![]() |
VOOZH | about |
dotnet add package Stratara.Resilience --version 3.1.4
NuGet\Install-Package Stratara.Resilience -Version 3.1.4
<PackageReference Include="Stratara.Resilience" Version="3.1.4" />
<PackageVersion Include="Stratara.Resilience" Version="3.1.4" />Directory.Packages.props
<PackageReference Include="Stratara.Resilience" />Project file
paket add Stratara.Resilience --version 3.1.4
#r "nuget: Stratara.Resilience, 3.1.4"
#:package Stratara.Resilience@3.1.4
#addin nuget:?package=Stratara.Resilience&version=3.1.4Install as a Cake Addin
#tool nuget:?package=Stratara.Resilience&version=3.1.4Install as a Cake Tool
License: (Functional Source License — source-available; converts to MIT after 2 years). Not OSI-approved OSS.
Polly-based named resilience pipelines pre-wired for Stratara's messaging + outbox dispatch paths.
services.AddResiliencePipelines();
// Resolve a specific named pipeline at the call site:
var pipeline = sp.GetRequiredService<ResiliencePipelineProvider<string>>()
.GetPipeline(ResilienceNames.CommandDispatcher);
await pipeline.ExecuteAsync(async ct => {
await DoFlakyWorkAsync(ct);
}, cancellationToken);
| Name | Constant | Strategy |
|---|---|---|
| Message bus | ResilienceNames.MessageBus |
Exponential retry up to int.MaxValue, 10s → 60s, jitter |
| Command dispatcher | ResilienceNames.CommandDispatcher |
3 retries, exponential, 200ms, jitter |
| Event bundle dispatcher | ResilienceNames.EventBundleDispatcher |
3 retries, exponential, 200ms, jitter |
| Concurrency conflict | ResilienceNames.ConcurrencyConflict |
Retry only on ConcurrencyConflictException, 5 attempts, short exponential, jitter |
The ResilienceFactory that builds these is internal — interact via DI and ResilienceNames only.
Hook a named pipeline into the in-process mediator pipeline per request — no manual ExecuteAsync plumbing:
services.AddResiliencePipelines();
services.AddStrataraResilienceBehavior(); // after AddStrataraValidation() / AddStrataraTenantIsolation()
public sealed record ReserveStock(Guid Id, int Qty) : ICommand<bool>, IResilientRequest
{
public string ResiliencePipelineName => ResilienceNames.ConcurrencyConflict;
}
Any request implementing Stratara.Abstractions.Resilience.IResilientRequest is dispatched inside the
named pipeline it selects; unmarked requests pass straight through. Only mark handlers that are safe
to re-run — naturally idempotent work, or operations guarded by optimistic concurrency (point at the
ConcurrencyConflict pipeline). Register the behavior after the validation / tenant-isolation
behaviors so a retry re-runs the handler, not the guards.
Stratara.Abstractions — for IPipelineBehavior, IResilientRequest, and ConcurrencyConflictException.Microsoft.Extensions.DependencyInjection.AbstractionsMicrosoft.Extensions.Resilience (Polly).| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 net10.0 is compatible. net10.0-android net10.0-android was computed. net10.0-browser net10.0-browser was computed. net10.0-ios net10.0-ios was computed. net10.0-maccatalyst net10.0-maccatalyst was computed. net10.0-macos net10.0-macos was computed. net10.0-tvos net10.0-tvos was computed. net10.0-windows net10.0-windows was computed. |
Showing the top 1 NuGet packages that depend on Stratara.Resilience:
| Package | Downloads |
|---|---|
|
Stratara.Shared
Umbrella package for the Stratara framework — source-generated logger extensions for outbox / saga / projection / messaging, domain-event helpers, merge primitives, and the Tier-A/B abstractions surface re-exported as one transitive bundle. |
This package is not used by any popular GitHub repositories.
### Added
- **Command-workload isolation (heavy-command lane)** — long-running commands can now be routed to a
dedicated worker lane so they cannot starve interactive commands. Mark a command with the new
`Stratara.Abstractions.Mediator.IHeavyCommand` marker and the `ICommandOutboxDispatcher`
automatically publishes it to a separate heavy-command topic (`IMessagingIdentifier.HeavyCommandTopic` /
`HeavyCommandSubscription`, configurable under `Messaging:HeavyCommand`, defaulting to `heavy-command` /
`heavy-command-subscription`). Run a dedicated heavy-command worker with the new
`services.AddHeavyCommandWorker(degreeOfParallelism?)` extension, or the
`builder.AddHeavyCommandWorkerServices(degreeOfParallelism?)` host composite — in the same process as
the interactive worker (two lanes) or in a separately scaled host. Each worker's degree of parallelism
is configurable per lane. `IMessagingIdentifier` gains `HeavyCommandTopic`, `HeavyCommandSubscription`,
and the `GetCommandTopic(Type)` / `GetCommandSubscription(Type)` routing helpers. The interactive lane
(`AddMediatorWorker()`) is unchanged and remains the default; commands not marked heavy keep their
existing routing. If a heavy command is dispatched while no heavy worker is bound, the publish is
rejected and the command is preserved in the outbox until a heavy-command worker comes online — it is
never dropped. Works over both the RabbitMQ and Azure Service Bus message buses (Azure Service Bus
requires the heavy-command topic/subscription to be provisioned, like the existing command topic). New
log-event ID `105_005` (`CommandWorkerLaneStarted`) in `Stratara.Diagnostics`.
- **Observability metrics across the worker pipeline** (`Stratara.Diagnostics`) — the shared
`Stratara.Service` meter now publishes throughput and latency instruments so operators can see how the
event-sourcing pipeline is behaving instead of flying blind on a single counter. New instruments:
`event_source.events.appended` (counter, tagged by `event.type` / `aggregate.type`),
`outbox.published` (counter, tagged by `outbox.kind` = `command` / `event`), `command.duration`
(histogram, ms, tagged by `request.type` / `outcome`), `projection.events.processed` (counter) +
`projection.bundle.duration` (histogram, ms), `saga.events.processed` (counter) +
`saga.bundle.duration` (histogram, ms), and `saga.inflight` (up/down counter). They are recorded by the
event source, command worker, projection worker, saga worker, and outbox worker respectively. Because
projections and sagas are real-time bus subscribers without a persisted checkpoint, these report
**throughput and latency**, not consumer lag. No configuration is required — point any OpenTelemetry
metrics exporter at the `Stratara.Service` meter.
- **Operational health checks for the event store and outbox** (`Stratara.EventSourcing.EntityFrameworkCore`) —
two opt-in readiness checks added to any `IHealthChecksBuilder`: `AddEventStoreHealthCheck()` verifies
the write-side database is reachable, and `AddOutboxHealthCheck(degradedThreshold?, unhealthyThreshold?)`
reports the pending outbox backlog (exposed under the `pending` data key) and escalates to
`Degraded` / `Unhealthy` when the backlog crosses the supplied thresholds. Both are tagged `ready` by
default (so they map to a readiness endpoint, not liveness) and require the Stratara write store to be
registered. The write-store DbContext is now also resolvable as a scoped `IWriteDbContext` service to
support these checks.
- **Polly-backed mediator resilience behavior** (`Stratara.Resilience`) — an opt-in pipeline behavior
wraps the in-process dispatch of a request marked with the new
`Stratara.Abstractions.Resilience.IResilientRequest` in the named Polly pipeline the request selects
(`ResiliencePipelineName`). Register it with the new `AddStrataraResilienceBehavior()` (after
`AddStrataraValidation()` / `AddStrataraTenantIsolation()` so the retry wraps the handler, not the
guards); requests without the marker are unaffected. A new built-in pipeline
`ResilienceNames.ConcurrencyConflict` retries **only** on
`Stratara.Abstractions.Persistence.ConcurrencyConflictException` (5 attempts, short exponential
backoff) so a handler that re-reads and re-applies on an optimistic-concurrency clash succeeds without
bespoke retry loops; it is registered by `AddResiliencePipelines()` alongside the existing message-bus
and dispatcher pipelines. Only mark handlers that are safe to re-run (idempotent or concurrency-guarded).