pulumi/pkg/resource/deploy/deployment.go

385 lines
16 KiB
Go
Raw Normal View History

2018-05-22 21:43:36 +02:00
// Copyright 2016-2018, Pulumi Corporation.
//
// 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.
Begin resource modeling and planning This change introduces a new package, pkg/resource, that will form the foundation for actually performing deployment plans and applications. It contains the following key abstractions: * resource.Provider is a wrapper around the CRUD operations exposed by underlying resource plugins. It will eventually defer to resource.Plugin, which itself defers -- over an RPC interface -- to the actual plugin, one per package exposing resources. The provider will also understand how to load, cache, and overall manage the lifetime of each plugin. * resource.Resource is the actual resource object. This is created from the overall evaluation object graph, but is simplified. It contains only serializable properties, for example. Inter-resource references are translated into serializable monikers as part of creating the resource. * resource.Moniker is a serializable string that uniquely identifies a resource in the Mu system. This is in contrast to resource IDs, which are generated by resource providers and generally opaque to the Mu system. See marapongo/mu#69 for more information about monikers and some of their challenges (namely, designing a stable algorithm). * resource.Snapshot is a "snapshot" taken from a graph of resources. This is a transitive closure of state representing one possible configuration of a given environment. This is what plans are created from. Eventually, two snapshots will be diffable, in order to perform incremental updates. One way of thinking about this is that a snapshot of the old world's state is advanced, one step at a time, until it reaches a desired snapshot of the new world's state. * resource.Plan is a plan for carrying out desired CRUD operations on a target environment. Each plan consists of zero-to-many Steps, each of which has a CRUD operation type, a resource target, and a next step. This is an enumerator because it is possible the plan will evolve -- and introduce new steps -- as it is carried out (hence, the Next() method). At the moment, this is linearized; eventually, we want to make this more "graph-like" so that we can exploit available parallelism within the dependencies. There are tons of TODOs remaining. However, the `mu plan` command is functioning with these new changes -- including colorization FTW -- so I'm landing it now. This is part of marapongo/mu#38 and marapongo/mu#41.
2017-02-17 21:31:48 +01:00
package deploy
Begin resource modeling and planning This change introduces a new package, pkg/resource, that will form the foundation for actually performing deployment plans and applications. It contains the following key abstractions: * resource.Provider is a wrapper around the CRUD operations exposed by underlying resource plugins. It will eventually defer to resource.Plugin, which itself defers -- over an RPC interface -- to the actual plugin, one per package exposing resources. The provider will also understand how to load, cache, and overall manage the lifetime of each plugin. * resource.Resource is the actual resource object. This is created from the overall evaluation object graph, but is simplified. It contains only serializable properties, for example. Inter-resource references are translated into serializable monikers as part of creating the resource. * resource.Moniker is a serializable string that uniquely identifies a resource in the Mu system. This is in contrast to resource IDs, which are generated by resource providers and generally opaque to the Mu system. See marapongo/mu#69 for more information about monikers and some of their challenges (namely, designing a stable algorithm). * resource.Snapshot is a "snapshot" taken from a graph of resources. This is a transitive closure of state representing one possible configuration of a given environment. This is what plans are created from. Eventually, two snapshots will be diffable, in order to perform incremental updates. One way of thinking about this is that a snapshot of the old world's state is advanced, one step at a time, until it reaches a desired snapshot of the new world's state. * resource.Plan is a plan for carrying out desired CRUD operations on a target environment. Each plan consists of zero-to-many Steps, each of which has a CRUD operation type, a resource target, and a next step. This is an enumerator because it is possible the plan will evolve -- and introduce new steps -- as it is carried out (hence, the Next() method). At the moment, this is linearized; eventually, we want to make this more "graph-like" so that we can exploit available parallelism within the dependencies. There are tons of TODOs remaining. However, the `mu plan` command is functioning with these new changes -- including colorization FTW -- so I'm landing it now. This is part of marapongo/mu#38 and marapongo/mu#41.
2017-02-17 21:31:48 +01:00
import (
"context"
"math"
"sync"
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
"github.com/blang/semver"
uuid "github.com/gofrs/uuid"
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
"github.com/pkg/errors"
"github.com/pulumi/pulumi/pkg/v2/codegen/schema"
"github.com/pulumi/pulumi/pkg/v2/resource/deploy/providers"
"github.com/pulumi/pulumi/pkg/v2/resource/graph"
"github.com/pulumi/pulumi/sdk/v2/go/common/diag"
"github.com/pulumi/pulumi/sdk/v2/go/common/resource"
"github.com/pulumi/pulumi/sdk/v2/go/common/resource/plugin"
"github.com/pulumi/pulumi/sdk/v2/go/common/tokens"
"github.com/pulumi/pulumi/sdk/v2/go/common/util/contract"
"github.com/pulumi/pulumi/sdk/v2/go/common/util/result"
Begin resource modeling and planning This change introduces a new package, pkg/resource, that will form the foundation for actually performing deployment plans and applications. It contains the following key abstractions: * resource.Provider is a wrapper around the CRUD operations exposed by underlying resource plugins. It will eventually defer to resource.Plugin, which itself defers -- over an RPC interface -- to the actual plugin, one per package exposing resources. The provider will also understand how to load, cache, and overall manage the lifetime of each plugin. * resource.Resource is the actual resource object. This is created from the overall evaluation object graph, but is simplified. It contains only serializable properties, for example. Inter-resource references are translated into serializable monikers as part of creating the resource. * resource.Moniker is a serializable string that uniquely identifies a resource in the Mu system. This is in contrast to resource IDs, which are generated by resource providers and generally opaque to the Mu system. See marapongo/mu#69 for more information about monikers and some of their challenges (namely, designing a stable algorithm). * resource.Snapshot is a "snapshot" taken from a graph of resources. This is a transitive closure of state representing one possible configuration of a given environment. This is what plans are created from. Eventually, two snapshots will be diffable, in order to perform incremental updates. One way of thinking about this is that a snapshot of the old world's state is advanced, one step at a time, until it reaches a desired snapshot of the new world's state. * resource.Plan is a plan for carrying out desired CRUD operations on a target environment. Each plan consists of zero-to-many Steps, each of which has a CRUD operation type, a resource target, and a next step. This is an enumerator because it is possible the plan will evolve -- and introduce new steps -- as it is carried out (hence, the Next() method). At the moment, this is linearized; eventually, we want to make this more "graph-like" so that we can exploit available parallelism within the dependencies. There are tons of TODOs remaining. However, the `mu plan` command is functioning with these new changes -- including colorization FTW -- so I'm landing it now. This is part of marapongo/mu#38 and marapongo/mu#41.
2017-02-17 21:31:48 +01:00
)
// BackendClient provides an interface for retrieving information about other stacks.
type BackendClient interface {
// GetStackOutputs returns the outputs (if any) for the named stack or an error if the stack cannot be found.
GetStackOutputs(ctx context.Context, name string) (resource.PropertyMap, error)
// GetStackResourceOutputs returns the resource outputs for a stack, or an error if the stack
// cannot be found. Resources are retrieved from the latest stack snapshot, which may include
// ongoing updates. They are returned in a `PropertyMap` mapping resource URN to another
// `Propertymap` with members `type` (containing the Pulumi type ID for the resource) and
// `outputs` (containing the resource outputs themselves).
GetStackResourceOutputs(ctx context.Context, stackName string) (resource.PropertyMap, error)
}
// Options controls the deployment process.
type Options struct {
Events Events // an optional events callback interface.
Parallel int // the degree of parallelism for resource operations (<=1 for serial).
Refresh bool // whether or not to refresh before executing the deployment.
RefreshOnly bool // whether or not to exit after refreshing.
RefreshTargets []resource.URN // The specific resources to refresh during a refresh op.
ReplaceTargets []resource.URN // Specific resources to replace.
2019-09-20 04:28:14 +02:00
DestroyTargets []resource.URN // Specific resources to destroy.
UpdateTargets []resource.URN // Specific resources to update.
TargetDependents bool // true if we're allowing things to proceed, even with unspecified targets
TrustDependencies bool // whether or not to trust the resource dependency graph.
UseLegacyDiff bool // whether or not to use legacy diffing behavior.
}
// DegreeOfParallelism returns the degree of parallelism that should be used during the
// deployment process.
func (o Options) DegreeOfParallelism() int {
if o.Parallel <= 1 {
return 1
}
return o.Parallel
}
// InfiniteParallelism returns whether or not the requested level of parallelism is unbounded.
func (o Options) InfiniteParallelism() bool {
return o.Parallel == math.MaxInt32
}
2019-06-11 00:20:44 +02:00
// StepExecutorEvents is an interface that can be used to hook resource lifecycle events.
type StepExecutorEvents interface {
OnResourceStepPre(step Step) (interface{}, error)
OnResourceStepPost(ctx interface{}, step Step, status resource.Status, err error) error
OnResourceOutputs(step Step) error
}
2019-06-11 00:20:44 +02:00
// PolicyEvents is an interface that can be used to hook policy violation events.
type PolicyEvents interface {
OnPolicyViolation(resource.URN, plugin.AnalyzeDiagnostic)
}
// Events is an interface that can be used to hook interesting engine events.
2019-06-11 00:20:44 +02:00
type Events interface {
StepExecutorEvents
PolicyEvents
}
Add a list of in-flight operations to the deployment (#1759) * Add a list of in-flight operations to the deployment This commit augments 'DeploymentV2' with a list of operations that are currently in flight. This information is used by the engine to keep track of whether or not a particular deployment is in a valid state. The SnapshotManager is responsible for inserting and removing operations from the in-flight operation list. When the engine registers an intent to perform an operation, SnapshotManager inserts an Operation into this list and saves it to the snapshot. When an operation completes, the SnapshotManager removes it from the snapshot. From this, the engine can infer that if it ever sees a deployment with pending operations, the Pulumi CLI must have crashed or otherwise abnormally terminated before seeing whether or not an operation completed successfully. To remedy this state, this commit also adds code to 'pulumi stack import' that clears all pending operations from a deployment, as well as code to plan generation that will reject any deployments that have pending operations present. At the CLI level, if we see that we are in a state where pending operations were in-flight when the engine died, we'll issue a human-friendly error message that indicates which resources are in a bad state and how to recover their stack. * CR: Multi-line string literals, renaming in-flight -> pending * CR: Add enum to apitype for operation type, also name status -> type for clarity * Fix the yaml type * Fix missed renames * Add implementation for lifecycle_test.go * Rebase against master
2018-08-11 06:39:59 +02:00
// PlanPendingOperationsError is an error returned from `NewPlan` if there exist pending operations in the
// snapshot that we are preparing to operate upon. The engine does not allow any operations to be pending
// when operating on a snapshot.
type PlanPendingOperationsError struct {
Operations []resource.Operation
}
func (p PlanPendingOperationsError) Error() string {
return "one or more operations are currently pending"
}
type resourceMap struct {
m sync.Map
}
func (m *resourceMap) set(urn resource.URN, state *resource.State) {
m.m.Store(urn, state)
}
func (m *resourceMap) get(urn resource.URN) (*resource.State, bool) {
s, ok := m.m.Load(urn)
if !ok {
return nil, false
}
return s.(*resource.State), true
}
func (m *resourceMap) mapRange(callback func(urn resource.URN, state *resource.State) bool) {
m.m.Range(func(k, v interface{}) bool {
return callback(k.(resource.URN), v.(*resource.State))
})
}
// A Deployment manages the iterative computation and execution of a deployment based on a stream of goal states.
// A running deployment emits events that indicate its progress. These events must be used to record the new state
// of the deployment target.
type Deployment struct {
ctx *plugin.Context // the plugin context (for provider operations).
target *Target // the deployment target.
prev *Snapshot // the old resource snapshot for comparison.
olds map[resource.URN]*resource.State // a map of all old resources.
imports []Import // resources to import, if this is an import deployment.
isImport bool // true if this is an import deployment.
schemaLoader schema.Loader // the schema cache for this deployment, if any.
source Source // the source of new resources.
localPolicyPackPaths []string // the policy packs to run during this deployment's generation.
preview bool // true if this deployment is to be previewed.
depGraph *graph.DependencyGraph // the dependency graph of the old snapshot
providers *providers.Registry // the provider registry for this deployment.
news *resourceMap // the set of new resources generated by the deployment
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}
// addDefaultProviders adds any necessary default provider definitions and references to the given snapshot. Version
// information for these providers is sourced from the snapshot's manifest; inputs parameters are sourced from the
// stack's configuration.
func addDefaultProviders(target *Target, source Source, prev *Snapshot) error {
if prev == nil {
return nil
}
// Pull the versions we'll use for default providers from the snapshot's manifest.
defaultProviderVersions := make(map[tokens.Package]*semver.Version)
for _, p := range prev.Manifest.Plugins {
defaultProviderVersions[tokens.Package(p.Name)] = p.Version
}
// Determine the necessary set of default providers and inject references to default providers as appropriate.
//
// We do this by scraping the snapshot for custom resources that does not reference a provider and adding
// default providers for these resources' packages. Each of these resources is rewritten to reference the default
// provider for its package.
//
// The configuration for each default provider is pulled from the stack's configuration information.
var defaultProviders []*resource.State
defaultProviderRefs := make(map[tokens.Package]providers.Reference)
for _, res := range prev.Resources {
if providers.IsProviderType(res.URN.Type()) || !res.Custom || res.Provider != "" {
continue
}
pkg := res.URN.Type().Package()
ref, ok := defaultProviderRefs[pkg]
if !ok {
inputs, err := target.GetPackageConfig(pkg)
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
if err != nil {
return errors.Errorf("could not fetch configuration for default provider '%v'", pkg)
}
if version, ok := defaultProviderVersions[pkg]; ok {
inputs["version"] = resource.NewStringProperty(version.String())
}
uuid, err := uuid.NewV4()
if err != nil {
return err
}
urn, id := defaultProviderURN(target, source, pkg), resource.ID(uuid.String())
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
ref, err = providers.NewReference(urn, id)
contract.Assert(err == nil)
provider := &resource.State{
Type: urn.Type(),
URN: urn,
Custom: true,
ID: id,
Inputs: inputs,
Outputs: inputs,
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}
defaultProviders = append(defaultProviders, provider)
defaultProviderRefs[pkg] = ref
}
res.Provider = ref.String()
}
// If any default providers are necessary, prepend their definitions to the snapshot's resources. This trivially
// guarantees that all default provider references name providers that precede the referent in the snapshot.
if len(defaultProviders) != 0 {
prev.Resources = append(defaultProviders, prev.Resources...)
}
return nil
}
// migrateProviders is responsible for adding default providers to old snapshots and filling in output properties for
// providers that do not have them.
func migrateProviders(target *Target, prev *Snapshot, source Source) error {
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
// Add any necessary default provider references to the previous snapshot in order to accommodate stacks that were
// created prior to the changes that added first-class providers. We do this here rather than in the migration
// package s.t. the inputs to any default providers (which we fetch from the stacks's configuration) are as
// accurate as possible.
if err := addDefaultProviders(target, source, prev); err != nil {
return err
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}
// Migrate provider resources from the old, output-less format to the new format where all inputs are reflected as
// outputs.
if prev != nil {
for _, res := range prev.Resources {
// If we have no old outputs for a provider, use its old inputs as its old outputs. This handles the
// scenario where the CLI is being upgraded from a version that did not reflect provider inputs to
// provider outputs, and a provider is being upgraded from a version that did not implement DiffConfig to
// a version that does.
if providers.IsProviderType(res.URN.Type()) && len(res.Inputs) != 0 && len(res.Outputs) == 0 {
res.Outputs = res.Inputs
}
}
}
return nil
}
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
func buildResourceMap(prev *Snapshot, preview bool) ([]*resource.State, map[resource.URN]*resource.State, error) {
olds := make(map[resource.URN]*resource.State)
if prev == nil {
return nil, olds, nil
}
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
if prev.PendingOperations != nil && !preview {
return nil, nil, PlanPendingOperationsError{prev.PendingOperations}
}
for _, oldres := range prev.Resources {
// Ignore resources that are pending deletion; these should not be recorded in the LUT.
if oldres.Delete {
continue
}
urn := oldres.URN
if olds[urn] != nil {
return nil, nil, errors.Errorf("unexpected duplicate resource '%s'", urn)
}
olds[urn] = oldres
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}
return prev.Resources, olds, nil
}
// NewDeployment creates a new deployment from a resource snapshot plus a package to evaluate.
//
// From the old and new states, it understands how to orchestrate an evaluation and analyze the resulting resources.
// The deployment may be used to simply inspect a series of operations, or actually perform them; these operations are
// generated based on analysis of the old and new states. If a resource exists in new, but not old, for example, it
// results in a create; if it exists in both, but is different, it results in an update; and so on and so forth.
//
// Note that a deployment uses internal concurrency and parallelism in various ways, so it must be closed if for some
// reason it isn't carried out to its final conclusion. This will result in cancellation and reclamation of resources.
func NewDeployment(ctx *plugin.Context, target *Target, prev *Snapshot, source Source,
localPolicyPackPaths []string, preview bool, backendClient BackendClient) (*Deployment, error) {
contract.Assert(ctx != nil)
contract.Assert(target != nil)
contract.Assert(source != nil)
if err := migrateProviders(target, prev, source); err != nil {
return nil, err
}
// Produce a map of all old resources for fast access.
//
// NOTE: we can and do mutate prev.Resources, olds, and depGraph during execution after performing a refresh. See
// deploymentExecutor.refresh for details.
oldResources, olds, err := buildResourceMap(prev, preview)
if err != nil {
return nil, err
}
// Build the dependency graph for the old resources.
depGraph := graph.NewDependencyGraph(oldResources)
// Create a resource map for the deployment.
newResources := &resourceMap{}
// Create a new builtin provider. This provider implements features such as `getStack`.
builtins := newBuiltinProvider(backendClient, newResources)
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
// Create a new provider registry. Although we really only need to pass in any providers that were present in the
// old resource list, the registry itself will filter out other sorts of resources when processing the prior state,
// so we just pass all of the old resources.
reg, err := providers.NewRegistry(ctx.Host, oldResources, preview, builtins)
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
if err != nil {
return nil, err
}
return &Deployment{
ctx: ctx,
target: target,
prev: prev,
olds: olds,
source: source,
localPolicyPackPaths: localPolicyPackPaths,
preview: preview,
depGraph: depGraph,
providers: reg,
news: newResources,
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}, nil
}
func (d *Deployment) Ctx() *plugin.Context { return d.ctx }
func (d *Deployment) Target() *Target { return d.target }
func (d *Deployment) Diag() diag.Sink { return d.ctx.Diag }
func (d *Deployment) Prev() *Snapshot { return d.prev }
func (d *Deployment) Olds() map[resource.URN]*resource.State { return d.olds }
func (d *Deployment) Source() Source { return d.source }
Begin resource modeling and planning This change introduces a new package, pkg/resource, that will form the foundation for actually performing deployment plans and applications. It contains the following key abstractions: * resource.Provider is a wrapper around the CRUD operations exposed by underlying resource plugins. It will eventually defer to resource.Plugin, which itself defers -- over an RPC interface -- to the actual plugin, one per package exposing resources. The provider will also understand how to load, cache, and overall manage the lifetime of each plugin. * resource.Resource is the actual resource object. This is created from the overall evaluation object graph, but is simplified. It contains only serializable properties, for example. Inter-resource references are translated into serializable monikers as part of creating the resource. * resource.Moniker is a serializable string that uniquely identifies a resource in the Mu system. This is in contrast to resource IDs, which are generated by resource providers and generally opaque to the Mu system. See marapongo/mu#69 for more information about monikers and some of their challenges (namely, designing a stable algorithm). * resource.Snapshot is a "snapshot" taken from a graph of resources. This is a transitive closure of state representing one possible configuration of a given environment. This is what plans are created from. Eventually, two snapshots will be diffable, in order to perform incremental updates. One way of thinking about this is that a snapshot of the old world's state is advanced, one step at a time, until it reaches a desired snapshot of the new world's state. * resource.Plan is a plan for carrying out desired CRUD operations on a target environment. Each plan consists of zero-to-many Steps, each of which has a CRUD operation type, a resource target, and a next step. This is an enumerator because it is possible the plan will evolve -- and introduce new steps -- as it is carried out (hence, the Next() method). At the moment, this is linearized; eventually, we want to make this more "graph-like" so that we can exploit available parallelism within the dependencies. There are tons of TODOs remaining. However, the `mu plan` command is functioning with these new changes -- including colorization FTW -- so I'm landing it now. This is part of marapongo/mu#38 and marapongo/mu#41.
2017-02-17 21:31:48 +01:00
func (d *Deployment) GetProvider(ref providers.Reference) (plugin.Provider, bool) {
return d.providers.GetProvider(ref)
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}
// generateURN generates a resource's URN from its parent, type, and name under the scope of the deployment's stack and
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
// project.
func (d *Deployment) generateURN(parent resource.URN, ty tokens.Type, name tokens.QName) resource.URN {
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
// Use the resource goal state name to produce a globally unique URN.
parentType := tokens.Type("")
if parent != "" && parent.Type() != resource.RootStackType {
// Skip empty parents and don't use the root stack type; otherwise, use the full qualified type.
parentType = parent.QualifiedType()
}
return resource.NewURN(d.Target().Name, d.source.Project(), parentType, ty, name)
Implement first-class providers. (#1695) ### First-Class Providers These changes implement support for first-class providers. First-class providers are provider plugins that are exposed as resources via the Pulumi programming model so that they may be explicitly and multiply instantiated. Each instance of a provider resource may be configured differently, and configuration parameters may be source from the outputs of other resources. ### Provider Plugin Changes In order to accommodate the need to verify and diff provider configuration and configure providers without complete configuration information, these changes adjust the high-level provider plugin interface. Two new methods for validating a provider's configuration and diffing changes to the same have been added (`CheckConfig` and `DiffConfig`, respectively), and the type of the configuration bag accepted by `Configure` has been changed to a `PropertyMap`. These changes have not yet been reflected in the provider plugin gRPC interface. We will do this in a set of follow-up changes. Until then, these methods are implemented by adapters: - `CheckConfig` validates that all configuration parameters are string or unknown properties. This is necessary because existing plugins only accept string-typed configuration values. - `DiffConfig` either returns "never replace" if all configuration values are known or "must replace" if any configuration value is unknown. The justification for this behavior is given [here](https://github.com/pulumi/pulumi/pull/1695/files#diff-a6cd5c7f337665f5bb22e92ca5f07537R106) - `Configure` converts the config bag to a legacy config map and configures the provider plugin if all config values are known. If any config value is unknown, the underlying plugin is not configured and the provider may only perform `Check`, `Read`, and `Invoke`, all of which return empty results. We justify this behavior becuase it is only possible during a preview and provides the best experience we can manage with the existing gRPC interface. ### Resource Model Changes Providers are now exposed as resources that participate in a stack's dependency graph. Like other resources, they are explicitly created, may have multiple instances, and may have dependencies on other resources. Providers are referred to using provider references, which are a combination of the provider's URN and its ID. This design addresses the need during a preview to refer to providers that have not yet been physically created and therefore have no ID. All custom resources that are not themselves providers must specify a single provider via a provider reference. The named provider will be used to manage that resource's CRUD operations. If a resource's provider reference changes, the resource must be replaced. Though its URN is not present in the resource's dependency list, the provider should be treated as a dependency of the resource when topologically sorting the dependency graph. Finally, `Invoke` operations must now specify a provider to use for the invocation via a provider reference. ### Engine Changes First-class providers support requires a few changes to the engine: - The engine must have some way to map from provider references to provider plugins. It must be possible to add providers from a stack's checkpoint to this map and to register new/updated providers during the execution of a plan in response to CRUD operations on provider resources. - In order to support updating existing stacks using existing Pulumi programs that may not explicitly instantiate providers, the engine must be able to manage the "default" providers for each package referenced by a checkpoint or Pulumi program. The configuration for a "default" provider is taken from the stack's configuration data. The former need is addressed by adding a provider registry type that is responsible for managing all of the plugins required by a plan. In addition to loading plugins froma checkpoint and providing the ability to map from a provider reference to a provider plugin, this type serves as the provider plugin for providers themselves (i.e. it is the "provider provider"). The latter need is solved via two relatively self-contained changes to plan setup and the eval source. During plan setup, the old checkpoint is scanned for custom resources that do not have a provider reference in order to compute the set of packages that require a default provider. Once this set has been computed, the required default provider definitions are conjured and prepended to the checkpoint's resource list. Each resource that requires a default provider is then updated to refer to the default provider for its package. While an eval source is running, each custom resource registration, resource read, and invoke that does not name a provider is trapped before being returned by the source iterator. If no default provider for the appropriate package has been registered, the eval source synthesizes an appropriate registration, waits for it to complete, and records the registered provider's reference. This reference is injected into the original request, which is then processed as usual. If a default provider was already registered, the recorded reference is used and no new registration occurs. ### SDK Changes These changes only expose first-class providers from the Node.JS SDK. - A new abstract class, `ProviderResource`, can be subclassed and used to instantiate first-class providers. - A new field in `ResourceOptions`, `provider`, can be used to supply a particular provider instance to manage a `CustomResource`'s CRUD operations. - A new type, `InvokeOptions`, can be used to specify options that control the behavior of a call to `pulumi.runtime.invoke`. This type includes a `provider` field that is analogous to `ResourceOptions.provider`.
2018-08-07 02:50:29 +02:00
}
// defaultProviderURN generates the URN for the global provider given a package.
func defaultProviderURN(target *Target, source Source, pkg tokens.Package) resource.URN {
return resource.NewURN(target.Name, source.Project(), "", providers.MakeProviderType(pkg), "default")
}
// generateEventURN generates a URN for the resource associated with the given event.
func (d *Deployment) generateEventURN(event SourceEvent) resource.URN {
contract.Require(event != nil, "event != nil")
switch e := event.(type) {
case RegisterResourceEvent:
goal := e.Goal()
return d.generateURN(goal.Parent, goal.Type, goal.Name)
case ReadResourceEvent:
return d.generateURN(e.Parent(), e.Type(), e.Name())
case RegisterResourceOutputsEvent:
return e.URN()
default:
return ""
}
}
// Execute executes a deployment to completion, using the given cancellation context and running a preview or update.
func (d *Deployment) Execute(ctx context.Context, opts Options, preview bool) result.Result {
deploymentExec := &deploymentExecutor{deployment: d}
return deploymentExec.Execute(ctx, opts, preview)
}