* refactor: make schedule request return value generic
* feat: add handler trait
* wip
* feat: use task handler
* fix: unit test
* refactor: separate scheduler mod
* chore: rename
* chore: Request use associate type
* refactor: use associate type
* refactor: use associate type to reduce generic parameters
* chore: further remove generic types
* chore: further remove a generic parameter
* feat: make args in coprocessor optional
* feat: supports kwargs for coprocessor as params passed by the users
* feat: supports params for /run-script
* fix: we should rewrite the coprocessor by removing kwargs
* fix: remove println
* fix: compile error after rebasing
* fix: improve http_handler_test
* test: http scripts api with user params
* refactor: tweak all to_owned
* feat: Add ContextProvider to Context
So procedures can query states of other procedures via the
ContextProvider and they don't need to hold a ProcedureManagerRef
* feat: Procedure supports acquring multiple lock keys
* test: Use multi-locks in test
* feat: Add keys_to_lock/unlock
* fix: correct date/time type format for postgresql
* fix: tests for timestamp
* refactor: use Utc datetime for timestamp::to_chrono_datetime
* Update src/servers/Cargo.toml
Co-authored-by: LFC <bayinamine@gmail.com>
---------
Co-authored-by: LFC <bayinamine@gmail.com>
* feat: trigger compaction on flush
* chore: rebase develop
* feat: add config item max_file_in_level0 and remove compaction_after_flush
* fix: cr comments
* chore: add unit test to cover Timestamp::new_inclusive
* fix: workaround to fix future is not Sync
* fix: future is not sync
* fix: some cr comments
* add DistLock trait and a implement based etcd
wip
impl lock grpc service for meta-srv
reuse the etcd client instead of repeatedly creating etcd client
add some docs and comments
add some comment
meta client support distribute lock
fix: dead lock
self-cr
* cr
* rename "expire" -> "expire_secs"
* feat: Runner executes procedure
* feat: Add rollback key type to ParsedKey
* feat: Write rollback key when procedure is unable to execute
* feat: Use loaded step to re-submit subprocedure
* feat: Track subprocedures in ProcedureMeta
* feat: Clean message cache after the root procedure is done
* feat: Runner returns execution result
* fix: Fix tests
* test: Test Runner
* test: Test procedures_in_tree
* chore: Refine test and comments
* feat: Remove support of lock inheritance
A deadlock happens if a subprocedure acquires the same lock key as
its parent.
The main concern is if the subprocedure directly inherits its parent's
lock, then how should we behave when multiple subprocedures acquire
this same lock? Each procedure may assume it has unique access to the
same object but it actually shares the resource with others.
Now subprocedures need to use different keys to lock objects, which is
reasonable. For example:
- A parent procedure wants to create a table so it locks the table with
a key like `catalog.schema.table`
- Subprocedures create regions for the table so they lock the regions
with keys `catalog.schema.table.region-0 ~ catalog.schema.table.region-n`
* style: Fix clippy
* feat: insert_procedure returns false on duplicate procedure
Also rename this method to try_insert_procedure
* chore: Address CR comments
* feat: implement "drop table" in distributed mode (both in SQL and gRPC)
refactor: create distributed table
some details:
- set table global value in Meta, as well as table routes value. Datanode only set table regional value
- complete instance SQL tests both in standalone and distributed mode
* fix: rebase develop
* fix: resolve PR comments
* feat: enable caching when using object store
* feat: support file cache for object store
* feat: maintaining the cached files with lru
* fix: improve the code
* empty commit
* improve the code
