diff --git a/daprdocs/content/en/concepts/dapr-services/placement.md b/daprdocs/content/en/concepts/dapr-services/placement.md index d94f9a8435d..c6d739957f5 100644 --- a/daprdocs/content/en/concepts/dapr-services/placement.md +++ b/daprdocs/content/en/concepts/dapr-services/placement.md @@ -13,7 +13,9 @@ The Placement service Docker container is started automatically as part of [`dap ## Kubernetes mode -The Placement service is deployed as part of `dapr init -k`, or via the Dapr Helm charts. For more information on running Dapr on Kubernetes, visit the [Kubernetes hosting page]({{< ref kubernetes >}}). +The Placement service is deployed as part of `dapr init -k`, or via the Dapr Helm charts. You can run Placement in high availability (HA) mode. [Learn more about setting HA mode in your Kubernetes service.]({{< ref "kubernetes-production.md#individual-service-ha-helm-configuration" >}}) + +For more information on running Dapr on Kubernetes, visit the [Kubernetes hosting page]({{< ref kubernetes >}}). ## Placement tables diff --git a/daprdocs/content/en/concepts/dapr-services/scheduler.md b/daprdocs/content/en/concepts/dapr-services/scheduler.md index 29060fe938e..2fba4ba713a 100644 --- a/daprdocs/content/en/concepts/dapr-services/scheduler.md +++ b/daprdocs/content/en/concepts/dapr-services/scheduler.md @@ -11,13 +11,21 @@ The diagram below shows how the Scheduler service is used via the jobs API when Diagram showing the Scheduler control plane service and the jobs API +## Actor reminders + +Prior to Dapr v1.15, [actor reminders]({{< ref "actors-timers-reminders.md#actor-reminders" >}}) were run using the Placement service. Now, by default, the [`SchedulerReminders` feature flag]({{< ref "support-preview-features.md#current-preview-features" >}}) is set to `true`, and all new actor reminders you create are run using the Scheduler service to make them more scalable. + +When you deploy Dapr v1.15, any _existing_ actor reminders are migrated from the Placement service to the Scheduler service as a one time operation for each actor type. You can prevent this migration by setting the `SchedulerReminders` flag to `false` in application configuration file for the actor type. + ## Self-hosted mode The Scheduler service Docker container is started automatically as part of `dapr init`. It can also be run manually as a process if you are running in [slim-init mode]({{< ref self-hosted-no-docker.md >}}). ## Kubernetes mode -The Scheduler service is deployed as part of `dapr init -k`, or via the Dapr Helm charts. For more information on running Dapr on Kubernetes, visit the [Kubernetes hosting page]({{< ref kubernetes >}}). +The Scheduler service is deployed as part of `dapr init -k`, or via the Dapr Helm charts. You can run Scheduler in high availability (HA) mode. [Learn more about setting HA mode in your Kubernetes service.]({{< ref "kubernetes-production.md#individual-service-ha-helm-configuration" >}}) + +For more information on running Dapr on Kubernetes, visit the [Kubernetes hosting page]({{< ref kubernetes >}}). ## Related links diff --git a/daprdocs/content/en/developing-applications/building-blocks/actors/actors-timers-reminders.md b/daprdocs/content/en/developing-applications/building-blocks/actors/actors-timers-reminders.md index 7a4cd1ec74c..8664045632c 100644 --- a/daprdocs/content/en/developing-applications/building-blocks/actors/actors-timers-reminders.md +++ b/daprdocs/content/en/developing-applications/building-blocks/actors/actors-timers-reminders.md @@ -107,6 +107,10 @@ Refer [api spec]({{< ref "actors_api.md#invoke-timer" >}}) for more details. ## Actor reminders +{{% alert title="Note" color="primary" %}} +In Dapr v1.15, actor reminders are stored by default in the [Scheduler service]({{< ref "scheduler.md#actor-reminders" >}}). +{{% /alert %}} + Reminders are a mechanism to trigger *persistent* callbacks on an actor at specified times. Their functionality is similar to timers. But unlike timers, reminders are triggered under all circumstances until the actor explicitly unregisters them or the actor is explicitly deleted or the number in invocations is exhausted. Specifically, reminders are triggered across actor deactivations and failovers because the Dapr actor runtime persists the information about the actors' reminders using Dapr actor state provider. You can create a persistent reminder for an actor by calling the HTTP/gRPC request to Dapr as shown below, or via Dapr SDK. @@ -148,7 +152,9 @@ If an invocation of the method fails, the timer is not removed. Timers are only ## Reminder data serialization format -Actor reminder data is serialized to JSON by default. Dapr v1.13 onwards supports a protobuf serialization format for reminders data which, depending on throughput and size of the payload, can result in significant performance improvements, giving developers a higher throughput and lower latency. Another benefit is storing smaller data in the actor underlying database, which can result in cost optimizations when using some cloud databases. A restriction with using protobuf serialization is that the reminder data can no longer be queried. +Actor reminder data is serialized to JSON by default. Dapr v1.13 onwards supports a protobuf serialization format for internal reminders data for workflow via both the Placement and Scheduler services. Depending on throughput and size of the payload, this can result in significant performance improvements, giving developers a higher throughput and lower latency. + +Another benefit is storing smaller data in the actor underlying database, which can result in cost optimizations when using some cloud databases. A restriction with using protobuf serialization is that the reminder data can no longer be queried. {{% alert title="Note" color="primary" %}} Protobuf serialization will become the default format in Dapr 1.14 diff --git a/daprdocs/content/en/developing-applications/building-blocks/jobs/jobs-overview.md b/daprdocs/content/en/developing-applications/building-blocks/jobs/jobs-overview.md index 486cfc5d623..63f90c102f6 100644 --- a/daprdocs/content/en/developing-applications/building-blocks/jobs/jobs-overview.md +++ b/daprdocs/content/en/developing-applications/building-blocks/jobs/jobs-overview.md @@ -59,10 +59,6 @@ The jobs API provides several features to make it easy for you to schedule jobs. The Scheduler service enables the scheduling of jobs to scale across multiple replicas, while guaranteeing that a job is only triggered by 1 scheduler service instance. -### Actor reminders - -Actors have actor reminders, but present some limitations involving scalability using the Placement service implementation. You can make reminders more scalable by using [`SchedulerReminders`]({{< ref support-preview-features.md >}}). This is set in the configuration for your actor application. - ## Try out the jobs API You can try out the jobs API in your application. After [Dapr is installed]({{< ref install-dapr-cli.md >}}), you can begin using the jobs API, starting with [the How-to: Schedule jobs guide]({{< ref howto-schedule-and-handle-triggered-jobs.md >}}). diff --git a/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-deploy.md b/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-deploy.md index 658d1475e5a..41af7c0d84a 100644 --- a/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-deploy.md +++ b/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-deploy.md @@ -231,6 +231,19 @@ You can install Dapr on Kubernetes using a Helm v3 chart. --wait ``` + To install in **high availability** mode and scale select services independently of global: + + ```bash + helm upgrade --install dapr dapr/dapr \ + --version={{% dapr-latest-version short="true" %}} \ + --namespace dapr-system \ + --create-namespace \ + --set global.ha.enabled=false \ + --set dapr_scheduler.ha=true \ + --set dapr_placement.ha=true \ + --wait + ``` + See [Guidelines for production ready deployments on Kubernetes]({{< ref kubernetes-production.md >}}) for more information on installing and upgrading Dapr using Helm. ### (optional) Install the Dapr dashboard as part of the control plane diff --git a/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-production.md b/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-production.md index 8e5ea993453..1151137efab 100644 --- a/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-production.md +++ b/daprdocs/content/en/operations/hosting/kubernetes/kubernetes-production.md @@ -95,6 +95,25 @@ For a new Dapr deployment, HA mode can be set with both: For an existing Dapr deployment, [you can enable HA mode in a few extra steps]({{< ref "#enabling-high-availability-in-an-existing-dapr-deployment" >}}). +### Individual service HA Helm configuration + +You can configure HA mode via Helm across all services by setting the `global.ha.enabled` flag to `true`. By default, `--set global.ha.enabled=true` is fully respected and cannot be overridden, making it impossible to simultaneously have either the placement or scheduler service as a single instance. + +> **Note:** HA for scheduler and placement services is not the default setting. + +To scale scheduler and placement to three instances independently of the `global.ha.enabled` flag, set `global.ha.enabled` to `false` and `dapr_scheduler.ha` and `dapr_placement.ha` to `true`. For example: + + ```bash + helm upgrade --install dapr dapr/dapr \ + --version={{% dapr-latest-version short="true" %}} \ + --namespace dapr-system \ + --create-namespace \ + --set global.ha.enabled=false \ + --set dapr_scheduler.ha=true \ + --set dapr_placement.ha=true \ + --wait + ``` + ## Setting cluster critical priority class name for control plane services In some scenarios, nodes may have memory and/or cpu pressure and the Dapr control plane pods might get selected diff --git a/daprdocs/content/en/operations/support/support-preview-features.md b/daprdocs/content/en/operations/support/support-preview-features.md index 221b24d8466..07ae1b9a679 100644 --- a/daprdocs/content/en/operations/support/support-preview-features.md +++ b/daprdocs/content/en/operations/support/support-preview-features.md @@ -22,4 +22,4 @@ For CLI there is no explicit opt-in, just the version that this was first made a | **Actor State TTL** | Allow actors to save records to state stores with Time To Live (TTL) set to automatically clean up old data. In its current implementation, actor state with TTL may not be reflected correctly by clients, read [Actor State Transactions]({{< ref actors_api.md >}}) for more information. | `ActorStateTTL` | [Actor State Transactions]({{< ref actors_api.md >}}) | v1.11 | | **Component Hot Reloading** | Allows for Dapr-loaded components to be "hot reloaded". A component spec is reloaded when it is created/updated/deleted in Kubernetes or on file when running in self-hosted mode. Ignores changes to actor state stores and workflow backends. | `HotReload`| [Hot Reloading]({{< ref components-concept.md >}}) | v1.13 | | **Subscription Hot Reloading** | Allows for declarative subscriptions to be "hot reloaded". A subscription is reloaded either when it is created/updated/deleted in Kubernetes, or on file in self-hosted mode. In-flight messages are unaffected when reloading. | `HotReload`| [Hot Reloading]({{< ref "subscription-methods.md#declarative-subscriptions" >}}) | v1.14 | -| **Scheduler Actor Reminders** | Whilst the [Scheduler service]({{< ref "concepts/dapr-services/scheduler.md" >}}) is deployed by default, Scheduler actor reminders (actor reminders stored in the Scheduler control plane service as opposed to the Placement control plane service actor reminder system) are enabled through a preview feature and needs a feature flag. | `SchedulerReminders`| [Scheduler actor reminders]({{< ref "jobs-overview.md#actor-reminders" >}}) | v1.14 | +| **Scheduler Actor Reminders** | Scheduler actor reminders are actor reminders stored in the Scheduler control plane service, as opposed to the Placement control plane service actor reminder system. The `SchedulerReminders` preview feature defaults to `true`, but you can disable Scheduler actor reminders by setting it to `false`. | `SchedulerReminders`| [Scheduler actor reminders]({{< ref "scheduler.md#actor-reminders" >}}) | v1.14 | \ No newline at end of file diff --git a/daprdocs/static/images/scheduler/scheduler-architecture.png b/daprdocs/static/images/scheduler/scheduler-architecture.png index 5cf309bf4a6..1b87d1ffd08 100644 Binary files a/daprdocs/static/images/scheduler/scheduler-architecture.png and b/daprdocs/static/images/scheduler/scheduler-architecture.png differ