_backend::DEFAULT_META_TABLE_NAME)]
meta_table_name: String,
}
impl StoreConfig {
/// Builds a [`KvBackendRef`] from the store configuration.
pub async fn build(&self) -> Result<KvBackendRef, BoxedError> {
let max_txn_ops = self.max_txn_ops;
let store_addrs = &self.store_addrs;
if store_addrs.is_empty() {
EmptyStoreAddrsSnafu.fail().map_err(BoxedError::new)
} else {
let kvbackend = match self.backend {
BackendImpl::EtcdStore => {
let etcd_client = create_etcd_client(store_addrs)
.await
.map_err(BoxedError::new)?;
Ok(EtcdStore::with_etcd_client(etcd_client, max_txn_ops))
}
#[cfg(feature = "pg_kvbackend")]
BackendImpl::PostgresStore => {
let table_name = &self.meta_table_name;
let pool = meta_srv::bootstrap::create_postgres_pool(store_addrs)
.await
.map_err(BoxedError::new)?;
Ok(common_meta::kv_backend::rds::PgStore::with_pg_pool(
pool,
table_name,
max_txn_ops,
)
.await
.map_err(BoxedError::new)?)
}
#[cfg(feature = "mysql_kvbackend")]
BackendImpl::MysqlStore => {
let table_name = &self.meta_table_name;
let pool = meta_srv::bootstrap::create_mysql_pool(store_addrs)
.await
.map_err(BoxedError::new)?;
Ok(common_meta::kv_backend::rds::MySqlStore::with_mysql_pool(
pool,
table_name,
max_txn_ops,
)
.await
.map_err(BoxedError::new)?)
}
BackendImpl::MemoryStore => UnsupportedMemoryBackendSnafu
.fail()
.map_err(BoxedError::new),
};
if self.store_key_prefix.is_empty() {
kvbackend
} else {
let chroot_kvbackend =
ChrootKvBackend::new(self.store_key_prefix.as_bytes().to_vec(), kvbackend?);
Ok(Arc::new(chroot_kvbackend))
}
}
}
}

View File

@@ -0,0 +1,38 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
mod get;
mod utils;
use clap::Subcommand;
use common_error::ext::BoxedError;
use get::GetCommand;
use crate::Tool;
/// Subcommand for metadata control.
#[derive(Subcommand)]
pub enum ControlCommand {
/// Get the metadata from the metasrv.
#[clap(subcommand)]
Get(GetCommand),
}
impl ControlCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
match self {
ControlCommand::Get(cmd) => cmd.build().await,
}
}
}

View File

@@ -0,0 +1,242 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use std::cmp::min;
use async_trait::async_trait;
use clap::{Parser, Subcommand};
use client::{DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME};
use common_catalog::format_full_table_name;
use common_error::ext::BoxedError;
use common_meta::key::table_info::TableInfoKey;
use common_meta::key::table_name::TableNameKey;
use common_meta::key::table_route::TableRouteKey;
use common_meta::key::TableMetadataManager;
use common_meta::kv_backend::KvBackendRef;
use common_meta::range_stream::{PaginationStream, DEFAULT_PAGE_SIZE};
use common_meta::rpc::store::RangeRequest;
use futures::TryStreamExt;
use crate::error::InvalidArgumentsSnafu;
use crate::metadata::common::StoreConfig;
use crate::metadata::control::utils::{decode_key_value, json_fromatter};
use crate::Tool;
/// Subcommand for get command.
#[derive(Subcommand)]
pub enum GetCommand {
Key(GetKeyCommand),
Table(GetTableCommand),
}
impl GetCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
match self {
GetCommand::Key(cmd) => cmd.build().await,
GetCommand::Table(cmd) => cmd.build().await,
}
}
}
/// Get key-value pairs from the metadata store.
#[derive(Debug, Default, Parser)]
pub struct GetKeyCommand {
/// The key to get from the metadata store. If empty, returns all key-value pairs.
#[clap(default_value = "")]
key: String,
/// Whether to perform a prefix query. If true, returns all key-value pairs where the key starts with the given prefix.
#[clap(long, default_value = "false")]
prefix: bool,
/// The maximum number of key-value pairs to return. If 0, returns all key-value pairs.
#[clap(long, default_value = "0")]
limit: u64,
#[clap(flatten)]
store: StoreConfig,
}
impl GetKeyCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
let kvbackend = self.store.build().await?;
Ok(Box::new(GetKeyTool {
kvbackend,
key: self.key.clone(),
prefix: self.prefix,
limit: self.limit,
}))
}
}
struct GetKeyTool {
kvbackend: KvBackendRef,
key: String,
prefix: bool,
limit: u64,
}
#[async_trait]
impl Tool for GetKeyTool {
async fn do_work(&self) -> Result<(), BoxedError> {
let mut req = RangeRequest::default();
if self.prefix {
req = req.with_prefix(self.key.as_bytes());
} else {
req = req.with_key(self.key.as_bytes());
}
let page_size = if self.limit > 0 {
min(self.limit as usize, DEFAULT_PAGE_SIZE)
} else {
DEFAULT_PAGE_SIZE
};
let pagination_stream =
PaginationStream::new(self.kvbackend.clone(), req, page_size, decode_key_value);
let mut stream = Box::pin(pagination_stream.into_stream());
let mut counter = 0;
while let Some((key, value)) = stream.try_next().await.map_err(BoxedError::new)? {
print!("{}\n{}\n", key, value);
counter += 1;
if self.limit > 0 && counter >= self.limit {
break;
}
}
Ok(())
}
}
/// Get table metadata from the metadata store via table id.
#[derive(Debug, Default, Parser)]
pub struct GetTableCommand {
/// Get table metadata by table id.
#[clap(long)]
table_id: Option<u32>,
/// Get table metadata by table name.
#[clap(long)]
table_name: Option<String>,
/// The schema name of the table.
#[clap(long)]
schema_name: Option<String>,
/// Pretty print the output.
#[clap(long, default_value = "false")]
pretty: bool,
#[clap(flatten)]
store: StoreConfig,
}
impl GetTableCommand {
pub fn validate(&self) -> Result<(), BoxedError> {
if self.table_id.is_none() && self.table_name.is_none() {
return Err(BoxedError::new(
InvalidArgumentsSnafu {
msg: "You must specify either --table-id or --table-name.",
}
.build(),
));
}
Ok(())
}
}
struct GetTableTool {
kvbackend: KvBackendRef,
table_id: Option<u32>,
table_name: Option<String>,
schema_name: Option<String>,
pretty: bool,
}
#[async_trait]
impl Tool for GetTableTool {
async fn do_work(&self) -> Result<(), BoxedError> {
let table_metadata_manager = TableMetadataManager::new(self.kvbackend.clone());
let table_name_manager = table_metadata_manager.table_name_manager();
let table_info_manager = table_metadata_manager.table_info_manager();
let table_route_manager = table_metadata_manager.table_route_manager();
let table_id = if let Some(table_name) = &self.table_name {
let catalog = DEFAULT_CATALOG_NAME.to_string();
let schema_name = self
.schema_name
.clone()
.unwrap_or_else(|| DEFAULT_SCHEMA_NAME.to_string());
let key = TableNameKey::new(&catalog, &schema_name, table_name);
let Some(table_name) = table_name_manager.get(key).await.map_err(BoxedError::new)?
else {
println!(
"Table({}) not found",
format_full_table_name(&catalog, &schema_name, table_name)
);
return Ok(());
};
table_name.table_id()
} else {
// Safety: we have validated that table_id or table_name is not None
self.table_id.unwrap()
};
let table_info = table_info_manager
.get(table_id)
.await
.map_err(BoxedError::new)?;
if let Some(table_info) = table_info {
println!(
"{}\n{}",
TableInfoKey::new(table_id),
json_fromatter(self.pretty, &*table_info)
);
} else {
println!("Table info not found");
}
let table_route = table_route_manager
.table_route_storage()
.get(table_id)
.await
.map_err(BoxedError::new)?;
if let Some(table_route) = table_route {
println!(
"{}\n{}",
TableRouteKey::new(table_id),
json_fromatter(self.pretty, &table_route)
);
} else {
println!("Table route not found");
}
Ok(())
}
}
impl GetTableCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
self.validate()?;
let kvbackend = self.store.build().await?;
Ok(Box::new(GetTableTool {
kvbackend,
table_id: self.table_id,
table_name: self.table_name.clone(),
schema_name: self.schema_name.clone(),
pretty: self.pretty,
}))
}
}

View File

@@ -0,0 +1,36 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use common_meta::error::Result as CommonMetaResult;
use common_meta::rpc::KeyValue;
use serde::Serialize;
/// Decodes a key-value pair into a string.
pub fn decode_key_value(kv: KeyValue) -> CommonMetaResult<(String, String)> {
let key = String::from_utf8_lossy(&kv.key).to_string();
let value = String::from_utf8_lossy(&kv.value).to_string();
Ok((key, value))
}
/// Formats a value as a JSON string.
pub fn json_fromatter<T>(pretty: bool, value: &T) -> String
where
T: Serialize,
{
if pretty {
serde_json::to_string_pretty(value).unwrap()
} else {
serde_json::to_string(value).unwrap()
}
}

View File

@@ -12,96 +12,38 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
use std::path::Path;
use async_trait::async_trait;
use clap::Parser;
use clap::{Parser, Subcommand};
use common_base::secrets::{ExposeSecret, SecretString};
use common_error::ext::BoxedError;
use common_meta::kv_backend::chroot::ChrootKvBackend;
use common_meta::kv_backend::etcd::EtcdStore;
use common_meta::kv_backend::KvBackendRef;
use common_meta::snapshot::MetadataSnapshotManager;
use meta_srv::bootstrap::create_etcd_client;
use meta_srv::metasrv::BackendImpl;
use object_store::services::{Fs, S3};
use object_store::ObjectStore;
use snafu::ResultExt;
use snafu::{OptionExt, ResultExt};
use crate::error::{KvBackendNotSetSnafu, OpenDalSnafu, S3ConfigNotSetSnafu};
use crate::error::{InvalidFilePathSnafu, OpenDalSnafu, S3ConfigNotSetSnafu};
use crate::metadata::common::StoreConfig;
use crate::Tool;
#[derive(Debug, Default, Parser)]
struct MetaConnection {
/// The endpoint of store. one of etcd, pg or mysql.
#[clap(long, alias = "store-addr", value_delimiter = ',', num_args = 1..)]
store_addrs: Vec<String>,
/// The database backend.
#[clap(long, value_enum)]
backend: Option<BackendImpl>,
#[clap(long, default_value = "")]
store_key_prefix: String,
#[cfg(any(feature = "pg_kvbackend", feature = "mysql_kvbackend"))]
#[clap(long,default_value = common_meta::kv_backend::DEFAULT_META_TABLE_NAME)]
meta_table_name: String,
#[clap(long, default_value = "128")]
max_txn_ops: usize,
/// Subcommand for metadata snapshot operations, including saving snapshots, restoring from snapshots, and viewing snapshot information.
#[derive(Subcommand)]
pub enum SnapshotCommand {
/// Save a snapshot of the current metadata state to a specified location.
Save(SaveCommand),
/// Restore metadata from a snapshot.
Restore(RestoreCommand),
/// Explore metadata from a snapshot.
Info(InfoCommand),
}
impl MetaConnection {
pub async fn build(&self) -> Result<KvBackendRef, BoxedError> {
let max_txn_ops = self.max_txn_ops;
let store_addrs = &self.store_addrs;
if store_addrs.is_empty() {
KvBackendNotSetSnafu { backend: "all" }
.fail()
.map_err(BoxedError::new)
} else {
let kvbackend = match self.backend {
Some(BackendImpl::EtcdStore) => {
let etcd_client = create_etcd_client(store_addrs)
.await
.map_err(BoxedError::new)?;
Ok(EtcdStore::with_etcd_client(etcd_client, max_txn_ops))
}
#[cfg(feature = "pg_kvbackend")]
Some(BackendImpl::PostgresStore) => {
let table_name = &self.meta_table_name;
let pool = meta_srv::bootstrap::create_postgres_pool(store_addrs)
.await
.map_err(BoxedError::new)?;
Ok(common_meta::kv_backend::rds::PgStore::with_pg_pool(
pool,
table_name,
max_txn_ops,
)
.await
.map_err(BoxedError::new)?)
}
#[cfg(feature = "mysql_kvbackend")]
Some(BackendImpl::MysqlStore) => {
let table_name = &self.meta_table_name;
let pool = meta_srv::bootstrap::create_mysql_pool(store_addrs)
.await
.map_err(BoxedError::new)?;
Ok(common_meta::kv_backend::rds::MySqlStore::with_mysql_pool(
pool,
table_name,
max_txn_ops,
)
.await
.map_err(BoxedError::new)?)
}
_ => KvBackendNotSetSnafu { backend: "all" }
.fail()
.map_err(BoxedError::new),
};
if self.store_key_prefix.is_empty() {
kvbackend
} else {
let chroot_kvbackend =
ChrootKvBackend::new(self.store_key_prefix.as_bytes().to_vec(), kvbackend?);
Ok(Arc::new(chroot_kvbackend))
}
impl SnapshotCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
match self {
SnapshotCommand::Save(cmd) => cmd.build().await,
SnapshotCommand::Restore(cmd) => cmd.build().await,
SnapshotCommand::Info(cmd) => cmd.build().await,
}
}
}
@@ -170,10 +112,10 @@ impl S3Config {
/// It will dump the metadata snapshot to local file or s3 bucket.
/// The snapshot file will be in binary format.
#[derive(Debug, Default, Parser)]
pub struct MetaSnapshotCommand {
/// The connection to the metadata store.
pub struct SaveCommand {
/// The store configuration.
#[clap(flatten)]
connection: MetaConnection,
store: StoreConfig,
/// The s3 config.
#[clap(flatten)]
s3_config: S3Config,
@@ -196,9 +138,9 @@ fn create_local_file_object_store(root: &str) -> Result<ObjectStore, BoxedError>
Ok(object_store)
}
impl MetaSnapshotCommand {
impl SaveCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
let kvbackend = self.connection.build().await?;
let kvbackend = self.store.build().await?;
let output_dir = &self.output_dir;
let object_store = self.s3_config.build(output_dir).map_err(BoxedError::new)?;
if let Some(store) = object_store {
@@ -218,7 +160,7 @@ impl MetaSnapshotCommand {
}
}
pub struct MetaSnapshotTool {
struct MetaSnapshotTool {
inner: MetadataSnapshotManager,
target_file: String,
}
@@ -234,14 +176,16 @@ impl Tool for MetaSnapshotTool {
}
}
/// Restore metadata snapshot tool.
/// This tool is used to restore metadata snapshot from etcd, pg or mysql.
/// It will restore the metadata snapshot from local file or s3 bucket.
/// Restore metadata from a snapshot file.
///
/// This command restores the metadata state from a previously saved snapshot.
/// The snapshot can be loaded from either a local file system or an S3 bucket,
/// depending on the provided configuration.
#[derive(Debug, Default, Parser)]
pub struct MetaRestoreCommand {
/// The connection to the metadata store.
pub struct RestoreCommand {
/// The store configuration.
#[clap(flatten)]
connection: MetaConnection,
store: StoreConfig,
/// The s3 config.
#[clap(flatten)]
s3_config: S3Config,
@@ -255,9 +199,9 @@ pub struct MetaRestoreCommand {
force: bool,
}
impl MetaRestoreCommand {
impl RestoreCommand {
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
let kvbackend = self.connection.build().await?;
let kvbackend = self.store.build().await?;
let input_dir = &self.input_dir;
let object_store = self.s3_config.build(input_dir).map_err(BoxedError::new)?;
if let Some(store) = object_store {
@@ -279,7 +223,7 @@ impl MetaRestoreCommand {
}
}
pub struct MetaRestoreTool {
struct MetaRestoreTool {
inner: MetadataSnapshotManager,
source_file: String,
force: bool,
@@ -327,3 +271,93 @@ impl Tool for MetaRestoreTool {
}
}
}
/// Explore metadata from a snapshot file.
///
/// This command allows filtering the metadata by a specific key and limiting the number of results.
/// It prints the filtered metadata to the console.
#[derive(Debug, Default, Parser)]
pub struct InfoCommand {
/// The s3 config.
#[clap(flatten)]
s3_config: S3Config,
/// The name of the target snapshot file. we will add the file extension automatically.
#[clap(long, default_value = "metadata_snapshot")]
file_name: String,
/// The query string to filter the metadata.
#[clap(long, default_value = "*")]
inspect_key: String,
/// The limit of the metadata to query.
#[clap(long)]
limit: Option<usize>,
}
struct MetaInfoTool {
inner: ObjectStore,
source_file: String,
inspect_key: String,
limit: Option<usize>,
}
#[async_trait]
impl Tool for MetaInfoTool {
#[allow(clippy::print_stdout)]
async fn do_work(&self) -> std::result::Result<(), BoxedError> {
let result = MetadataSnapshotManager::info(
&self.inner,
&self.source_file,
&self.inspect_key,
self.limit,
)
.await
.map_err(BoxedError::new)?;
for item in result {
println!("{}", item);
}
Ok(())
}
}
impl InfoCommand {
fn decide_object_store_root_for_local_store(
file_path: &str,
) -> Result<(&str, &str), BoxedError> {
let path = Path::new(file_path);
let parent = path
.parent()
.and_then(|p| p.to_str())
.context(InvalidFilePathSnafu { msg: file_path })
.map_err(BoxedError::new)?;
let file_name = path
.file_name()
.and_then(|f| f.to_str())
.context(InvalidFilePathSnafu { msg: file_path })
.map_err(BoxedError::new)?;
let root = if parent.is_empty() { "." } else { parent };
Ok((root, file_name))
}
pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
let object_store = self.s3_config.build("").map_err(BoxedError::new)?;
if let Some(store) = object_store {
let tool = MetaInfoTool {
inner: store,
source_file: self.file_name.clone(),
inspect_key: self.inspect_key.clone(),
limit: self.limit,
};
Ok(Box::new(tool))
} else {
let (root, file_name) =
Self::decide_object_store_root_for_local_store(&self.file_name)?;
let object_store = create_local_file_object_store(root)?;
let tool = MetaInfoTool {
inner: object_store,
source_file: file_name.to_string(),
inspect_key: self.inspect_key.clone(),
limit: self.limit,
};
Ok(Box::new(tool))
}
}
}

View File

@@ -162,12 +162,23 @@ impl Client {
.as_bytes() as usize
}
pub fn make_flight_client(&self) -> Result<FlightClient> {
pub fn make_flight_client(
&self,
send_compression: bool,
accept_compression: bool,
) -> Result<FlightClient> {
let (addr, channel) = self.find_channel()?;
let client = FlightServiceClient::new(channel)
let mut client = FlightServiceClient::new(channel)
.max_decoding_message_size(self.max_grpc_recv_message_size())
.max_encoding_message_size(self.max_grpc_send_message_size());
// todo(hl): support compression methods.
if send_compression {
client = client.send_compressed(CompressionEncoding::Zstd);
}
if accept_compression {
client = client.accept_compressed(CompressionEncoding::Zstd);
}
Ok(FlightClient { addr, client })
}

View File

@@ -49,7 +49,16 @@ impl NodeManager for NodeClients {
async fn datanode(&self, datanode: &Peer) -> DatanodeRef {
let client = self.get_client(datanode).await;
Arc::new(RegionRequester::new(client))
let ChannelConfig {
send_compression,
accept_compression,
..
} = self.channel_manager.config();
Arc::new(RegionRequester::new(
client,
*send_compression,
*accept_compression,
))
}
async fn flownode(&self, flownode: &Peer) -> FlownodeRef {

View File

@@ -287,7 +287,7 @@ impl Database {
let mut request = tonic::Request::new(request);
Self::put_hints(request.metadata_mut(), hints)?;
let mut client = self.client.make_flight_client()?;
let mut client = self.client.make_flight_client(false, false)?;
let response = client.mut_inner().do_get(request).await.or_else(|e| {
let tonic_code = e.code();
@@ -409,7 +409,7 @@ impl Database {
MetadataValue::from_str(db_to_put).context(InvalidTonicMetadataValueSnafu)?,
);
let mut client = self.client.make_flight_client()?;
let mut client = self.client.make_flight_client(false, false)?;
let response = client.mut_inner().do_put(request).await?;
let response = response
.into_inner()

View File

@@ -12,7 +12,7 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use api::v1::flow::{FlowRequest, FlowResponse};
use api::v1::flow::{DirtyWindowRequest, DirtyWindowRequests, FlowRequest, FlowResponse};
use api::v1::region::InsertRequests;
use common_error::ext::BoxedError;
use common_meta::node_manager::Flownode;
@@ -44,6 +44,16 @@ impl Flownode for FlowRequester {
.map_err(BoxedError::new)
.context(common_meta::error::ExternalSnafu)
}
async fn handle_mark_window_dirty(
&self,
req: DirtyWindowRequest,
) -> common_meta::error::Result<FlowResponse> {
self.handle_mark_window_dirty(req)
.await
.map_err(BoxedError::new)
.context(common_meta::error::ExternalSnafu)
}
}
impl FlowRequester {
@@ -91,4 +101,20 @@ impl FlowRequester {
.into_inner();
Ok(response)
}
async fn handle_mark_window_dirty(&self, req: DirtyWindowRequest) -> Result<FlowResponse> {
let (addr, mut client) = self.client.raw_flow_client()?;
let response = client
.handle_mark_dirty_time_window(DirtyWindowRequests {
requests: vec![req],
})
.await
.or_else(|e| {
let code = e.code();
let err: crate::error::Error = e.into();
Err(BoxedError::new(err)).context(FlowServerSnafu { addr, code })
})?
.into_inner();
Ok(response)
}
}

View File

@@ -46,6 +46,8 @@ use crate::{metrics, Client, Error};
#[derive(Debug)]
pub struct RegionRequester {
client: Client,
send_compression: bool,
accept_compression: bool,
}
#[async_trait]
@@ -89,12 +91,18 @@ impl Datanode for RegionRequester {
}
impl RegionRequester {
pub fn new(client: Client) -> Self {
Self { client }
pub fn new(client: Client, send_compression: bool, accept_compression: bool) -> Self {
Self {
client,
send_compression,
accept_compression,
}
}
pub async fn do_get_inner(&self, ticket: Ticket) -> Result<SendableRecordBatchStream> {
let mut flight_client = self.client.make_flight_client()?;
let mut flight_client = self
.client
.make_flight_client(self.send_compression, self.accept_compression)?;
let response = flight_client
.mut_inner()
.do_get(ticket)

View File

@@ -146,6 +146,7 @@ mod tests {
let output_dir = tempfile::tempdir().unwrap();
let cli = cli::Command::parse_from([
"cli",
"data",
"export",
"--addr",
"127.0.0.1:4000",

View File

@@ -323,6 +323,7 @@ impl StartCommand {
cached_meta_backend.clone(),
layered_cache_registry.clone(),
None,
None,
);
let table_metadata_manager =

View File

@@ -20,6 +20,7 @@ use async_trait::async_trait;
use cache::{build_fundamental_cache_registry, with_default_composite_cache_registry};
use catalog::information_extension::DistributedInformationExtension;
use catalog::kvbackend::{CachedKvBackendBuilder, KvBackendCatalogManager, MetaKvBackend};
use catalog::process_manager::ProcessManager;
use clap::Parser;
use client::client_manager::NodeClients;
use common_base::Plugins;
@@ -38,6 +39,7 @@ use frontend::heartbeat::HeartbeatTask;
use frontend::instance::builder::FrontendBuilder;
use frontend::server::Services;
use meta_client::{MetaClientOptions, MetaClientType};
use servers::addrs;
use servers::export_metrics::ExportMetricsTask;
use servers::tls::{TlsMode, TlsOption};
use snafu::{OptionExt, ResultExt};
@@ -342,11 +344,17 @@ impl StartCommand {
let information_extension =
Arc::new(DistributedInformationExtension::new(meta_client.clone()));
let process_manager = Arc::new(ProcessManager::new(
addrs::resolve_addr(&opts.grpc.bind_addr, Some(&opts.grpc.server_addr)),
Some(meta_client.clone()),
));
let catalog_manager = KvBackendCatalogManager::new(
information_extension,
cached_meta_backend.clone(),
layered_cache_registry.clone(),
None,
Some(process_manager.clone()),
);
let executor = HandlerGroupExecutor::new(vec![
@@ -364,12 +372,16 @@ impl StartCommand {
// frontend to datanode need not timeout.
// Some queries are expected to take long time.
let channel_config = ChannelConfig {
let mut channel_config = ChannelConfig {
timeout: None,
tcp_nodelay: opts.datanode.client.tcp_nodelay,
connect_timeout: Some(opts.datanode.client.connect_timeout),
..Default::default()
};
if opts.grpc.flight_compression.transport_compression() {
channel_config.accept_compression = true;
channel_config.send_compression = true;
}
let client = NodeClients::new(channel_config);
let instance = FrontendBuilder::new(
@@ -379,6 +391,7 @@ impl StartCommand {
catalog_manager,
Arc::new(client),
meta_client,
process_manager,
)
.with_plugin(plugins.clone())
.with_local_cache_invalidator(layered_cache_registry)

View File

@@ -237,12 +237,20 @@ impl StartCommand {
tokio_console_addr: global_options.tokio_console_addr.clone(),
};
#[allow(deprecated)]
if let Some(addr) = &self.rpc_bind_addr {
opts.bind_addr.clone_from(addr);
opts.grpc.bind_addr.clone_from(addr);
} else if !opts.bind_addr.is_empty() {
opts.grpc.bind_addr.clone_from(&opts.bind_addr);
}
#[allow(deprecated)]
if let Some(addr) = &self.rpc_server_addr {
opts.server_addr.clone_from(addr);
opts.grpc.server_addr.clone_from(addr);
} else if !opts.server_addr.is_empty() {
opts.grpc.server_addr.clone_from(&opts.server_addr);
}
if let Some(addrs) = &self.store_addrs {
@@ -319,7 +327,7 @@ impl StartCommand {
let plugin_opts = opts.plugins;
let mut opts = opts.component;
opts.detect_server_addr();
opts.grpc.detect_server_addr();
info!("Metasrv options: {:#?}", opts);
@@ -363,7 +371,7 @@ mod tests {
};
let options = cmd.load_options(&Default::default()).unwrap().component;
assert_eq!("127.0.0.1:3002".to_string(), options.bind_addr);
assert_eq!("127.0.0.1:3002".to_string(), options.grpc.bind_addr);
assert_eq!(vec!["127.0.0.1:2380".to_string()], options.store_addrs);
assert_eq!(SelectorType::LoadBased, options.selector);
}
@@ -396,8 +404,8 @@ mod tests {
};
let options = cmd.load_options(&Default::default()).unwrap().component;
assert_eq!("127.0.0.1:3002".to_string(), options.bind_addr);
assert_eq!("127.0.0.1:3002".to_string(), options.server_addr);
assert_eq!("127.0.0.1:3002".to_string(), options.grpc.bind_addr);
assert_eq!("127.0.0.1:3002".to_string(), options.grpc.server_addr);
assert_eq!(vec!["127.0.0.1:2379".to_string()], options.store_addrs);
assert_eq!(SelectorType::LeaseBased, options.selector);
assert_eq!("debug", options.logging.level.as_ref().unwrap());
@@ -509,10 +517,10 @@ mod tests {
let opts = command.load_options(&Default::default()).unwrap().component;
// Should be read from env, env > default values.
assert_eq!(opts.bind_addr, "127.0.0.1:14002");
assert_eq!(opts.grpc.bind_addr, "127.0.0.1:14002");
// Should be read from config file, config file > env > default values.
assert_eq!(opts.server_addr, "127.0.0.1:3002");
assert_eq!(opts.grpc.server_addr, "127.0.0.1:3002");
// Should be read from cli, cli > config file > env > default values.
assert_eq!(opts.http.addr, "127.0.0.1:14000");

View File

@@ -21,6 +21,7 @@ use async_trait::async_trait;
use cache::{build_fundamental_cache_registry, with_default_composite_cache_registry};
use catalog::information_schema::InformationExtension;
use catalog::kvbackend::KvBackendCatalogManager;
use catalog::process_manager::ProcessManager;
use clap::Parser;
use client::api::v1::meta::RegionRole;
use common_base::readable_size::ReadableSize;
@@ -526,11 +527,14 @@ impl StartCommand {
datanode.region_server(),
procedure_manager.clone(),
));
let process_manager = Arc::new(ProcessManager::new(opts.grpc.server_addr.clone(), None));
let catalog_manager = KvBackendCatalogManager::new(
information_extension.clone(),
kv_backend.clone(),
layered_cache_registry.clone(),
Some(procedure_manager.clone()),
Some(process_manager.clone()),
);
let table_metadata_manager =
@@ -620,6 +624,7 @@ impl StartCommand {
catalog_manager.clone(),
node_manager.clone(),
ddl_task_executor.clone(),
process_manager,
)
.with_plugin(plugins.clone())
.try_build()
@@ -647,7 +652,7 @@ impl StartCommand {
node_manager,
)
.await
.context(error::StartFlownodeSnafu)?;
.context(StartFlownodeSnafu)?;
flow_streaming_engine.set_frontend_invoker(invoker).await;
let export_metrics_task = ExportMetricsTask::try_new(&opts.export_metrics, Some(&plugins))

View File

@@ -12,7 +12,6 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::path::Path;
use std::time::Duration;
use cmd::options::GreptimeOptions;
@@ -58,12 +57,7 @@ fn test_load_datanode_example_config() {
metadata_cache_tti: Duration::from_secs(300),
}),
wal: DatanodeWalConfig::RaftEngine(RaftEngineConfig {
dir: Some(
Path::new(DEFAULT_DATA_HOME)
.join(WAL_DIR)
.to_string_lossy()
.to_string(),
),
dir: Some(format!("{}/{}", DEFAULT_DATA_HOME, WAL_DIR)),
sync_period: Some(Duration::from_secs(10)),
recovery_parallelism: 2,
..Default::default()
@@ -86,16 +80,13 @@ fn test_load_datanode_example_config() {
],
logging: LoggingOptions {
level: Some("info".to_string()),
dir: Path::new(DEFAULT_DATA_HOME)
.join(DEFAULT_LOGGING_DIR)
.to_string_lossy()
.to_string(),
dir: format!("{}/{}", DEFAULT_DATA_HOME, DEFAULT_LOGGING_DIR),
otlp_endpoint: Some(DEFAULT_OTLP_ENDPOINT.to_string()),
tracing_sample_ratio: Some(Default::default()),
..Default::default()
},
export_metrics: ExportMetricsOption {
self_import: Some(Default::default()),
self_import: None,
remote_write: Some(Default::default()),
..Default::default()
},
@@ -132,10 +123,7 @@ fn test_load_frontend_example_config() {
}),
logging: LoggingOptions {
level: Some("info".to_string()),
dir: Path::new(DEFAULT_DATA_HOME)
.join(DEFAULT_LOGGING_DIR)
.to_string_lossy()
.to_string(),
dir: format!("{}/{}", DEFAULT_DATA_HOME, DEFAULT_LOGGING_DIR),
otlp_endpoint: Some(DEFAULT_OTLP_ENDPOINT.to_string()),
tracing_sample_ratio: Some(Default::default()),
..Default::default()
@@ -148,7 +136,7 @@ fn test_load_frontend_example_config() {
},
},
export_metrics: ExportMetricsOption {
self_import: Some(Default::default()),
self_import: None,
remote_write: Some(Default::default()),
..Default::default()
},
@@ -176,12 +164,13 @@ fn test_load_metasrv_example_config() {
component: MetasrvOptions {
selector: SelectorType::default(),
data_home: DEFAULT_DATA_HOME.to_string(),
server_addr: "127.0.0.1:3002".to_string(),
grpc: GrpcOptions {
bind_addr: "127.0.0.1:3002".to_string(),
server_addr: "127.0.0.1:3002".to_string(),
..Default::default()
},
logging: LoggingOptions {
dir: Path::new(DEFAULT_DATA_HOME)
.join(DEFAULT_LOGGING_DIR)
.to_string_lossy()
.to_string(),
dir: format!("{}/{}", DEFAULT_DATA_HOME, DEFAULT_LOGGING_DIR),
level: Some("info".to_string()),
otlp_endpoint: Some(DEFAULT_OTLP_ENDPOINT.to_string()),
tracing_sample_ratio: Some(Default::default()),
@@ -195,7 +184,7 @@ fn test_load_metasrv_example_config() {
},
},
export_metrics: ExportMetricsOption {
self_import: Some(Default::default()),
self_import: None,
remote_write: Some(Default::default()),
..Default::default()
},
@@ -216,12 +205,7 @@ fn test_load_standalone_example_config() {
component: StandaloneOptions {
default_timezone: Some("UTC".to_string()),
wal: DatanodeWalConfig::RaftEngine(RaftEngineConfig {
dir: Some(
Path::new(DEFAULT_DATA_HOME)
.join(WAL_DIR)
.to_string_lossy()
.to_string(),
),
dir: Some(format!("{}/{}", DEFAULT_DATA_HOME, WAL_DIR)),
sync_period: Some(Duration::from_secs(10)),
recovery_parallelism: 2,
..Default::default()
@@ -244,10 +228,7 @@ fn test_load_standalone_example_config() {
},
logging: LoggingOptions {
level: Some("info".to_string()),
dir: Path::new(DEFAULT_DATA_HOME)
.join(DEFAULT_LOGGING_DIR)
.to_string_lossy()
.to_string(),
dir: format!("{}/{}", DEFAULT_DATA_HOME, DEFAULT_LOGGING_DIR),
otlp_endpoint: Some(DEFAULT_OTLP_ENDPOINT.to_string()),
tracing_sample_ratio: Some(Default::default()),
..Default::default()

View File

@@ -0,0 +1,240 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//! [CancellationHandle] is used to compose with manual implementation of [futures::future::Future]
//! or [futures::stream::Stream] to facilitate cancellation.
//! See example in [frontend::stream_wrapper::CancellableStreamWrapper] and [CancellableFuture].
use std::fmt::{Debug, Display, Formatter};
use std::future::Future;
use std::pin::Pin;
use std::sync::atomic::{AtomicBool, Ordering};
use std::sync::Arc;
use std::task::{Context, Poll};
use futures::task::AtomicWaker;
use pin_project::pin_project;
#[derive(Default)]
pub struct CancellationHandle {
waker: AtomicWaker,
cancelled: AtomicBool,
}
impl Debug for CancellationHandle {
fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
f.debug_struct("CancellationHandle")
.field("cancelled", &self.is_cancelled())
.finish()
}
}
impl CancellationHandle {
pub fn waker(&self) -> &AtomicWaker {
&self.waker
}
/// Cancels a future or stream.
pub fn cancel(&self) {
if self
.cancelled
.compare_exchange(false, true, Ordering::Acquire, Ordering::Relaxed)
.is_ok()
{
self.waker.wake();
}
}
/// Is this handle cancelled.
pub fn is_cancelled(&self) -> bool {
self.cancelled.load(Ordering::Relaxed)
}
}
#[pin_project]
#[derive(Debug, Clone)]
pub struct CancellableFuture<T> {
#[pin]
fut: T,
handle: Arc<CancellationHandle>,
}
impl<T> CancellableFuture<T> {
pub fn new(fut: T, handle: Arc<CancellationHandle>) -> Self {
Self { fut, handle }
}
}
impl<T> Future for CancellableFuture<T>
where
T: Future,
{
type Output = Result<T::Output, Cancelled>;
fn poll(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Self::Output> {
let this = self.as_mut().project();
// Check if the task has been aborted
if this.handle.is_cancelled() {
return Poll::Ready(Err(Cancelled));
}
if let Poll::Ready(x) = this.fut.poll(cx) {
return Poll::Ready(Ok(x));
}
this.handle.waker().register(cx.waker());
if this.handle.is_cancelled() {
return Poll::Ready(Err(Cancelled));
}
Poll::Pending
}
}
#[derive(Copy, Clone, Debug)]
pub struct Cancelled;
impl Display for Cancelled {
fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
write!(f, "Future has been cancelled")
}
}
#[cfg(test)]
mod tests {
use std::sync::Arc;
use std::time::Duration;
use tokio::time::{sleep, timeout};
use crate::cancellation::{CancellableFuture, CancellationHandle, Cancelled};
#[tokio::test]
async fn test_cancellable_future_completes_normally() {
let handle = Arc::new(CancellationHandle::default());
let future = async { 42 };
let cancellable = CancellableFuture::new(future, handle);
let result = cancellable.await;
assert!(result.is_ok());
assert_eq!(result.unwrap(), 42);
}
#[tokio::test]
async fn test_cancellable_future_cancelled_before_start() {
let handle = Arc::new(CancellationHandle::default());
handle.cancel();
let future = async { 42 };
let cancellable = CancellableFuture::new(future, handle);
let result = cancellable.await;
assert!(result.is_err());
assert!(matches!(result.unwrap_err(), Cancelled));
}
#[tokio::test]
async fn test_cancellable_future_cancelled_during_execution() {
let handle = Arc::new(CancellationHandle::default());
let handle_clone = handle.clone();
// Create a future that sleeps for a long time
let future = async {
sleep(Duration::from_secs(10)).await;
42
};
let cancellable = CancellableFuture::new(future, handle);
// Cancel the future after a short delay
tokio::spawn(async move {
sleep(Duration::from_millis(50)).await;
handle_clone.cancel();
});
let result = cancellable.await;
assert!(result.is_err());
assert!(matches!(result.unwrap_err(), Cancelled));
}
#[tokio::test]
async fn test_cancellable_future_completes_before_cancellation() {
let handle = Arc::new(CancellationHandle::default());
let handle_clone = handle.clone();
// Create a future that completes quickly
let future = async {
sleep(Duration::from_millis(10)).await;
42
};
let cancellable = CancellableFuture::new(future, handle);
// Try to cancel after the future should have completed
tokio::spawn(async move {
sleep(Duration::from_millis(100)).await;
handle_clone.cancel();
});
let result = cancellable.await;
assert!(result.is_ok());
assert_eq!(result.unwrap(), 42);
}
#[tokio::test]
async fn test_cancellation_handle_is_cancelled() {
let handle = CancellationHandle::default();
assert!(!handle.is_cancelled());
handle.cancel();
assert!(handle.is_cancelled());
}
#[tokio::test]
async fn test_multiple_cancellable_futures_with_same_handle() {
let handle = Arc::new(CancellationHandle::default());
let future1 = CancellableFuture::new(async { 1 }, handle.clone());
let future2 = CancellableFuture::new(async { 2 }, handle.clone());
// Cancel before starting
handle.cancel();
let (result1, result2) = tokio::join!(future1, future2);
assert!(result1.is_err());
assert!(result2.is_err());
assert!(matches!(result1.unwrap_err(), Cancelled));
assert!(matches!(result2.unwrap_err(), Cancelled));
}
#[tokio::test]
async fn test_cancellable_future_with_timeout() {
let handle = Arc::new(CancellationHandle::default());
let future = async {
sleep(Duration::from_secs(1)).await;
42
};
let cancellable = CancellableFuture::new(future, handle.clone());
// Use timeout to ensure the test doesn't hang
let result = timeout(Duration::from_millis(100), cancellable).await;
// Should timeout because the future takes 1 second but we timeout after 100ms
assert!(result.is_err());
}
#[tokio::test]
async fn test_cancelled_display() {
let cancelled = Cancelled;
assert_eq!(format!("{}", cancelled), "Future has been cancelled");
}
}

View File

@@ -14,6 +14,7 @@
pub mod bit_vec;
pub mod bytes;
pub mod cancellation;
pub mod plugins;
pub mod range_read;
#[allow(clippy::all)]

View File

@@ -102,6 +102,8 @@ pub const INFORMATION_SCHEMA_FLOW_TABLE_ID: u32 = 33;
pub const INFORMATION_SCHEMA_PROCEDURE_INFO_TABLE_ID: u32 = 34;
/// id for information_schema.region_statistics
pub const INFORMATION_SCHEMA_REGION_STATISTICS_TABLE_ID: u32 = 35;
/// id for information_schema.process_list
pub const INFORMATION_SCHEMA_PROCESS_LIST_TABLE_ID: u32 = 36;
// ----- End of information_schema tables -----

View File

@@ -7,5 +7,13 @@ license.workspace = true
[dependencies]
async-trait.workspace = true
common-error.workspace = true
common-grpc.workspace = true
common-macro.workspace = true
common-meta.workspace = true
greptime-proto.workspace = true
meta-client.workspace = true
snafu.workspace = true
tonic.workspace = true
[dev-dependencies]
tokio.workspace = true

View File

@@ -27,6 +27,35 @@ pub enum Error {
location: Location,
source: BoxedError,
},
#[snafu(display("Failed to list nodes from metasrv"))]
Meta {
source: Box<meta_client::error::Error>,
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Failed to parse process id: {}", s))]
ParseProcessId {
s: String,
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Failed to invoke frontend service"))]
InvokeFrontend {
#[snafu(source)]
error: tonic::Status,
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Failed to invoke list process service"))]
CreateChannel {
source: common_grpc::error::Error,
#[snafu(implicit)]
location: Location,
},
}
pub type Result<T> = std::result::Result<T, Error>;
@@ -36,6 +65,10 @@ impl ErrorExt for Error {
use Error::*;
match self {
External { source, .. } => source.status_code(),
Meta { source, .. } => source.status_code(),
ParseProcessId { .. } => StatusCode::InvalidArguments,
InvokeFrontend { .. } => StatusCode::Unexpected,
CreateChannel { source, .. } => source.status_code(),
}
}

View File

@@ -12,4 +12,41 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::fmt::{Display, Formatter};
use std::str::FromStr;
use snafu::OptionExt;
pub mod error;
pub mod selector;
#[derive(Debug, Clone, Eq, PartialEq)]
pub struct DisplayProcessId {
pub server_addr: String,
pub id: u64,
}
impl Display for DisplayProcessId {
fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
write!(f, "{}/{}", self.server_addr, self.id)
}
}
impl TryFrom<&str> for DisplayProcessId {
type Error = error::Error;
fn try_from(value: &str) -> Result<Self, Self::Error> {
let mut split = value.split('/');
let server_addr = split
.next()
.context(error::ParseProcessIdSnafu { s: value })?
.to_string();
let id = split
.next()
.context(error::ParseProcessIdSnafu { s: value })?;
let id = u64::from_str(id)
.ok()
.context(error::ParseProcessIdSnafu { s: value })?;
Ok(DisplayProcessId { server_addr, id })
}
}

View File

@@ -0,0 +1,112 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use std::time::Duration;
use common_grpc::channel_manager::{ChannelConfig, ChannelManager};
use common_meta::cluster::{ClusterInfo, NodeInfo, Role};
use greptime_proto::v1::frontend::{
frontend_client, KillProcessRequest, KillProcessResponse, ListProcessRequest,
ListProcessResponse,
};
use meta_client::MetaClientRef;
use snafu::ResultExt;
use tonic::Response;
use crate::error;
use crate::error::{MetaSnafu, Result};
pub type FrontendClientPtr = Box<dyn FrontendClient>;
#[async_trait::async_trait]
pub trait FrontendClient: Send {
async fn list_process(&mut self, req: ListProcessRequest) -> Result<ListProcessResponse>;
async fn kill_process(&mut self, req: KillProcessRequest) -> Result<KillProcessResponse>;
}
#[async_trait::async_trait]
impl FrontendClient for frontend_client::FrontendClient<tonic::transport::channel::Channel> {
async fn list_process(&mut self, req: ListProcessRequest) -> Result<ListProcessResponse> {
frontend_client::FrontendClient::<tonic::transport::channel::Channel>::list_process(
self, req,
)
.await
.context(error::InvokeFrontendSnafu)
.map(Response::into_inner)
}
async fn kill_process(&mut self, req: KillProcessRequest) -> Result<KillProcessResponse> {
frontend_client::FrontendClient::<tonic::transport::channel::Channel>::kill_process(
self, req,
)
.await
.context(error::InvokeFrontendSnafu)
.map(Response::into_inner)
}
}
#[async_trait::async_trait]
pub trait FrontendSelector {
async fn select<F>(&self, predicate: F) -> Result<Vec<FrontendClientPtr>>
where
F: Fn(&NodeInfo) -> bool + Send;
}
#[derive(Debug, Clone)]
pub struct MetaClientSelector {
meta_client: MetaClientRef,
channel_manager: ChannelManager,
}
#[async_trait::async_trait]
impl FrontendSelector for MetaClientSelector {
async fn select<F>(&self, predicate: F) -> Result<Vec<FrontendClientPtr>>
where
F: Fn(&NodeInfo) -> bool + Send,
{
let nodes = self
.meta_client
.list_nodes(Some(Role::Frontend))
.await
.map_err(Box::new)
.context(MetaSnafu)?;
nodes
.into_iter()
.filter(predicate)
.map(|node| {
let channel = self
.channel_manager
.get(node.peer.addr)
.context(error::CreateChannelSnafu)?;
let client = frontend_client::FrontendClient::new(channel);
Ok(Box::new(client) as FrontendClientPtr)
})
.collect::<Result<Vec<_>>>()
}
}
impl MetaClientSelector {
pub fn new(meta_client: MetaClientRef) -> Self {
let cfg = ChannelConfig::new()
.connect_timeout(Duration::from_secs(30))
.timeout(Duration::from_secs(30));
let channel_manager = ChannelManager::with_config(cfg);
Self {
meta_client,
channel_manager,
}
}
}

View File

@@ -0,0 +1,18 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
pub mod approximate;
#[cfg(feature = "geo")]
pub mod geo;
pub mod vector;

View File

@@ -0,0 +1,32 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use crate::function_registry::FunctionRegistry;
pub(crate) mod hll;
mod uddsketch;
pub(crate) struct ApproximateFunction;
impl ApproximateFunction {
pub fn register(registry: &FunctionRegistry) {
// uddsketch
registry.register_aggr(uddsketch::UddSketchState::state_udf_impl());
registry.register_aggr(uddsketch::UddSketchState::merge_udf_impl());
// hll
registry.register_aggr(hll::HllState::state_udf_impl());
registry.register_aggr(hll::HllState::merge_udf_impl());
}
}

View File

@@ -0,0 +1,27 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use crate::function_registry::FunctionRegistry;
mod encoding;
mod geo_path;
pub(crate) struct GeoFunction;
impl GeoFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register_aggr(geo_path::GeoPathAccumulator::uadf_impl());
registry.register_aggr(encoding::JsonPathAccumulator::uadf_impl());
}
}

View File

@@ -19,9 +19,12 @@ use common_error::status_code::StatusCode;
use common_macro::{as_aggr_func_creator, AggrFuncTypeStore};
use common_query::error::{self, InvalidInputStateSnafu, Result};
use common_query::logical_plan::accumulator::AggrFuncTypeStore;
use common_query::logical_plan::{Accumulator, AggregateFunctionCreator};
use common_query::logical_plan::{
create_aggregate_function, Accumulator, AggregateFunctionCreator,
};
use common_query::prelude::AccumulatorCreatorFunction;
use common_time::Timestamp;
use datafusion_expr::AggregateUDF;
use datatypes::prelude::ConcreteDataType;
use datatypes::value::{ListValue, Value};
use datatypes::vectors::VectorRef;
@@ -47,6 +50,16 @@ impl JsonPathAccumulator {
timestamp_type,
}
}
/// Create a new `AggregateUDF` for the `json_encode_path` aggregate function.
pub fn uadf_impl() -> AggregateUDF {
create_aggregate_function(
"json_encode_path".to_string(),
3,
Arc::new(JsonPathEncodeFunctionCreator::default()),
)
.into()
}
}
impl Accumulator for JsonPathAccumulator {

View File

@@ -47,7 +47,7 @@ impl GeoPathAccumulator {
Self::default()
}
pub fn udf_impl() -> AggregateUDF {
pub fn uadf_impl() -> AggregateUDF {
create_udaf(
GEO_PATH_NAME,
// Input types: lat, lng, timestamp

View File

@@ -0,0 +1,29 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use crate::aggrs::vector::product::VectorProduct;
use crate::aggrs::vector::sum::VectorSum;
use crate::function_registry::FunctionRegistry;
mod product;
mod sum;
pub(crate) struct VectorFunction;
impl VectorFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register_aggr(VectorSum::uadf_impl());
registry.register_aggr(VectorProduct::uadf_impl());
}
}

View File

@@ -16,8 +16,11 @@ use std::sync::Arc;
use common_macro::{as_aggr_func_creator, AggrFuncTypeStore};
use common_query::error::{CreateAccumulatorSnafu, Error, InvalidFuncArgsSnafu};
use common_query::logical_plan::{Accumulator, AggregateFunctionCreator};
use common_query::logical_plan::{
create_aggregate_function, Accumulator, AggregateFunctionCreator,
};
use common_query::prelude::AccumulatorCreatorFunction;
use datafusion_expr::AggregateUDF;
use datatypes::prelude::{ConcreteDataType, Value, *};
use datatypes::vectors::VectorRef;
use nalgebra::{Const, DVectorView, Dyn, OVector};
@@ -75,6 +78,16 @@ impl AggregateFunctionCreator for VectorProductCreator {
}
impl VectorProduct {
/// Create a new `AggregateUDF` for the `vec_product` aggregate function.
pub fn uadf_impl() -> AggregateUDF {
create_aggregate_function(
"vec_product".to_string(),
1,
Arc::new(VectorProductCreator::default()),
)
.into()
}
fn inner(&mut self, len: usize) -> &mut OVector<f32, Dyn> {
self.product.get_or_insert_with(|| {
OVector::from_iterator_generic(Dyn(len), Const::<1>, (0..len).map(|_| 1.0))

View File

@@ -16,8 +16,11 @@ use std::sync::Arc;
use common_macro::{as_aggr_func_creator, AggrFuncTypeStore};
use common_query::error::{CreateAccumulatorSnafu, Error, InvalidFuncArgsSnafu};
use common_query::logical_plan::{Accumulator, AggregateFunctionCreator};
use common_query::logical_plan::{
create_aggregate_function, Accumulator, AggregateFunctionCreator,
};
use common_query::prelude::AccumulatorCreatorFunction;
use datafusion_expr::AggregateUDF;
use datatypes::prelude::{ConcreteDataType, Value, *};
use datatypes::vectors::VectorRef;
use nalgebra::{Const, DVectorView, Dyn, OVector};
@@ -25,6 +28,7 @@ use snafu::ensure;
use crate::scalars::vector::impl_conv::{as_veclit, as_veclit_if_const, veclit_to_binlit};
/// The accumulator for the `vec_sum` aggregate function.
#[derive(Debug, Default)]
pub struct VectorSum {
sum: Option<OVector<f32, Dyn>>,
@@ -74,6 +78,16 @@ impl AggregateFunctionCreator for VectorSumCreator {
}
impl VectorSum {
/// Create a new `AggregateUDF` for the `vec_sum` aggregate function.
pub fn uadf_impl() -> AggregateUDF {
create_aggregate_function(
"vec_sum".to_string(),
1,
Arc::new(VectorSumCreator::default()),
)
.into()
}
fn inner(&mut self, len: usize) -> &mut OVector<f32, Dyn> {
self.sum
.get_or_insert_with(|| OVector::zeros_generic(Dyn(len), Const::<1>))

View File

@@ -0,0 +1,63 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
use datafusion_expr::ScalarUDF;
use crate::function::{FunctionContext, FunctionRef};
use crate::scalars::udf::create_udf;
/// A factory for creating `ScalarUDF` that require a function context.
#[derive(Clone)]
pub struct ScalarFunctionFactory {
name: String,
factory: Arc<dyn Fn(FunctionContext) -> ScalarUDF + Send + Sync>,
}
impl ScalarFunctionFactory {
/// Returns the name of the function.
pub fn name(&self) -> &str {
&self.name
}
/// Returns a `ScalarUDF` when given a function context.
pub fn provide(&self, ctx: FunctionContext) -> ScalarUDF {
(self.factory)(ctx)
}
}
impl From<ScalarUDF> for ScalarFunctionFactory {
fn from(df_udf: ScalarUDF) -> Self {
let name = df_udf.name().to_string();
let func = Arc::new(move |_ctx| df_udf.clone());
Self {
name,
factory: func,
}
}
}
impl From<FunctionRef> for ScalarFunctionFactory {
fn from(func: FunctionRef) -> Self {
let name = func.name().to_string();
let func = Arc::new(move |ctx: FunctionContext| {
create_udf(func.clone(), ctx.query_ctx, ctx.state)
});
Self {
name,
factory: func,
}
}
}

View File

@@ -16,11 +16,14 @@
use std::collections::HashMap;
use std::sync::{Arc, RwLock};
use datafusion_expr::AggregateUDF;
use once_cell::sync::Lazy;
use crate::admin::AdminFunction;
use crate::function::{AsyncFunctionRef, FunctionRef};
use crate::scalars::aggregate::{AggregateFunctionMetaRef, AggregateFunctions};
use crate::aggrs::approximate::ApproximateFunction;
use crate::aggrs::vector::VectorFunction as VectorAggrFunction;
use crate::function::{AsyncFunctionRef, Function, FunctionRef};
use crate::function_factory::ScalarFunctionFactory;
use crate::scalars::date::DateFunction;
use crate::scalars::expression::ExpressionFunction;
use crate::scalars::hll_count::HllCalcFunction;
@@ -31,18 +34,19 @@ use crate::scalars::matches_term::MatchesTermFunction;
use crate::scalars::math::MathFunction;
use crate::scalars::timestamp::TimestampFunction;
use crate::scalars::uddsketch_calc::UddSketchCalcFunction;
use crate::scalars::vector::VectorFunction;
use crate::scalars::vector::VectorFunction as VectorScalarFunction;
use crate::system::SystemFunction;
#[derive(Default)]
pub struct FunctionRegistry {
functions: RwLock<HashMap<String, FunctionRef>>,
functions: RwLock<HashMap<String, ScalarFunctionFactory>>,
async_functions: RwLock<HashMap<String, AsyncFunctionRef>>,
aggregate_functions: RwLock<HashMap<String, AggregateFunctionMetaRef>>,
aggregate_functions: RwLock<HashMap<String, AggregateUDF>>,
}
impl FunctionRegistry {
pub fn register(&self, func: FunctionRef) {
pub fn register(&self, func: impl Into<ScalarFunctionFactory>) {
let func = func.into();
let _ = self
.functions
.write()
@@ -50,6 +54,10 @@ impl FunctionRegistry {
.insert(func.name().to_string(), func);
}
pub fn register_scalar(&self, func: impl Function + 'static) {
self.register(Arc::new(func) as FunctionRef);
}
pub fn register_async(&self, func: AsyncFunctionRef) {
let _ = self
.async_functions
@@ -58,6 +66,14 @@ impl FunctionRegistry {
.insert(func.name().to_string(), func);
}
pub fn register_aggr(&self, func: AggregateUDF) {
let _ = self
.aggregate_functions
.write()
.unwrap()
.insert(func.name().to_string(), func);
}
pub fn get_async_function(&self, name: &str) -> Option<AsyncFunctionRef> {
self.async_functions.read().unwrap().get(name).cloned()
}
@@ -71,27 +87,16 @@ impl FunctionRegistry {
.collect()
}
pub fn register_aggregate_function(&self, func: AggregateFunctionMetaRef) {
let _ = self
.aggregate_functions
.write()
.unwrap()
.insert(func.name(), func);
}
pub fn get_aggr_function(&self, name: &str) -> Option<AggregateFunctionMetaRef> {
self.aggregate_functions.read().unwrap().get(name).cloned()
}
pub fn get_function(&self, name: &str) -> Option<FunctionRef> {
#[cfg(test)]
pub fn get_function(&self, name: &str) -> Option<ScalarFunctionFactory> {
self.functions.read().unwrap().get(name).cloned()
}
pub fn functions(&self) -> Vec<FunctionRef> {
pub fn scalar_functions(&self) -> Vec<ScalarFunctionFactory> {
self.functions.read().unwrap().values().cloned().collect()
}
pub fn aggregate_functions(&self) -> Vec<AggregateFunctionMetaRef> {
pub fn aggregate_functions(&self) -> Vec<AggregateUDF> {
self.aggregate_functions
.read()
.unwrap()
@@ -112,9 +117,6 @@ pub static FUNCTION_REGISTRY: Lazy<Arc<FunctionRegistry>> = Lazy::new(|| {
UddSketchCalcFunction::register(&function_registry);
HllCalcFunction::register(&function_registry);
// Aggregate functions
AggregateFunctions::register(&function_registry);
// Full text search function
MatchesFunction::register(&function_registry);
MatchesTermFunction::register(&function_registry);
@@ -127,15 +129,21 @@ pub static FUNCTION_REGISTRY: Lazy<Arc<FunctionRegistry>> = Lazy::new(|| {
JsonFunction::register(&function_registry);
// Vector related functions
VectorFunction::register(&function_registry);
VectorScalarFunction::register(&function_registry);
VectorAggrFunction::register(&function_registry);
// Geo functions
#[cfg(feature = "geo")]
crate::scalars::geo::GeoFunctions::register(&function_registry);
#[cfg(feature = "geo")]
crate::aggrs::geo::GeoFunction::register(&function_registry);
// Ip functions
IpFunctions::register(&function_registry);
// Approximate functions
ApproximateFunction::register(&function_registry);
Arc::new(function_registry)
});
@@ -147,12 +155,11 @@ mod tests {
#[test]
fn test_function_registry() {
let registry = FunctionRegistry::default();
let func = Arc::new(TestAndFunction);
assert!(registry.get_function("test_and").is_none());
assert!(registry.functions().is_empty());
registry.register(func);
assert!(registry.scalar_functions().is_empty());
registry.register_scalar(TestAndFunction);
let _ = registry.get_function("test_and").unwrap();
assert_eq!(1, registry.functions().len());
assert_eq!(1, registry.scalar_functions().len());
}
}

View File

@@ -18,13 +18,14 @@
mod admin;
mod flush_flow;
mod macros;
pub mod scalars;
mod system;
pub mod aggr;
pub mod aggrs;
pub mod function;
pub mod function_factory;
pub mod function_registry;
pub mod handlers;
pub mod helper;
pub mod scalars;
pub mod state;
pub mod utils;

View File

@@ -12,7 +12,6 @@
// See the License for the specific language governing permissions and
// limitations under the License.
pub mod aggregate;
pub(crate) mod date;
pub mod expression;
#[cfg(feature = "geo")]

View File

@@ -1,89 +0,0 @@
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//! # Deprecate Warning:
//!
//! This module is deprecated and will be removed in the future.
//! All UDAF implementation here are not maintained and should
//! not be used before they are refactored into the `src/aggr`
//! version.
use std::sync::Arc;
use common_query::logical_plan::AggregateFunctionCreatorRef;
use crate::function_registry::FunctionRegistry;
use crate::scalars::vector::product::VectorProductCreator;
use crate::scalars::vector::sum::VectorSumCreator;
/// A function creates `AggregateFunctionCreator`.
/// "Aggregator" *is* AggregatorFunction. Since the later one is long, we named an short alias for it.
/// The two names might be used interchangeably.
type AggregatorCreatorFunction = Arc<dyn Fn() -> AggregateFunctionCreatorRef + Send + Sync>;
/// `AggregateFunctionMeta` dynamically creates AggregateFunctionCreator.
#[derive(Clone)]
pub struct AggregateFunctionMeta {
name: String,
args_count: u8,
creator: AggregatorCreatorFunction,
}
pub type AggregateFunctionMetaRef = Arc<AggregateFunctionMeta>;
impl AggregateFunctionMeta {
pub fn new(name: &str, args_count: u8, creator: AggregatorCreatorFunction) -> Self {
Self {
name: name.to_string(),
args_count,
creator,
}
}
pub fn name(&self) -> String {
self.name.to_string()
}
pub fn args_count(&self) -> u8 {
self.args_count
}
pub fn create(&self) -> AggregateFunctionCreatorRef {
(self.creator)()
}
}
pub(crate) struct AggregateFunctions;
impl AggregateFunctions {
pub fn register(registry: &FunctionRegistry) {
registry.register_aggregate_function(Arc::new(AggregateFunctionMeta::new(
"vec_sum",
1,
Arc::new(|| Arc::new(VectorSumCreator::default())),
)));
registry.register_aggregate_function(Arc::new(AggregateFunctionMeta::new(
"vec_product",
1,
Arc::new(|| Arc::new(VectorProductCreator::default())),
)));
#[cfg(feature = "geo")]
registry.register_aggregate_function(Arc::new(AggregateFunctionMeta::new(
"json_encode_path",
3,
Arc::new(|| Arc::new(super::geo::encoding::JsonPathEncodeFunctionCreator::default())),
)));
}
}

View File

@@ -12,7 +12,6 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
mod date_add;
mod date_format;
mod date_sub;
@@ -27,8 +26,8 @@ pub(crate) struct DateFunction;
impl DateFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(DateAddFunction));
registry.register(Arc::new(DateSubFunction));
registry.register(Arc::new(DateFormatFunction));
registry.register_scalar(DateAddFunction);
registry.register_scalar(DateSubFunction);
registry.register_scalar(DateFormatFunction);
}
}

View File

@@ -17,8 +17,6 @@ mod ctx;
mod is_null;
mod unary;
use std::sync::Arc;
pub use binary::scalar_binary_op;
pub use ctx::EvalContext;
pub use unary::scalar_unary_op;
@@ -30,6 +28,6 @@ pub(crate) struct ExpressionFunction;
impl ExpressionFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(IsNullFunction));
registry.register_scalar(IsNullFunction);
}
}

View File

@@ -12,11 +12,9 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
pub(crate) mod encoding;
mod geohash;
mod h3;
mod helpers;
pub(crate) mod helpers;
mod measure;
mod relation;
mod s2;
@@ -29,57 +27,57 @@ pub(crate) struct GeoFunctions;
impl GeoFunctions {
pub fn register(registry: &FunctionRegistry) {
// geohash
registry.register(Arc::new(geohash::GeohashFunction));
registry.register(Arc::new(geohash::GeohashNeighboursFunction));
registry.register_scalar(geohash::GeohashFunction);
registry.register_scalar(geohash::GeohashNeighboursFunction);
// h3 index
registry.register(Arc::new(h3::H3LatLngToCell));
registry.register(Arc::new(h3::H3LatLngToCellString));
registry.register_scalar(h3::H3LatLngToCell);
registry.register_scalar(h3::H3LatLngToCellString);
// h3 index inspection
registry.register(Arc::new(h3::H3CellBase));
registry.register(Arc::new(h3::H3CellIsPentagon));
registry.register(Arc::new(h3::H3StringToCell));
registry.register(Arc::new(h3::H3CellToString));
registry.register(Arc::new(h3::H3CellCenterLatLng));
registry.register(Arc::new(h3::H3CellResolution));
registry.register_scalar(h3::H3CellBase);
registry.register_scalar(h3::H3CellIsPentagon);
registry.register_scalar(h3::H3StringToCell);
registry.register_scalar(h3::H3CellToString);
registry.register_scalar(h3::H3CellCenterLatLng);
registry.register_scalar(h3::H3CellResolution);
// h3 hierarchical grid
registry.register(Arc::new(h3::H3CellCenterChild));
registry.register(Arc::new(h3::H3CellParent));
registry.register(Arc::new(h3::H3CellToChildren));
registry.register(Arc::new(h3::H3CellToChildrenSize));
registry.register(Arc::new(h3::H3CellToChildPos));
registry.register(Arc::new(h3::H3ChildPosToCell));
registry.register(Arc::new(h3::H3CellContains));
registry.register_scalar(h3::H3CellCenterChild);
registry.register_scalar(h3::H3CellParent);
registry.register_scalar(h3::H3CellToChildren);
registry.register_scalar(h3::H3CellToChildrenSize);
registry.register_scalar(h3::H3CellToChildPos);
registry.register_scalar(h3::H3ChildPosToCell);
registry.register_scalar(h3::H3CellContains);
// h3 grid traversal
registry.register(Arc::new(h3::H3GridDisk));
registry.register(Arc::new(h3::H3GridDiskDistances));
registry.register(Arc::new(h3::H3GridDistance));
registry.register(Arc::new(h3::H3GridPathCells));
registry.register_scalar(h3::H3GridDisk);
registry.register_scalar(h3::H3GridDiskDistances);
registry.register_scalar(h3::H3GridDistance);
registry.register_scalar(h3::H3GridPathCells);
// h3 measurement
registry.register(Arc::new(h3::H3CellDistanceSphereKm));
registry.register(Arc::new(h3::H3CellDistanceEuclideanDegree));
registry.register_scalar(h3::H3CellDistanceSphereKm);
registry.register_scalar(h3::H3CellDistanceEuclideanDegree);
// s2
registry.register(Arc::new(s2::S2LatLngToCell));
registry.register(Arc::new(s2::S2CellLevel));
registry.register(Arc::new(s2::S2CellToToken));
registry.register(Arc::new(s2::S2CellParent));
registry.register_scalar(s2::S2LatLngToCell);
registry.register_scalar(s2::S2CellLevel);
registry.register_scalar(s2::S2CellToToken);
registry.register_scalar(s2::S2CellParent);
// spatial data type
registry.register(Arc::new(wkt::LatLngToPointWkt));
registry.register_scalar(wkt::LatLngToPointWkt);
// spatial relation
registry.register(Arc::new(relation::STContains));
registry.register(Arc::new(relation::STWithin));
registry.register(Arc::new(relation::STIntersects));
registry.register_scalar(relation::STContains);
registry.register_scalar(relation::STWithin);
registry.register_scalar(relation::STIntersects);
// spatial measure
registry.register(Arc::new(measure::STDistance));
registry.register(Arc::new(measure::STDistanceSphere));
registry.register(Arc::new(measure::STArea));
registry.register_scalar(measure::STDistance);
registry.register_scalar(measure::STDistanceSphere);
registry.register_scalar(measure::STArea);
}
}

View File

@@ -37,7 +37,7 @@ macro_rules! ensure_columns_len {
};
}
pub(super) use ensure_columns_len;
pub(crate) use ensure_columns_len;
macro_rules! ensure_columns_n {
($columns:ident, $n:literal) => {
@@ -58,7 +58,7 @@ macro_rules! ensure_columns_n {
};
}
pub(super) use ensure_columns_n;
pub(crate) use ensure_columns_n;
macro_rules! ensure_and_coerce {
($compare:expr, $coerce:expr) => {{
@@ -72,4 +72,4 @@ macro_rules! ensure_and_coerce {
}};
}
pub(super) use ensure_and_coerce;
pub(crate) use ensure_and_coerce;

View File

@@ -16,7 +16,6 @@
use std::fmt;
use std::fmt::Display;
use std::sync::Arc;
use common_query::error::{DowncastVectorSnafu, InvalidFuncArgsSnafu, Result};
use common_query::prelude::{Signature, Volatility};
@@ -27,7 +26,7 @@ use datatypes::vectors::{BinaryVector, MutableVector, UInt64VectorBuilder, Vecto
use hyperloglogplus::HyperLogLog;
use snafu::OptionExt;
use crate::aggr::HllStateType;
use crate::aggrs::approximate::hll::HllStateType;
use crate::function::{Function, FunctionContext};
use crate::function_registry::FunctionRegistry;
@@ -44,7 +43,7 @@ pub struct HllCalcFunction;
impl HllCalcFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(HllCalcFunction));
registry.register_scalar(HllCalcFunction);
}
}
@@ -117,6 +116,8 @@ impl Function for HllCalcFunction {
#[cfg(test)]
mod tests {
use std::sync::Arc;
use datatypes::vectors::BinaryVector;
use super::*;

View File

@@ -17,8 +17,6 @@ mod ipv4;
mod ipv6;
mod range;
use std::sync::Arc;
use cidr::{Ipv4ToCidr, Ipv6ToCidr};
use ipv4::{Ipv4NumToString, Ipv4StringToNum};
use ipv6::{Ipv6NumToString, Ipv6StringToNum};
@@ -31,15 +29,15 @@ pub(crate) struct IpFunctions;
impl IpFunctions {
pub fn register(registry: &FunctionRegistry) {
// Register IPv4 functions
registry.register(Arc::new(Ipv4NumToString));
registry.register(Arc::new(Ipv4StringToNum));
registry.register(Arc::new(Ipv4ToCidr));
registry.register(Arc::new(Ipv4InRange));
registry.register_scalar(Ipv4NumToString);
registry.register_scalar(Ipv4StringToNum);
registry.register_scalar(Ipv4ToCidr);
registry.register_scalar(Ipv4InRange);
// Register IPv6 functions
registry.register(Arc::new(Ipv6NumToString));
registry.register(Arc::new(Ipv6StringToNum));
registry.register(Arc::new(Ipv6ToCidr));
registry.register(Arc::new(Ipv6InRange));
registry.register_scalar(Ipv6NumToString);
registry.register_scalar(Ipv6StringToNum);
registry.register_scalar(Ipv6ToCidr);
registry.register_scalar(Ipv6InRange);
}
}

View File

@@ -12,7 +12,6 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
pub mod json_get;
mod json_is;
mod json_path_exists;
@@ -33,23 +32,23 @@ pub(crate) struct JsonFunction;
impl JsonFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(JsonToStringFunction));
registry.register(Arc::new(ParseJsonFunction));
registry.register_scalar(JsonToStringFunction);
registry.register_scalar(ParseJsonFunction);
registry.register(Arc::new(JsonGetInt));
registry.register(Arc::new(JsonGetFloat));
registry.register(Arc::new(JsonGetString));
registry.register(Arc::new(JsonGetBool));
registry.register_scalar(JsonGetInt);
registry.register_scalar(JsonGetFloat);
registry.register_scalar(JsonGetString);
registry.register_scalar(JsonGetBool);
registry.register(Arc::new(JsonIsNull));
registry.register(Arc::new(JsonIsInt));
registry.register(Arc::new(JsonIsFloat));
registry.register(Arc::new(JsonIsString));
registry.register(Arc::new(JsonIsBool));
registry.register(Arc::new(JsonIsArray));
registry.register(Arc::new(JsonIsObject));
registry.register_scalar(JsonIsNull);
registry.register_scalar(JsonIsInt);
registry.register_scalar(JsonIsFloat);
registry.register_scalar(JsonIsString);
registry.register_scalar(JsonIsBool);
registry.register_scalar(JsonIsArray);
registry.register_scalar(JsonIsObject);
registry.register(Arc::new(json_path_exists::JsonPathExistsFunction));
registry.register(Arc::new(json_path_match::JsonPathMatchFunction));
registry.register_scalar(json_path_exists::JsonPathExistsFunction);
registry.register_scalar(json_path_match::JsonPathMatchFunction);
}
}

View File

@@ -38,11 +38,11 @@ use crate::function_registry::FunctionRegistry;
///
/// Usage: matches(`<col>`, `<pattern>`) -> boolean
#[derive(Clone, Debug, Default)]
pub(crate) struct MatchesFunction;
pub struct MatchesFunction;
impl MatchesFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(MatchesFunction));
registry.register_scalar(MatchesFunction);
}
}

View File

@@ -77,7 +77,7 @@ pub struct MatchesTermFunction;
impl MatchesTermFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(MatchesTermFunction));
registry.register_scalar(MatchesTermFunction);
}
}

View File

@@ -18,7 +18,6 @@ mod pow;
mod rate;
use std::fmt;
use std::sync::Arc;
pub use clamp::{ClampFunction, ClampMaxFunction, ClampMinFunction};
use common_query::error::{GeneralDataFusionSnafu, Result};
@@ -39,13 +38,13 @@ pub(crate) struct MathFunction;
impl MathFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(ModuloFunction));
registry.register(Arc::new(PowFunction));
registry.register(Arc::new(RateFunction));
registry.register(Arc::new(RangeFunction));
registry.register(Arc::new(ClampFunction));
registry.register(Arc::new(ClampMinFunction));
registry.register(Arc::new(ClampMaxFunction));
registry.register_scalar(ModuloFunction);
registry.register_scalar(PowFunction);
registry.register_scalar(RateFunction);
registry.register_scalar(RangeFunction);
registry.register_scalar(ClampFunction);
registry.register_scalar(ClampMinFunction);
registry.register_scalar(ClampMaxFunction);
}
}

View File

@@ -12,7 +12,6 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
mod to_unixtime;
use to_unixtime::ToUnixtimeFunction;
@@ -23,6 +22,6 @@ pub(crate) struct TimestampFunction;
impl TimestampFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(ToUnixtimeFunction));
registry.register_scalar(ToUnixtimeFunction);
}
}

View File

@@ -16,7 +16,6 @@
use std::fmt;
use std::fmt::Display;
use std::sync::Arc;
use common_query::error::{DowncastVectorSnafu, InvalidFuncArgsSnafu, Result};
use common_query::prelude::{Signature, Volatility};
@@ -44,7 +43,7 @@ pub struct UddSketchCalcFunction;
impl UddSketchCalcFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(UddSketchCalcFunction));
registry.register_scalar(UddSketchCalcFunction);
}
}

View File

@@ -17,10 +17,8 @@ mod distance;
mod elem_product;
mod elem_sum;
pub mod impl_conv;
pub(crate) mod product;
mod scalar_add;
mod scalar_mul;
pub(crate) mod sum;
mod vector_add;
mod vector_dim;
mod vector_div;
@@ -30,37 +28,34 @@ mod vector_norm;
mod vector_sub;
mod vector_subvector;
use std::sync::Arc;
use crate::function_registry::FunctionRegistry;
pub(crate) struct VectorFunction;
impl VectorFunction {
pub fn register(registry: &FunctionRegistry) {
// conversion
registry.register(Arc::new(convert::ParseVectorFunction));
registry.register(Arc::new(convert::VectorToStringFunction));
registry.register_scalar(convert::ParseVectorFunction);
registry.register_scalar(convert::VectorToStringFunction);
// distance
registry.register(Arc::new(distance::CosDistanceFunction));
registry.register(Arc::new(distance::DotProductFunction));
registry.register(Arc::new(distance::L2SqDistanceFunction));
registry.register_scalar(distance::CosDistanceFunction);
registry.register_scalar(distance::DotProductFunction);
registry.register_scalar(distance::L2SqDistanceFunction);
// scalar calculation
registry.register(Arc::new(scalar_add::ScalarAddFunction));
registry.register(Arc::new(scalar_mul::ScalarMulFunction));
registry.register_scalar(scalar_add::ScalarAddFunction);
registry.register_scalar(scalar_mul::ScalarMulFunction);
// vector calculation
registry.register(Arc::new(vector_add::VectorAddFunction));
registry.register(Arc::new(vector_sub::VectorSubFunction));
registry.register(Arc::new(vector_mul::VectorMulFunction));
registry.register(Arc::new(vector_div::VectorDivFunction));
registry.register(Arc::new(vector_norm::VectorNormFunction));
registry.register(Arc::new(vector_dim::VectorDimFunction));
registry.register(Arc::new(vector_kth_elem::VectorKthElemFunction));
registry.register(Arc::new(vector_subvector::VectorSubvectorFunction));
registry.register(Arc::new(elem_sum::ElemSumFunction));
registry.register(Arc::new(elem_product::ElemProductFunction));
registry.register_scalar(vector_add::VectorAddFunction);
registry.register_scalar(vector_sub::VectorSubFunction);
registry.register_scalar(vector_mul::VectorMulFunction);
registry.register_scalar(vector_div::VectorDivFunction);
registry.register_scalar(vector_norm::VectorNormFunction);
registry.register_scalar(vector_dim::VectorDimFunction);
registry.register_scalar(vector_kth_elem::VectorKthElemFunction);
registry.register_scalar(vector_subvector::VectorSubvectorFunction);
registry.register_scalar(elem_sum::ElemSumFunction);
registry.register_scalar(elem_product::ElemProductFunction);
}
}

View File

@@ -23,7 +23,8 @@ use std::sync::Arc;
use build::BuildFunction;
use database::{
CurrentSchemaFunction, DatabaseFunction, ReadPreferenceFunction, SessionUserFunction,
ConnectionIdFunction, CurrentSchemaFunction, DatabaseFunction, PgBackendPidFunction,
ReadPreferenceFunction, SessionUserFunction,
};
use pg_catalog::PGCatalogFunction;
use procedure_state::ProcedureStateFunction;
@@ -36,13 +37,15 @@ pub(crate) struct SystemFunction;
impl SystemFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(BuildFunction));
registry.register(Arc::new(VersionFunction));
registry.register(Arc::new(CurrentSchemaFunction));
registry.register(Arc::new(DatabaseFunction));
registry.register(Arc::new(SessionUserFunction));
registry.register(Arc::new(ReadPreferenceFunction));
registry.register(Arc::new(TimezoneFunction));
registry.register_scalar(BuildFunction);
registry.register_scalar(VersionFunction);
registry.register_scalar(CurrentSchemaFunction);
registry.register_scalar(DatabaseFunction);
registry.register_scalar(SessionUserFunction);
registry.register_scalar(ReadPreferenceFunction);
registry.register_scalar(PgBackendPidFunction);
registry.register_scalar(ConnectionIdFunction);
registry.register_scalar(TimezoneFunction);
registry.register_async(Arc::new(ProcedureStateFunction));
PGCatalogFunction::register(registry);
}

View File

@@ -18,7 +18,8 @@ use std::sync::Arc;
use common_query::error::Result;
use common_query::prelude::{Signature, Volatility};
use datatypes::prelude::{ConcreteDataType, ScalarVector};
use datatypes::vectors::{StringVector, VectorRef};
use datatypes::vectors::{StringVector, UInt64Vector, VectorRef};
use derive_more::Display;
use crate::function::{Function, FunctionContext};
@@ -32,10 +33,20 @@ pub struct SessionUserFunction;
pub struct ReadPreferenceFunction;
#[derive(Display)]
#[display("{}", self.name())]
pub struct PgBackendPidFunction;
#[derive(Display)]
#[display("{}", self.name())]
pub struct ConnectionIdFunction;
const DATABASE_FUNCTION_NAME: &str = "database";
const CURRENT_SCHEMA_FUNCTION_NAME: &str = "current_schema";
const SESSION_USER_FUNCTION_NAME: &str = "session_user";
const READ_PREFERENCE_FUNCTION_NAME: &str = "read_preference";
const PG_BACKEND_PID: &str = "pg_backend_pid";
const CONNECTION_ID: &str = "connection_id";
impl Function for DatabaseFunction {
fn name(&self) -> &str {
@@ -117,6 +128,46 @@ impl Function for ReadPreferenceFunction {
}
}
impl Function for PgBackendPidFunction {
fn name(&self) -> &str {
PG_BACKEND_PID
}
fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
Ok(ConcreteDataType::uint64_datatype())
}
fn signature(&self) -> Signature {
Signature::nullary(Volatility::Immutable)
}
fn eval(&self, func_ctx: &FunctionContext, _columns: &[VectorRef]) -> Result<VectorRef> {
let pid = func_ctx.query_ctx.process_id();
Ok(Arc::new(UInt64Vector::from_slice([pid])) as _)
}
}
impl Function for ConnectionIdFunction {
fn name(&self) -> &str {
CONNECTION_ID
}
fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
Ok(ConcreteDataType::uint64_datatype())
}
fn signature(&self) -> Signature {
Signature::nullary(Volatility::Immutable)
}
fn eval(&self, func_ctx: &FunctionContext, _columns: &[VectorRef]) -> Result<VectorRef> {
let pid = func_ctx.query_ctx.process_id();
Ok(Arc::new(UInt64Vector::from_slice([pid])) as _)
}
}
impl fmt::Display for DatabaseFunction {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(f, "DATABASE")

View File

@@ -16,8 +16,6 @@ mod pg_get_userbyid;
mod table_is_visible;
mod version;
use std::sync::Arc;
use pg_get_userbyid::PGGetUserByIdFunction;
use table_is_visible::PGTableIsVisibleFunction;
use version::PGVersionFunction;
@@ -35,8 +33,8 @@ pub(super) struct PGCatalogFunction;
impl PGCatalogFunction {
pub fn register(registry: &FunctionRegistry) {
registry.register(Arc::new(PGTableIsVisibleFunction));
registry.register(Arc::new(PGGetUserByIdFunction));
registry.register(Arc::new(PGVersionFunction));
registry.register_scalar(PGTableIsVisibleFunction);
registry.register_scalar(PGGetUserByIdFunction);
registry.register_scalar(PGVersionFunction);
}
}

View File

@@ -296,6 +296,8 @@ pub struct ChannelConfig {
pub max_recv_message_size: ReadableSize,
// Max gRPC sending(encoding) message size
pub max_send_message_size: ReadableSize,
pub send_compression: bool,
pub accept_compression: bool,
}
impl Default for ChannelConfig {
@@ -316,6 +318,8 @@ impl Default for ChannelConfig {
client_tls: None,
max_recv_message_size: DEFAULT_MAX_GRPC_RECV_MESSAGE_SIZE,
max_send_message_size: DEFAULT_MAX_GRPC_SEND_MESSAGE_SIZE,
send_compression: false,
accept_compression: false,
}
}
}
@@ -566,6 +570,8 @@ mod tests {
client_tls: None,
max_recv_message_size: DEFAULT_MAX_GRPC_RECV_MESSAGE_SIZE,
max_send_message_size: DEFAULT_MAX_GRPC_SEND_MESSAGE_SIZE,
send_compression: false,
accept_compression: false,
},
default_cfg
);
@@ -610,6 +616,8 @@ mod tests {
}),
max_recv_message_size: DEFAULT_MAX_GRPC_RECV_MESSAGE_SIZE,
max_send_message_size: DEFAULT_MAX_GRPC_SEND_MESSAGE_SIZE,
send_compression: false,
accept_compression: false,
},
cfg
);

View File

@@ -64,6 +64,7 @@ impl Default for FlightEncoder {
}
impl FlightEncoder {
/// Creates new [FlightEncoder] with compression disabled.
pub fn with_compression_disabled() -> Self {
let write_options = writer::IpcWriteOptions::default()
.try_with_compression(None)

View File

@@ -35,6 +35,9 @@ pub const FLOWNODE_LEASE_SECS: u64 = DATANODE_LEASE_SECS;
/// The lease seconds of metasrv leader.
pub const META_LEASE_SECS: u64 = 5;
/// The keep-alive interval of the Postgres connection.
pub const POSTGRES_KEEP_ALIVE_SECS: u64 = 30;
/// In a lease, there are two opportunities for renewal.
pub const META_KEEP_ALIVE_INTERVAL_SECS: u64 = META_LEASE_SECS / 2;

View File

@@ -1001,7 +1001,7 @@ impl ErrorExt for Error {
}
#[cfg(any(feature = "pg_kvbackend", feature = "mysql_kvbackend"))]
RdsTransactionRetryFailed { .. } => StatusCode::Internal,
Error::DatanodeTableInfoNotFound { .. } => StatusCode::Internal,
DatanodeTableInfoNotFound { .. } => StatusCode::Internal,
}
}

View File

@@ -70,11 +70,12 @@ impl MetadataKey<'_, ViewInfoKey> for ViewInfoKey {
}
.build()
})?;
let captures = VIEW_INFO_KEY_PATTERN
.captures(key)
.context(InvalidViewInfoSnafu {
err_msg: format!("Invalid ViewInfoKey '{key}'"),
})?;
let captures =
VIEW_INFO_KEY_PATTERN
.captures(key)
.with_context(|| InvalidViewInfoSnafu {
err_msg: format!("Invalid ViewInfoKey '{key}'"),
})?;
// Safety: pass the regex check above
let view_id = captures[1].parse::<TableId>().unwrap();
Ok(ViewInfoKey { view_id })

View File

@@ -15,7 +15,7 @@
use std::sync::Arc;
use api::region::RegionResponse;
use api::v1::flow::{FlowRequest, FlowResponse};
use api::v1::flow::{DirtyWindowRequest, FlowRequest, FlowResponse};
use api::v1::region::{InsertRequests, RegionRequest};
pub use common_base::AffectedRows;
use common_query::request::QueryRequest;
@@ -42,6 +42,9 @@ pub trait Flownode: Send + Sync {
async fn handle(&self, request: FlowRequest) -> Result<FlowResponse>;
async fn handle_inserts(&self, request: InsertRequests) -> Result<FlowResponse>;
/// Handles requests to mark time window as dirty.
async fn handle_mark_window_dirty(&self, req: DirtyWindowRequest) -> Result<FlowResponse>;
}
pub type FlownodeRef = Arc<dyn Flownode>;

View File

@@ -14,6 +14,7 @@
pub mod file;
use std::borrow::Cow;
use std::fmt::{Display, Formatter};
use std::path::{Path, PathBuf};
use std::time::Instant;
@@ -271,6 +272,49 @@ impl MetadataSnapshotManager {
Ok((filename.to_string(), num_keyvalues as u64))
}
fn format_output(key: Cow<'_, str>, value: Cow<'_, str>) -> String {
format!("{} => {}", key, value)
}
pub async fn info(
object_store: &ObjectStore,
file_path: &str,
query_str: &str,
limit: Option<usize>,
) -> Result<Vec<String>> {
let path = Path::new(file_path);
let file_name = path
.file_name()
.and_then(|s| s.to_str())
.context(InvalidFilePathSnafu { file_path })?;
let filename = FileName::try_from(file_name)?;
let data = object_store
.read(file_path)
.await
.context(ReadObjectSnafu { file_path })?;
let document = Document::from_slice(&filename.extension.format, &data.to_bytes())?;
let metadata_content = document.into_metadata_content()?.values();
let mut results = Vec::with_capacity(limit.unwrap_or(256));
for kv in metadata_content {
let key_str = String::from_utf8_lossy(&kv.key);
if let Some(prefix) = query_str.strip_suffix('*') {
if key_str.starts_with(prefix) {
let value_str = String::from_utf8_lossy(&kv.value);
results.push(Self::format_output(key_str, value_str));
}
} else if key_str == query_str {
let value_str = String::from_utf8_lossy(&kv.value);
results.push(Self::format_output(key_str, value_str));
}
if results.len() == limit.unwrap_or(usize::MAX) {
break;
}
}
Ok(results)
}
}
#[cfg(test)]

View File

@@ -111,6 +111,11 @@ impl MetadataContent {
pub fn into_iter(self) -> impl Iterator<Item = KeyValue> {
self.values.into_iter()
}
/// Returns the key-value pairs as a vector.
pub fn values(self) -> Vec<KeyValue> {
self.values
}
}
/// The key-value pair of the backup file.

View File

@@ -15,7 +15,7 @@
use std::sync::Arc;
use api::region::RegionResponse;
use api::v1::flow::{FlowRequest, FlowResponse};
use api::v1::flow::{DirtyWindowRequest, FlowRequest, FlowResponse};
use api::v1::region::{InsertRequests, RegionRequest};
pub use common_base::AffectedRows;
use common_query::request::QueryRequest;
@@ -67,6 +67,14 @@ pub trait MockFlownodeHandler: Sync + Send + Clone {
) -> Result<FlowResponse> {
unimplemented!()
}
async fn handle_mark_window_dirty(
&self,
_peer: &Peer,
_req: DirtyWindowRequest,
) -> Result<FlowResponse> {
unimplemented!()
}
}
/// A mock struct implements [NodeManager] only implement the `datanode` method.
@@ -134,6 +142,10 @@ impl<T: MockFlownodeHandler> Flownode for MockNode<T> {
async fn handle_inserts(&self, requests: InsertRequests) -> Result<FlowResponse> {
self.handler.handle_inserts(&self.peer, requests).await
}
async fn handle_mark_window_dirty(&self, req: DirtyWindowRequest) -> Result<FlowResponse> {
self.handler.handle_mark_window_dirty(&self.peer, req).await
}
}
#[async_trait::async_trait]

View File

@@ -173,6 +173,7 @@ pub enum Error {
#[snafu(implicit)]
location: Location,
},
#[snafu(display("Stream timeout"))]
StreamTimeout {
#[snafu(implicit)]
@@ -180,6 +181,7 @@ pub enum Error {
#[snafu(source)]
error: tokio::time::error::Elapsed,
},
#[snafu(display("RecordBatch slice index overflow: {visit_index} > {size}"))]
RecordBatchSliceIndexOverflow {
#[snafu(implicit)]
@@ -187,6 +189,12 @@ pub enum Error {
size: usize,
visit_index: usize,
},
#[snafu(display("Stream has been cancelled"))]
StreamCancelled {
#[snafu(implicit)]
location: Location,
},
}
impl ErrorExt for Error {
@@ -221,6 +229,8 @@ impl ErrorExt for Error {
}
Error::StreamTimeout { .. } => StatusCode::Cancelled,
Error::StreamCancelled { .. } => StatusCode::Cancelled,
}
}

View File

@@ -372,6 +372,7 @@ impl DatanodeBuilder {
opts.max_concurrent_queries,
//TODO: revaluate the hardcoded timeout on the next version of datanode concurrency limiter.
Duration::from_millis(100),
opts.grpc.flight_compression,
);
let object_store_manager = Self::build_object_store_manager(&opts.storage).await?;

View File

@@ -50,6 +50,7 @@ use query::QueryEngineRef;
use servers::error::{self as servers_error, ExecuteGrpcRequestSnafu, Result as ServerResult};
use servers::grpc::flight::{FlightCraft, FlightRecordBatchStream, TonicStream};
use servers::grpc::region_server::RegionServerHandler;
use servers::grpc::FlightCompression;
use session::context::{QueryContextBuilder, QueryContextRef};
use snafu::{ensure, OptionExt, ResultExt};
use store_api::metric_engine_consts::{
@@ -80,6 +81,7 @@ use crate::event_listener::RegionServerEventListenerRef;
#[derive(Clone)]
pub struct RegionServer {
inner: Arc<RegionServerInner>,
flight_compression: FlightCompression,
}
pub struct RegionStat {
@@ -93,6 +95,7 @@ impl RegionServer {
query_engine: QueryEngineRef,
runtime: Runtime,
event_listener: RegionServerEventListenerRef,
flight_compression: FlightCompression,
) -> Self {
Self::with_table_provider(
query_engine,
@@ -101,6 +104,7 @@ impl RegionServer {
Arc::new(DummyTableProviderFactory),
0,
Duration::from_millis(0),
flight_compression,
)
}
@@ -111,6 +115,7 @@ impl RegionServer {
table_provider_factory: TableProviderFactoryRef,
max_concurrent_queries: usize,
concurrent_query_limiter_timeout: Duration,
flight_compression: FlightCompression,
) -> Self {
Self {
inner: Arc::new(RegionServerInner::new(
@@ -123,6 +128,7 @@ impl RegionServer {
concurrent_query_limiter_timeout,
),
)),
flight_compression,
}
}
@@ -536,7 +542,11 @@ impl FlightCraft for RegionServer {
.trace(tracing_context.attach(info_span!("RegionServer::handle_read")))
.await?;
let stream = Box::pin(FlightRecordBatchStream::new(result, tracing_context));
let stream = Box::pin(FlightRecordBatchStream::new(
result,
tracing_context,
self.flight_compression,
));
Ok(Response::new(stream))
}
}

View File

@@ -19,16 +19,16 @@ use std::time::Duration;
use api::region::RegionResponse;
use async_trait::async_trait;
use common_error::ext::BoxedError;
use common_function::function::FunctionRef;
use common_function::scalars::aggregate::AggregateFunctionMetaRef;
use common_function::function_factory::ScalarFunctionFactory;
use common_query::Output;
use common_runtime::runtime::{BuilderBuild, RuntimeTrait};
use common_runtime::Runtime;
use datafusion_expr::LogicalPlan;
use datafusion_expr::{AggregateUDF, LogicalPlan};
use query::dataframe::DataFrame;
use query::planner::LogicalPlanner;
use query::query_engine::{DescribeResult, QueryEngineState};
use query::{QueryEngine, QueryEngineContext};
use servers::grpc::FlightCompression;
use session::context::QueryContextRef;
use store_api::metadata::RegionMetadataRef;
use store_api::region_engine::{
@@ -76,9 +76,9 @@ impl QueryEngine for MockQueryEngine {
unimplemented!()
}
fn register_aggregate_function(&self, _func: AggregateFunctionMetaRef) {}
fn register_aggregate_function(&self, _func: AggregateUDF) {}
fn register_function(&self, _func: FunctionRef) {}
fn register_scalar_function(&self, _func: ScalarFunctionFactory) {}
fn read_table(&self, _table: TableRef) -> query::error::Result<DataFrame> {
unimplemented!()
@@ -98,6 +98,7 @@ pub fn mock_region_server() -> RegionServer {
Arc::new(MockQueryEngine),
Runtime::builder().build().unwrap(),
Box::new(NoopRegionServerEventListener),
FlightCompression::default(),
)
}

View File

@@ -326,6 +326,14 @@ impl Value {
}
}
/// Cast Value to [Duration]. Return None if value is not a valid duration data type.
pub fn as_duration(&self) -> Option<Duration> {
match self {
Value::Duration(d) => Some(*d),
_ => None,
}
}
/// Returns the logical type of the value.
pub fn logical_type_id(&self) -> LogicalTypeId {
match self {

View File

@@ -316,7 +316,7 @@ impl StreamingEngine {
);
METRIC_FLOW_ROWS
.with_label_values(&["out"])
.with_label_values(&["out-streaming"])
.inc_by(total_rows as u64);
let now = self.tick_manager.tick();

Some files were not shown because too many files have changed in this diff Show More