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.
|
2017-08-30 03:24:12 +02:00
|
|
|
|
2017-08-23 01:56:15 +02:00
|
|
|
package engine
|
|
|
|
|
|
|
|
import (
|
2018-08-07 01:46:17 +02:00
|
|
|
"sync"
|
2017-08-23 01:56:15 +02:00
|
|
|
"time"
|
|
|
|
|
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"
|
|
|
|
|
2017-09-22 04:18:21 +02:00
|
|
|
"github.com/pulumi/pulumi/pkg/diag"
|
|
|
|
"github.com/pulumi/pulumi/pkg/resource"
|
|
|
|
"github.com/pulumi/pulumi/pkg/resource/deploy"
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
"github.com/pulumi/pulumi/pkg/resource/plugin"
|
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/pulumi/pulumi/pkg/tokens"
|
2017-09-22 04:18:21 +02:00
|
|
|
"github.com/pulumi/pulumi/pkg/util/contract"
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
"github.com/pulumi/pulumi/pkg/workspace"
|
2017-08-23 01:56:15 +02:00
|
|
|
)
|
|
|
|
|
2018-01-18 20:10:15 +01:00
|
|
|
// UpdateOptions contains all the settings for customizing how an update (deploy, preview, or destroy) is performed.
|
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
|
|
|
// nolint: structcheck, host is used in a different file
|
2018-01-18 20:10:15 +01:00
|
|
|
type UpdateOptions struct {
|
2018-04-14 07:26:01 +02:00
|
|
|
// an optional set of analyzers to run as part of this deployment.
|
|
|
|
Analyzers []string
|
|
|
|
|
|
|
|
// the degree of parallelism for resource operations (<=1 for serial).
|
|
|
|
Parallel int
|
|
|
|
|
|
|
|
// true if debugging output it enabled
|
|
|
|
Debug bool
|
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
|
|
|
|
2018-08-23 02:52:46 +02:00
|
|
|
// true if the plan should refresh before executing.
|
|
|
|
Refresh bool
|
|
|
|
|
2018-08-09 23:45:39 +02:00
|
|
|
// true if we should report events for steps that involve default providers.
|
|
|
|
reportDefaultProviderSteps bool
|
|
|
|
|
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
|
|
|
// the plugin host to use for this update
|
|
|
|
host plugin.Host
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
|
|
|
|
2018-01-20 21:07:03 +01:00
|
|
|
// ResourceChanges contains the aggregate resource changes by operation type.
|
|
|
|
type ResourceChanges map[deploy.StepOp]int
|
|
|
|
|
2018-05-05 20:57:09 +02:00
|
|
|
// HasChanges returns true if there are any non-same changes in the resulting summary.
|
|
|
|
func (changes ResourceChanges) HasChanges() bool {
|
|
|
|
var c int
|
|
|
|
for op, count := range changes {
|
|
|
|
if op != deploy.OpSame {
|
|
|
|
c += count
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return c > 0
|
|
|
|
}
|
|
|
|
|
2018-04-20 03:59:14 +02:00
|
|
|
func Update(u UpdateInfo, ctx *Context, opts UpdateOptions, dryRun bool) (ResourceChanges, error) {
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
contract.Require(u != nil, "update")
|
2018-04-20 03:59:14 +02:00
|
|
|
contract.Require(ctx != nil, "ctx")
|
2017-09-09 22:43:51 +02:00
|
|
|
|
2018-04-20 03:59:14 +02:00
|
|
|
defer func() { ctx.Events <- cancelEvent() }()
|
2017-10-23 00:52:00 +02:00
|
|
|
|
2018-05-05 02:01:35 +02:00
|
|
|
info, err := newPlanContext(u, "update", ctx.ParentSpan)
|
2017-08-23 01:56:15 +02:00
|
|
|
if err != nil {
|
2018-01-20 21:07:03 +01:00
|
|
|
return nil, err
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
2018-04-20 03:59:14 +02:00
|
|
|
defer info.Close()
|
2017-10-05 23:08:46 +02:00
|
|
|
|
2018-08-23 00:32:54 +02:00
|
|
|
emitter, err := makeEventEmitter(ctx.Events, u)
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
2018-04-20 03:59:14 +02:00
|
|
|
return update(ctx, info, planOptions{
|
2018-01-18 20:10:15 +01:00
|
|
|
UpdateOptions: opts,
|
2018-03-29 17:57:25 +02:00
|
|
|
SourceFunc: newUpdateSource,
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
Events: emitter,
|
|
|
|
Diag: newEventSink(emitter),
|
2018-04-14 07:26:01 +02:00
|
|
|
}, dryRun)
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
|
|
|
|
2018-04-14 07:26:01 +02:00
|
|
|
func newUpdateSource(
|
|
|
|
opts planOptions, proj *workspace.Project, pwd, main string,
|
|
|
|
target *deploy.Target, plugctx *plugin.Context, dryRun bool) (deploy.Source, error) {
|
|
|
|
|
2018-03-29 17:57:25 +02:00
|
|
|
// Figure out which plugins to load by inspecting the program contents.
|
|
|
|
plugins, err := plugctx.Host.GetRequiredPlugins(plugin.ProgInfo{
|
|
|
|
Proj: proj,
|
|
|
|
Pwd: pwd,
|
|
|
|
Program: main,
|
2018-05-15 05:32:53 +02:00
|
|
|
}, plugin.AllPlugins)
|
2018-03-29 17:57:25 +02:00
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
}
|
2018-03-29 17:57:25 +02:00
|
|
|
|
|
|
|
// Now ensure that we have loaded up any plugins that the program will need in advance.
|
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
|
|
|
const kinds = plugin.AnalyzerPlugins | plugin.LanguagePlugins
|
|
|
|
if err = plugctx.Host.EnsurePlugins(plugins, kinds); err != nil {
|
2018-03-29 17:57:25 +02:00
|
|
|
return nil, 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
|
|
|
// Collect the version information for default providers.
|
|
|
|
defaultProviderVersions := make(map[tokens.Package]*semver.Version)
|
|
|
|
for _, p := range plugins {
|
|
|
|
if p.Kind != workspace.ResourcePlugin {
|
|
|
|
continue
|
|
|
|
}
|
|
|
|
defaultProviderVersions[tokens.Package(p.Name)] = p.Version
|
|
|
|
}
|
|
|
|
|
2018-03-29 17:57:25 +02:00
|
|
|
// If that succeeded, create a new source that will perform interpretation of the compiled program.
|
|
|
|
// TODO[pulumi/pulumi#88]: we are passing `nil` as the arguments map; we need to allow a way to pass these.
|
|
|
|
return deploy.NewEvalSource(plugctx, &deploy.EvalRunInfo{
|
|
|
|
Proj: proj,
|
|
|
|
Pwd: pwd,
|
|
|
|
Program: main,
|
|
|
|
Target: target,
|
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
|
|
|
}, defaultProviderVersions, dryRun), nil
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
|
|
|
|
2018-04-20 03:59:14 +02:00
|
|
|
func update(ctx *Context, info *planContext, opts planOptions, dryRun bool) (ResourceChanges, error) {
|
2018-05-18 23:58:06 +02:00
|
|
|
result, err := plan(ctx, info, opts, dryRun)
|
2017-08-23 01:56:15 +02:00
|
|
|
if err != nil {
|
2018-01-20 21:07:03 +01:00
|
|
|
return nil, err
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
2018-01-20 21:07:03 +01:00
|
|
|
|
|
|
|
var resourceChanges ResourceChanges
|
2017-08-23 01:56:15 +02:00
|
|
|
if result != nil {
|
|
|
|
defer contract.IgnoreClose(result)
|
2017-12-12 21:31:09 +01:00
|
|
|
|
|
|
|
// Make the current working directory the same as the program's, and restore it upon exit.
|
2018-08-22 04:33:59 +02:00
|
|
|
done, chErr := result.Chdir()
|
|
|
|
if chErr != nil {
|
|
|
|
return nil, chErr
|
2017-12-12 21:31:09 +01:00
|
|
|
}
|
|
|
|
defer done()
|
|
|
|
|
2018-04-14 07:26:01 +02:00
|
|
|
if dryRun {
|
2018-01-25 03:22:41 +01:00
|
|
|
// If a dry run, just print the plan, don't actually carry out the deployment.
|
2018-04-20 03:59:14 +02:00
|
|
|
resourceChanges, err = printPlan(ctx, result, dryRun)
|
2017-08-23 01:56:15 +02:00
|
|
|
} else {
|
|
|
|
// Otherwise, we will actually deploy the latest bits.
|
2018-04-14 07:26:01 +02:00
|
|
|
opts.Events.preludeEvent(dryRun, result.Ctx.Update.GetTarget().Config)
|
2017-08-23 01:56:15 +02:00
|
|
|
|
2017-10-02 23:26:51 +02:00
|
|
|
// Walk the plan, reporting progress and executing the actual operations as we go.
|
2017-08-23 01:56:15 +02:00
|
|
|
start := time.Now()
|
2018-04-20 03:59:14 +02:00
|
|
|
actions := newUpdateActions(ctx, info.Update, opts)
|
2018-04-30 19:27:04 +02:00
|
|
|
|
2018-08-22 04:33:59 +02:00
|
|
|
err = result.Walk(ctx, actions, false)
|
2018-01-20 21:07:03 +01:00
|
|
|
resourceChanges = ResourceChanges(actions.Ops)
|
2017-10-05 23:08:46 +02:00
|
|
|
|
2018-08-22 04:33:59 +02:00
|
|
|
if len(resourceChanges) != 0 {
|
|
|
|
// Print out the total number of steps performed (and their kinds), the duration, and any summary info.
|
|
|
|
opts.Events.updateSummaryEvent(actions.MaybeCorrupt, time.Since(start), resourceChanges)
|
2017-09-09 21:42:04 +02:00
|
|
|
}
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
|
|
|
}
|
2018-08-22 04:33:59 +02:00
|
|
|
return resourceChanges, err
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
|
|
|
|
2018-04-26 02:20:08 +02:00
|
|
|
// pluginActions listens for plugin events and persists the set of loaded plugins
|
|
|
|
// to the snapshot.
|
|
|
|
type pluginActions struct {
|
|
|
|
Context *Context
|
|
|
|
}
|
|
|
|
|
|
|
|
func (p *pluginActions) OnPluginLoad(loadedPlug workspace.PluginInfo) error {
|
|
|
|
return p.Context.SnapshotManager.RecordPlugin(loadedPlug)
|
|
|
|
}
|
|
|
|
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
// updateActions pretty-prints the plan application process as it goes.
|
|
|
|
type updateActions struct {
|
2018-04-20 03:59:14 +02:00
|
|
|
Context *Context
|
2018-04-17 08:04:56 +02:00
|
|
|
Steps int
|
|
|
|
Ops map[deploy.StepOp]int
|
|
|
|
Seen map[resource.URN]deploy.Step
|
2018-08-07 01:46:17 +02:00
|
|
|
MapLock sync.Mutex
|
2018-04-17 08:04:56 +02:00
|
|
|
MaybeCorrupt bool
|
|
|
|
Update UpdateInfo
|
|
|
|
Opts planOptions
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
|
|
|
|
2018-04-20 03:59:14 +02:00
|
|
|
func newUpdateActions(context *Context, u UpdateInfo, opts planOptions) *updateActions {
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
return &updateActions{
|
2018-04-20 03:59:14 +02:00
|
|
|
Context: context,
|
|
|
|
Ops: make(map[deploy.StepOp]int),
|
|
|
|
Seen: make(map[resource.URN]deploy.Step),
|
|
|
|
Update: u,
|
|
|
|
Opts: opts,
|
2017-11-17 03:21:41 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
func (acts *updateActions) OnResourceStepPre(step deploy.Step) (interface{}, error) {
|
2018-02-03 01:02:50 +01:00
|
|
|
// Ensure we've marked this step as observed.
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Lock()
|
2018-02-03 01:02:50 +01:00
|
|
|
acts.Seen[step.URN()] = step
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Unlock()
|
2018-02-03 01:02:50 +01:00
|
|
|
|
2018-08-09 23:45:39 +02:00
|
|
|
// Check for a default provider step and skip reporting if necessary.
|
|
|
|
if acts.Opts.reportDefaultProviderSteps || !isDefaultProviderStep(step) {
|
|
|
|
acts.Opts.Events.resourcePreEvent(step, false /*planning*/, acts.Opts.Debug)
|
2018-08-07 01:46:17 +02:00
|
|
|
}
|
|
|
|
|
2017-10-21 18:31:01 +02:00
|
|
|
// Inform the snapshot service that we are about to perform a step.
|
2018-04-26 02:20:08 +02:00
|
|
|
return acts.Context.SnapshotManager.BeginMutation(step)
|
Bring back component outputs
This change brings back component outputs to the overall system again.
In doing so, it generally overhauls the way we do resource RPCs a bit:
* Instead of RegisterResource and CompleteResource, we call these
BeginRegisterResource and EndRegisterResource, which begins to model
these as effectively "asynchronous" resource requests. This should also
help with parallelism (https://github.com/pulumi/pulumi/issues/106).
* Flip the CLI/engine a little on its head. Rather than it driving the
planning and deployment process, we move more to a model where it
simply observes it. This is done by implementing an event handler
interface with three events: OnResourceStepPre, OnResourceStepPost,
and OnResourceComplete. The first two are invoked immediately before
and after any step operation, and the latter is invoked whenever a
EndRegisterResource comes in. The reason for the asymmetry here is
that the checkpointing logic in the deployment engine is largely
untouched (intentionally, as this is a sensitive part of the system),
and so the "begin"/"end" nature doesn't flow through faithfully.
* Also make the engine more event-oriented in its terminology and the
way it handles the incoming BeginRegisterResource and
EndRegisterResource events from the language host. This is the first
step down a long road of incrementally refactoring the engine to work
this way, a necessary prerequisite for parallelism.
2017-11-29 16:42:14 +01:00
|
|
|
}
|
2017-08-23 01:56:15 +02:00
|
|
|
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
func (acts *updateActions) OnResourceStepPost(ctx interface{},
|
2017-11-30 00:05:58 +01:00
|
|
|
step deploy.Step, status resource.Status, err error) error {
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Lock()
|
2018-02-03 01:02:50 +01:00
|
|
|
assertSeen(acts.Seen, step)
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Unlock()
|
2018-02-03 01:02:50 +01:00
|
|
|
|
2018-04-20 03:59:14 +02:00
|
|
|
// If we've already been terminated, exit without writing the checkpoint. We explicitly want to leave the
|
|
|
|
// checkpoint in an inconsistent state in this event.
|
|
|
|
if acts.Context.Cancel.TerminateErr() != nil {
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
|
2018-08-09 23:45:39 +02:00
|
|
|
reportStep := acts.Opts.reportDefaultProviderSteps || !isDefaultProviderStep(step)
|
|
|
|
|
2017-10-03 19:27:59 +02:00
|
|
|
// Report the result of the step.
|
2017-08-23 01:56:15 +02:00
|
|
|
if err != nil {
|
2018-02-04 10:18:06 +01:00
|
|
|
if status == resource.StatusUnknown {
|
2017-10-02 23:26:51 +02:00
|
|
|
acts.MaybeCorrupt = true
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
2018-02-04 10:18:06 +01:00
|
|
|
|
2018-08-09 23:45:39 +02:00
|
|
|
errorURN := resource.URN("")
|
|
|
|
if reportStep {
|
|
|
|
errorURN = step.URN()
|
|
|
|
}
|
|
|
|
|
2018-02-04 10:18:06 +01:00
|
|
|
// Issue a true, bonafide error.
|
2018-08-31 21:33:01 +02:00
|
|
|
acts.Opts.Diag.Errorf(diag.GetPlanApplyFailedError(errorURN), err)
|
2018-08-09 23:45:39 +02:00
|
|
|
if reportStep {
|
|
|
|
acts.Opts.Events.resourceOperationFailedEvent(step, status, acts.Steps, acts.Opts.Debug)
|
|
|
|
}
|
|
|
|
} else if reportStep {
|
2018-08-23 02:52:46 +02:00
|
|
|
op, record := step.Op(), step.Logical()
|
|
|
|
if acts.Opts.isRefresh && op == deploy.OpRefresh {
|
|
|
|
// Refreshes are handled specially.
|
|
|
|
op, record = step.(*deploy.RefreshStep).ResultOp(), true
|
|
|
|
}
|
|
|
|
|
|
|
|
if record {
|
2017-11-29 20:27:32 +01:00
|
|
|
// Increment the counters.
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Lock()
|
2017-11-29 20:27:32 +01:00
|
|
|
acts.Steps++
|
2018-08-23 02:52:46 +02:00
|
|
|
acts.Ops[op]++
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Unlock()
|
2017-11-29 20:27:32 +01:00
|
|
|
}
|
|
|
|
|
2018-04-20 20:52:33 +02:00
|
|
|
// Also show outputs here for custom resources, since there might be some from the initial registration. We do
|
|
|
|
// not show outputs for component resources at this point: any that exist must be from a previous execution of
|
|
|
|
// the Pulumi program, as component resources only report outputs via calls to RegisterResourceOutputs.
|
2018-08-23 02:52:46 +02:00
|
|
|
if step.Res().Custom || acts.Opts.Refresh && step.Op() == deploy.OpRefresh {
|
|
|
|
acts.Opts.Events.resourceOutputsEvent(op, step, false /*planning*/, acts.Opts.Debug)
|
2018-04-20 20:44:28 +02:00
|
|
|
}
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|
2017-10-02 23:27:50 +02:00
|
|
|
|
Implement resource protection (#751)
This change implements resource protection, as per pulumi/pulumi#689.
The overall idea is that a resource can be marked as "protect: true",
which will prevent deletion of that resource for any reason whatsoever
(straight deletion, replacement, etc). This is expressed in the
program. To "unprotect" a resource, one must perform an update setting
"protect: false", and then afterwards, they can delete the resource.
For example:
let res = new MyResource("precious", { .. }, { protect: true });
Afterwards, the resource will display in the CLI with a lock icon, and
any attempts to remove it will fail in the usual ways (in planning or,
worst case, during an actual update).
This was done by adding a new ResourceOptions bag parameter to the
base Resource types. This is unfortunately a breaking change, but now
is the right time to take this one. We had been adding new settings
one by one -- like parent and dependsOn -- and this new approach will
set us up to add any number of additional settings down the road,
without needing to worry about breaking anything ever again.
This is related to protected stacks, as described in
pulumi/pulumi-service#399. Most likely this will serve as a foundational
building block that enables the coarser grained policy management.
2017-12-20 23:31:07 +01:00
|
|
|
// Write out the current snapshot. Note that even if a failure has occurred, we should still have a
|
2018-04-26 02:20:08 +02:00
|
|
|
// safe checkpoint. Note that any error that occurs when writing the checkpoint trumps the error
|
|
|
|
// reported above.
|
2018-05-02 19:36:55 +02:00
|
|
|
return ctx.(SnapshotMutation).End(step, err == nil || status == resource.StatusPartialFailure)
|
Bring back component outputs
This change brings back component outputs to the overall system again.
In doing so, it generally overhauls the way we do resource RPCs a bit:
* Instead of RegisterResource and CompleteResource, we call these
BeginRegisterResource and EndRegisterResource, which begins to model
these as effectively "asynchronous" resource requests. This should also
help with parallelism (https://github.com/pulumi/pulumi/issues/106).
* Flip the CLI/engine a little on its head. Rather than it driving the
planning and deployment process, we move more to a model where it
simply observes it. This is done by implementing an event handler
interface with three events: OnResourceStepPre, OnResourceStepPost,
and OnResourceComplete. The first two are invoked immediately before
and after any step operation, and the latter is invoked whenever a
EndRegisterResource comes in. The reason for the asymmetry here is
that the checkpointing logic in the deployment engine is largely
untouched (intentionally, as this is a sensitive part of the system),
and so the "begin"/"end" nature doesn't flow through faithfully.
* Also make the engine more event-oriented in its terminology and the
way it handles the incoming BeginRegisterResource and
EndRegisterResource events from the language host. This is the first
step down a long road of incrementally refactoring the engine to work
this way, a necessary prerequisite for parallelism.
2017-11-29 16:42:14 +01:00
|
|
|
}
|
2017-10-03 19:27:59 +02:00
|
|
|
|
General prep work for refresh
This change includes a bunch of refactorings I made in prep for
doing refresh (first, the command, see pulumi/pulumi#1081):
* The primary change is to change the way the engine's core update
functionality works with respect to deploy.Source. This is the
way we can plug in new sources of resource information during
planning (and, soon, diffing). The way I intend to model refresh
is by having a new kind of source, deploy.RefreshSource, which
will let us do virtually everything about an update/diff the same
way with refreshes, which avoid otherwise duplicative effort.
This includes changing the planOptions (nee deployOptions) to
take a new SourceFunc callback, which is responsible for creating
a source specific to the kind of plan being requested.
Preview, Update, and Destroy now are primarily differentiated by
the kind of deploy.Source that they return, rather than sprinkling
things like `if Destroying` throughout. This tidies up some logic
and, more importantly, gives us precisely the refresh hook we need.
* Originally, we used the deploy.NullSource for Destroy operations.
This simply returns nothing, which is how Destroy works. For some
reason, we were no longer doing this, and instead had some
`if Destroying` cases sprinkled throughout the deploy.EvalSource.
I think this is a vestige of some old way we did configuration, at
least judging by a comment, which is apparently no longer relevant.
* Move diff and diff-printing logic within the engine into its own
pkg/engine/diff.go file, to prepare for upcoming work.
* I keep noticing benign diffs anytime I regenerate protobufs. I
suspect this is because we're also on different versions. I changed
generate.sh to also dump the version into grpc_version.txt. At
least we can understand where the diffs are coming from, decide
whether to take them (i.e., a newer version), and ensure that as
a team we are monotonically increasing, and not going backwards.
* I also tidied up some tiny things I noticed while in there, like
comments, incorrect types, lint suppressions, and so on.
2018-03-28 16:45:23 +02:00
|
|
|
func (acts *updateActions) OnResourceOutputs(step deploy.Step) error {
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Lock()
|
2018-02-03 01:02:50 +01:00
|
|
|
assertSeen(acts.Seen, step)
|
2018-08-07 01:46:17 +02:00
|
|
|
acts.MapLock.Unlock()
|
2018-02-03 01:02:50 +01:00
|
|
|
|
2018-08-09 23:45:39 +02:00
|
|
|
// Check for a default provider step and skip reporting if necessary.
|
|
|
|
if acts.Opts.reportDefaultProviderSteps || !isDefaultProviderStep(step) {
|
2018-08-23 02:52:46 +02:00
|
|
|
acts.Opts.Events.resourceOutputsEvent(step.Op(), step, false /*planning*/, acts.Opts.Debug)
|
2018-08-09 23:45:39 +02:00
|
|
|
}
|
2018-04-17 08:04:56 +02:00
|
|
|
|
|
|
|
// There's a chance there are new outputs that weren't written out last time.
|
|
|
|
// We need to perform another snapshot write to ensure they get written out.
|
2018-04-26 02:20:08 +02:00
|
|
|
return acts.Context.SnapshotManager.RegisterResourceOutputs(step)
|
2017-08-23 01:56:15 +02:00
|
|
|
}
|