* Add a simple cache implementation
* Add support for ReadOnlySqlLedgerWithMutableCache
* Add support for ReadOnlySqlLedgerWithTranslationCache
* Add support for sandbox-classic
changelog_begin
changelog_end
* Remove the mutating schema
changelog_begin
- [Participant] All participants now use the new append-only schema. Existing databases will
automatically upgrade to the new schema the first time a participant/ledger is started.
changelog_end
* Fix post-commit validation test
* Remove append-only flags from CI
* Don't crash when using deprecated flag
* Increase timeout for reset service tests
* Fix typos in parameter names
* Restore removed test
* Restore removed CLI check
* Improve CLI parameter description
* Support contextual logging in KeyValueConsumption
CHANGELOG_BEGIN
CHANGELOG_END
* Fix formatting and imports
* Propagate the contextual logging per-readUpdates call
* No need for at-construction logging context in LogAppendingReadServiceFactory
* No need for logging context in CommitStrategySupport.newReadServiceFactory
* Fix JdbcIndexerSpec
* Propagate error version switch to KeyValueConsumption
CHANGELOG_BEGIN
CHANGELOG_END
* Introduce v2 (self-service) KV error codes behind the CLI switch
* Test v2 codes and fix them
* Keep newline separating methods in KeyValueConsumption
* TransactionRejections: don't wrap updates in `Some`
* Factor errorVersionsTable
* Reorder imports
* Split "convert rejection to proto models and back to expected grpc code"
* Remove unneeded Option returned by decodeTransactionRejectionEntry
* Formatting fix
* Fix 7537e93d
* kvutils: Remove `VersionedOffsetBuilder.apply`.
Doesn't do anything.
* ledger-on-memory: Use `VersionedOffsetBuilder`.
* indexer-benchmark: Use `VersionedOffsetBuilder`.
* ledger-on-sql: Use `VersionedOffsetBuilder`.
* kvutils: Use `VersionedOffsetBuilder` in tests.
* kvutils: Create a case class for `VersionedOffsetBuilder#split`.
* kvutils: Delete unused methods from the offset builders.
* kvutils: Use `Bytes#startWith` to check the offset version.
* kvutils: Move offset splitting into `VersionedOffset`.
* kvutils: Extract out versioned offset generators.
* kvutils: Replace `OffsetBuilder` with `VersionedOffsetMutator`.
This takes care of the last usages of `OffsetBuilder`, which were to
modify the lowest component of the offset.
CHANGELOG_BEGIN
CHANGELOG_END
* kvutils: Randomize the offset version in testing.
To make sure we don't use a hard-coded offset version anywhere.
* kvutils: `Random.between` is not available in Scala 2.12.
* kvutils: Move offset mutation methods to `VersionedOffset`.
* kvutils: Move the versioned offset construction into `VersionedOffset`.
The `VersionedOffsetBuilder` is still useful as it remembers the
version, meaning we only need to specify it once.
* Add deduplication and submission id to the proto submitter info so we can build the v2 completion info
* Use v2 rejections and store the enriched model in protobuf
* Update v1 SubmissionResult to v2 SubmissionResult
* Update v1 imports to v2 imports
* Enable API deduplication for ledgers that don't have it implemented yet (all kv)
* Update tests for new rejections
* Disable buf check for oneof
* Code review - fix formatting and test assertions
* Update imports for proto changes
* Add tests for rejection conversions
* Use entry id as submission id when submission id is missing
* Calculate deduplication period for rejection based on the set deduplicate until by calculating the duration relative to record time
* Code review consistent naming and comments
* Submission id is optional in the completion info
the submissionId is not set for participant.state.v1 entries, therefore for the transition from v1 to the v2 API we need to account for the missing submission ids by making it optional in the completion info.
* Update messages and grpc status codes for rejections to keep backward compatibility
* Deprecate proto deduplicate_until
* Set deduplication value in the commit context for all the deduplication periods. This change allows us to re-enable the committer deduplication tests
* Set ledger configuration for tests
* Command dedup: In the committer set the deduplicate until context value based on the given deduplication period [KVL-1049] (#10743)
CHANGELOG_BEGIN
participant-state:
- Migrate to use only v2 Read/Write Services. This includes the use of new models for rejections/updates/submission results.
- Calculate deduplicate until for committer side deduplication based on the submitter given deduplication period
- if using the deprecated deduplicateUntil then just set the given timestamp
- if using duration then calculate the new deduplicateUntil by using the formula (submissionTime + deduplicationDuration + minSkew)
- if using offset deduplication then calculate deduplicateUntil by using the formula (submissionTime + maxDeduplicationDuration + minSkew)
- if the deduplication period is not set then we don't set deduplicateUntil
- Emit completions with status `ALREADY_EXISTS` for duplicate commands
CHANGELOG_END
* Enable append-only schema in sandbox
changelog_begin
changelog_end
* Add test
* Unhide the compression flag in sandbox
* Make append-only config non-optional
* Upgrade Scalatest to v3.2.9.
Because of some coupling we also have to upgrade Scalaz to the latest
v7.2 point release, v7.2.33.
The Scalatest changes are quite involved because the JAR has been broken
up into several smaller JARs. Because Bazel expects us to specify all
dependencies and doesn't allow transitive dependencies to be used
directly, this means that we need to specify the explicit Scalatest
components that we use.
As you can imagine, this results in quite a big set of changes. They
are, however, constrained to dependency management; all the code remains
the same.
CHANGELOG_BEGIN
CHANGELOG_END
* http-json-oracle: Fix a Scalatest dependency.
* ledger-api-client: Fix a Scalatest dependency.
* participant-state: Give `ChangeId` its own file.
* participant-state: Introduce `v1.CompletionInfo` for a while.
`CompletionInfo` is identical to `SubmitterInfo`. Its purpose is to make
the migration to v2 easier. It should not make it into the final
version.
* participant-integration-api: Switch to the v2 participant state API.
This means that:
- the API server and indexer expect v2 API traits
- adapters are provided if you need to elevate your v1 API usage
- the indexer internally uses v2 Updates
- rejections are switched over to use the v2 format
- Sandbox Classic uses v2 as the internal representation too (because
it interacts directly with the underlying indexer representation,
and is therefore tightly coupled)
_kvutils_ and other users of the `StandaloneApiServer` and
`StandaloneIndexerServer` use the adapters.
CHANGELOG_BEGIN
- [Integration Kit] The API server and indexer have switched over to v2
of the participant-state API. You can continue to use the v1 API, but
you will need to wrap your ``ReadService`` and ``WriteService``
objects in the ``AdaptedV1ReadService`` and ``AdaptedV1WriteSerivce``
classes.
CHANGELOG_END
* participant-state: Remove v1.CompletionInfo.
It's served its purpose.
* kvutils: Remove an unnecessary line from `Runner`.
* ledger-api-common: Delete a TODO; we'll track it elsewhere.
* participant-integration-api: Use full words in `JdbcLedgerDao`.
Just a little bit of cleanup.
* ledger-api-common: Extract out the random submission ID generator.
And introduce a trait, because, well, this is still the JVM.
* participant-state: Remove the aliases to `Ref.LedgerString`.
Use the aliases directly in `Ref` instead.
CHANGELOG_BEGIN
CHANGELOG_END
* daml-lf/data: Improve comments on ID aliases.
* daml-lf/data: Fix a typo in a comment.
Co-authored-by: Miklos <57664299+miklos-da@users.noreply.github.com>
Co-authored-by: Miklos <57664299+miklos-da@users.noreply.github.com>
* participant-state: Remove the `ParticipantId` alias.
This alias adds nothing. By using `Ref.ParticipantId` directly, many
packages can remove their dependency on the _participant-state_ package.
CHANGELOG_BEGIN
CHANGELOG_END
* participant-state: Remove the `PackageId` and `Party` aliases.
They don't add anything. Let's just use `Ref`.
* kvutils: Restore missing compat imports.
* ledger-offset: Move `Offset` to a new package.
CHANGELOG_BEGIN
- [Integration Kit] The ``Offset`` type has been moved to a new Maven
package, ``ledger-offset``, from the ``participant-state`` package.
The Java package has been renamed to ``com.daml.ledger.offset``. If
you are using this type, you will need to update your dependencies and
imports.
CHANGELOG_END
* Avoid rewrapping offsets for no reason.
Co-authored-by: Miklos <57664299+miklos-da@users.noreply.github.com>
* participant-integration-api: Sort some imports.
* participant-integration-api: Fix dependencies for the Oracle tests.
I didn't add `ledger-offset`.
Co-authored-by: Miklos <57664299+miklos-da@users.noreply.github.com>
* ledger-configuration: Extract configuration from participant-state.
The configuration is often used without the state, and doesn't need to
be versioned in the same way.
CHANGELOG_BEGIN
- [Integration Kit] The ledger configuration classes, ``Configuration``,
``LedgerInitialConditions``, and ``TimeModel``, have been moved from
*participant-state* to a separate package named
*ledger-configuration*, in the Java package
``com.daml.ledger.configuration``. You will need to update your
dependencies and imports.
CHANGELOG_END
* participant-state: Remove the `LedgerId` aliases.
* ledger-configuration: Rename `TimeModel` to `LedgerTimeModel`.
This avoids confusion with the protobuf-generated `TimeModel` classes.
CHANGELOG_BEGIN
- [Integration Kit] ``TimeModel`` has been renamed to
``LedgerTimeModel``. If you are using the ledger configuration classes
directly, you may need to update your code.
CHANGELOG_END
* ledger-configuration: Remove colons in LedgerInitialConditions' docs.
* kvutils: Restore a missing compat import.
* participant-integration-api: Add ledger-configuration to Oracle tests.
* sandbox-common: Fix `--max-ledger-time-skew` docs.
Co-authored-by: Miklos <57664299+miklos-da@users.noreply.github.com>
Co-authored-by: Miklos <57664299+miklos-da@users.noreply.github.com>
* logging-entries: Split from contextualized-logging.
This allows us to introduce it to Daml-LF without bringing in the
Logback, Logstash, and gRPC dependencies.
CHANGELOG_BEGIN
CHANGELOG_END
* logging-entries: Fix dependencies for 2.12.
* logging-entries: Missed one more Scala 2.12 dependency.
* release: Publish logging-entries.
* 2 seconds rather than 250ms for default hikari connection pool timeout
CHANGELOG_BEGIN
[jdbc ledger] increase default hikari connection pool timeout to 2s from 250ms
CHANGELOG_END
* extract hikari connection timeout configuration
* default of 250ms
* read hikari connection timeout default from config rather than hardcoding for sandbox next
* remove unused import
changelog_begin
- [Ledger HTTP Json Service] Logging output can now be in JSON either via providing the cli option `--log-encoder json` or via setting the env var `LOG_FORMAT_JSON=true`
changelog_end
In replicated Postgres setups, running with SYNCHRONOUS_COMMIT=LOCAL is a
suitable option as it is a trade off to only locally commit synchronously
but not to wait for a network ack.
CHANGELOG_BEGIN
CHANGELOG_END
[ledger-api-server] Make database connection pool size configurable
CHANGELOG_BEGIN
[Daml Driver for PostgreSQL]: Added CLI option --database-connection-pool-size
to configure the size of the database connection pool
[Integration Kit]: Added the CLI options api-server-connection-pool-size and
indexer-server-connection-pool-size to configure the database connection pool size
for the Ledger API Server and the indexer respectively.
CHANGELOG_END
* ledger-on-memory: Rename `PreExecutingOwner` to just `Owner`.
* ledger-on-memory: Push the committer execution context up a little.
* ledger-on-memory: Use pre-execution in tests, not batching.
* ledger-on-memory: Remove support for batching submissions.
Pre-execution is pretty much always faster; let's use it.
CHANGELOG_BEGIN
CHANGELOG_END
* ledger-on-memory: Simplify construction.
Now we only support pre-execution, we can be less general.
* participant-integration-api: Reuse the services EC for data munging.
Rather than spawning a separate thread pool for manipulating data before
and after index queries, we can just re-use the services EC, which is
already a work-stealing thread pool.
* kvutils: Share the services EC with the indexer.
CHANGELOG_BEGIN
CHANGELOG_END
* Port the rest of //ledger/... to Scala 2.13
draw the rest of the fcking owl
Omitted for now are the ledger API test tool which has a dependency
only compatible with 2.12 and the generated code of the Scala
codegen (the codegen compiles and runs with 2.13, the generated code
does not).
changelog_begin
changelog_end
* Less symbols
changelog_begin
changelog_end
This PR updates scalafmt and enables trailingCommas =
multiple. Unfortunately, scalafmt broke the version field which means
we cannot fully preserve the rest of the config. I’ve made some
attempts to stay reasonably close to the original config but couldn’t
find an exact equivalent in a lot of cases. I don’t feel strongly
about any of the settings so happy to change them to something else.
As announced, this will be merged on Saturday to avoid too many conflicts.
changelog_begin
changelog_end
This is necessary to at least attempt an upgrade to 2.13 and
generally, I want to keep our rulesets up2date. rules-scala forces the
version of scalatest so we have to bump that at the same time.
This requires changes to basically all Scala test suites since the
import structure has changed and a bunch of things (primarily
scalacheck support) got split out.
Apologies for the giant PR, I don’t see a way to keep it smaller.
changelog_begin
changelog_end
* resources: Move builders into //ledger/ledger-resources.
Keep the actual constructors in a trait, but instantiate it when working
with ledger code.
This allows us to later introduce an extra "context" type parameter to
ResourceOwner.
* resources-akka: Move the builders in to //ledger/ledger-resources.
* resources: Introduce an abstract `Context` parameter for owners.
This replaces the concrete `ExecutionContext`. While it _can_ be an
execution context, it really doesn't matter as long as we can get at one
somehow.
This is being introduced so we can wrap the context in a container,
either for type tagging or to include extra information.
Because our current context _is_ `ExecutionContext`, and an implicit is
provided to extract it, we can end up with two ways to get the same
value. We use shadowing to prevent this. This problem should go away in
the near future when a new context type is added.
CHANGELOG_BEGIN
- [Integration Kit] The `ResourceOwner` type is now parameterized by a
`Context`, which is filled in by the corresponding `Context` class in
the _ledger-resources_ dependency. This allows us to pass extra
information through resource acquisition.
CHANGELOG_END
* ledger-resources: Move `ResourceOwner` here from `resources`.
* ledger-resources: Remove dependencies from outside //ledger.
* ledger-resource: Wrap the acquisition execution context in `Context`.
So we can add a logging context to it.
* resources: Pass the Context, not the ExecutionContext, to Resource.
* Avoid importing `HasExecutionContext`.
* ledger-resources: Publish to Maven Central.
* resources: Make the small changes suggested by @stefanobaghino-da.
Co-Authored-By: Stefano Baghino <43749967+stefanobaghino-da@users.noreply.github.com>
* ledger-resources: Pull out a trait for test resource contexts.
Saves a few lines of code.
* Restore some imports that were accidentally wildcarded.
* resources: Replace an `implicit def` with a couple of imports.
* participant-integration-api: Simplify the JdbcLedgerDaoBackend tests.
Try and use the right execution context where possible.
Co-authored-by: Stefano Baghino <43749967+stefanobaghino-da@users.noreply.github.com>
This PR creates 3 validation modes:
* `Strict`: Specifies that the committer should validate the package
before committing them to the ledger. When using this mode, the
packages committed to the ledger can be fully trusted and do not
have to be validated when loaded into the engine.
* `Lenient`: Specifies that the committer should perform a fast
validation of the packages before committing them to the ledger.
This mode is useful for ledger integrations that cannot handle
long-running submissions (> 10s). When using this mode, the
packages committed to the ledger cannot be trusted and must be
validated every time they are loaded into the engine.
* `No`: Specifies that the committer should not perform any
validation the packages before committing them to the ledger. This
should be used only by non distributed ledgers, like DAML-on-SQL,
where the validation done in the API server can be trusted.
This PR creates 3 preloading modes:
* `Synchronous` : Specifies that the packages should be preloading
into the engine before committed.
* `Asynchronous`: Specifies that the packages should be preloaded into
the engine asynchronously with the rest of the commit process. This
mode is useful for ledger integrations that cannot handle
long-running submissions (> 10s). Failure of the preloading process
will not affect the commit.
* `No`: Specifies that the packages should not be preloaded into
the engine.
CHANGELOG_BEGIN
- [Integration Kit] In kvutils, add metric
daml.kvutils.committer.package_upload.validate_timer to track
package validation time.
CHANGELOG_END
* Move in-mem writer's `ledgerStateAccess.inTransaction` down to committer
* Move `BatchedSubmissionValidator` and spec into `batch` subpackage
* Add `StateAccessingValidatingCommitter` and inherit it in batching one
* Document `StateAccessingValidatingCommitter`
* Generalize the committer for `InMemoryLedgerReaderWriter`
* `envelope` -> `submissionEnvelope` in validating committers
* Add `PreExecutingValidatingCommitter` and sub-components
* Add retry in case of conflict in `PreExecutingValidatingCommitter`
CHANGELOG_BEGIN
CHANGELOG_END
* Fix compilation error
* Hook pre-execution in `daml-on-memory-kv`
* Add fake time updates provider
* Fix `BatchedValidatingCommitterSpec`
* Don't use batched writer with pre-execution
* Fix conflict detection
* Fix out-of-time-bounds detection
* Prefix/unprefix serialized log entry IDs in pre-execution write sets
* Fix: produce an out-of-bounds rejection log entry in transaction rejected cases too
* Fix `SubmissionResult` return in case of repeated pre-exec conflict
* Fidelity level 1: sequential pre-execution
* Documentation for pre-execution support in DAML-on-Memory KV
* Add ledger-on-memory conformance test with pre-execution enabled
* Revert "Fix: produce an out-of-bounds rejection log entry in transaction rejected cases too"
This reverts commit 4df7e26b
* Fix test
* Improve naming and documentation
* Address review comments
* Fix test
* Fix wrong implementation used for `ParticipantStateIntegrationSpecBase` tests
* Address review comments
* Address review comments
* Address minor review comments
* Move public code into daml-integration-api
CHANGELOG_BEGIN
[DAML Integration Kit]: Removed sandbox specific code from the API intended to be used by ledger integrations. Use the maven coordinates ``com.daml:participant-integration-api:VERSION`` instead of ``com.daml:ledger-api-server`` or ``com.daml:sandbox``.
CHANGELOG_END
Also fixes#5635, removing usage of `ReadService` from the `StandaloneApiServer`.
The configuration stream of the Ledger API LedgerConfigurationService is now properly backed by the configuration entries instead of just serving the initial configuration.
CHANGELOG_BEGIN
[DAML Integration Kit]: ``StandaloneApiServer`` can now be run in a read-only mode.
- The type of the constructor parameter ``writeService`` of ``StandaloneApiServer`` changed to ``Option[WriteService]``. Passing ``None`` will not start any of the admin services, the command service, and the command submission service.
- The constructor parameter ``readService`` of ``StandaloneApiServer`` has been removed.
- A new constructor parameter ``ledgerId`` has been added to ``StandaloneApiServer``. It is used to verify that that ``StandaloneApiServer`` is run against an index storage for the same ledgerId. Initialization is aborted if this is not the case.
[DAML Integration Kit]: The ``LedgerConfigurationService`` now properly streams configuration changes.
CHANGELOG_END
* Extract caching from participant-state as a library
This will be used to keep a cache of values to cut on LF translation cost when serving transactions.
changelog_begin
changelog_end
* Add dependency where missing
- Only upload packages during the initial startup.
- Avoid loading packages during subsequent resets
- Share an engine between Ledger API Server and Committer
* Use a randomized H2 URL to simular in-memory
The reset service test assumes to get a completely new ledger for each
test case. But because we use H2 in-memory with db_close_delay=1 and the
same H2 database name, the second test case gets the remnants of the
first test case.
Since we know that sandbox in-memory uses an H2 in-memory URL, we can
simply use SandboxBackend.H2Database for ResetServiceInMemoryIT.
CHANGELOG_BEGIN
[Sandbox] Drastically lower the time needed to do a reset via the
ResetService.
CHANGELOG_END
* kvutils: Cache state value conversions from bytes.
This seems to have a decent speedup in ledger-on-memory.
CHANGELOG_BEGIN
- [Ledger Integration Kit] Submissions now look up ledger values from a
cache where possible, improving performance when there's contention over
certain resources (e.g. common packages). The cache size currently
defaults to 64 MB.
CHANGELOG_END
* kvutils: Make the SubmissionValidator statue value cache configurable.
* kvutils: Report state value cache metrics.
* kvutils: Add a suffix to a Long literal because WartRemover is unhappy.
Strangely, it doesn't fail on my machine.
* kvutils: Extract caching out into its own file.
* kvutils: Move the `bytesToStateValue` call into `cache.get`.
* kvutils: Move caching to its own package.
* kvutils: Inject the state value cache.
* kvutils: Default to no state value cache.
* kvutils: Accept a state value cache size in megabytes, not bytes.
* kvutils: Move cache building from `Config` to the `caching` package.
* kvutils: Replace Guava's cache with Caffeine.
* kvutils: Simplify caching configuration.
* sandbox: Enable state value caching by default.
CHANGELOG_BEGIN
- [Sandbox] State values deserialization is now cached, with a fixed
cache size of 128MB.
CHANGELOG_END
* Changelog commit.
CHANGELOG_BEGIN
- [Ledger Integration Kit] The state value cache is now opt-in, with a
default of no cache at all.
CHANGELOG_END
Packages com.digitalasset.daml and com.daml have been unified under com.daml
Ledger API and DAML-LF DEV protos have also been moved from `com/digitalasset`
to `com/daml` on the file system.
Protos for already released DAML LF versions (1.6, 1.7, 1.8) stay in the
package `com.digitalasset`.
CHANGELOG_BEGIN
[SDK] All Java and Scala packages starting with
``com.digitalasset.daml`` and ``com.digitalasset`` are now consolidated
under ``com.daml``. Simply changing imports should be enough to
migrate your code.
CHANGELOG_END
* sandbox: Move the events page size configuration value into config.
* sandbox: Pass `config` directly into JdbcIndexerFactory.
* sandbox: Reorder `eventsPageSize` before `metrics` in parameters.
* sandbox: Move `seeding` into `ApiServerConfig`.
CHANGELOG_BEGIN
CHANGELOG_END
* sandbox: Name all parameters of `JdbcLedgerDao.writeOwner`.
Co-Authored-By: stefano.baghino@digitalasset.com
Contributes to #4231.
Remove checkpoints from Participant Server storage:
- Removed code to store checkpoints in the index database.
- Remove existing checkpoint rows in ledger_entries and participant_command_completions.
- Removed ObservedTimeServiceBackend
This commit modifies the java migration V2_1__Rebuild_Acs. This is safe to do, because:
a) any even semi recent persistent sandbox had already gone through this migration and won't re-run it
b) A new database doesn't even have entries yet to migrate.
CHANGELOG_BEGIN
[DAML Ledger Integration Kit] Removed the ``Hearbeat`` state update.
[Sandbox] Checkpoints are no longer emitted in regular intervals in wall
clock time mode.
CHANGELOG_END
