terraform/backend/local/backend_plan.go

357 lines
12 KiB
Go
Raw Normal View History

package local
import (
"bytes"
"context"
"fmt"
"log"
"sort"
"strings"
"github.com/mitchellh/cli"
"github.com/mitchellh/colorstring"
"github.com/hashicorp/terraform/addrs"
"github.com/hashicorp/terraform/backend"
"github.com/hashicorp/terraform/command/format"
"github.com/hashicorp/terraform/plans"
"github.com/hashicorp/terraform/plans/planfile"
"github.com/hashicorp/terraform/states"
"github.com/hashicorp/terraform/states/statemgr"
"github.com/hashicorp/terraform/terraform"
"github.com/hashicorp/terraform/tfdiags"
)
func (b *Local) opPlan(
stopCtx context.Context,
cancelCtx context.Context,
op *backend.Operation,
runningOp *backend.RunningOperation) {
log.Printf("[INFO] backend/local: starting Plan operation")
var diags tfdiags.Diagnostics
if op.PlanFile != nil {
diags = diags.Append(tfdiags.Sourceless(
tfdiags.Error,
"Can't re-plan a saved plan",
"The plan command was given a saved plan file as its input. This command generates "+
"a new plan, and so it requires a configuration directory as its argument.",
))
b.ReportResult(runningOp, diags)
return
}
// Local planning requires a config, unless we're planning to destroy.
if !op.Destroy && !op.HasConfig() {
diags = diags.Append(tfdiags.Sourceless(
tfdiags.Error,
"No configuration files",
"Plan requires configuration to be present. Planning without a configuration would "+
"mark everything for destruction, which is normally not what is desired. If you "+
"would like to destroy everything, run plan with the -destroy option. Otherwise, "+
"create a Terraform configuration file (.tf file) and try again.",
))
b.ReportResult(runningOp, diags)
return
}
// Setup our count hook that keeps track of resource changes
countHook := new(CountHook)
if b.ContextOpts == nil {
b.ContextOpts = new(terraform.ContextOpts)
}
old := b.ContextOpts.Hooks
defer func() { b.ContextOpts.Hooks = old }()
b.ContextOpts.Hooks = append(b.ContextOpts.Hooks, countHook)
// Get our context
tfCtx, configSnap, opState, ctxDiags := b.context(op)
diags = diags.Append(ctxDiags)
if ctxDiags.HasErrors() {
b.ReportResult(runningOp, diags)
return
}
// the state was locked during succesfull context creation; unlock the state
// when the operation completes
defer func() {
err := op.StateLocker.Unlock(nil)
if err != nil {
b.ShowDiagnostics(err)
runningOp.Result = backend.OperationFailure
}
}()
runningOp.State = tfCtx.State()
// Perform the plan in a goroutine so we can be interrupted
var plan *plans.Plan
terraform: ugly huge change to weave in new HCL2-oriented types Due to how deeply the configuration types go into Terraform Core, there isn't a great way to switch out to HCL2 gradually. As a consequence, this huge commit gets us from the old state to a _compilable_ new state, but does not yet attempt to fix any tests and has a number of known missing parts and bugs. We will continue to iterate on this in forthcoming commits, heading back towards passing tests and making Terraform fully-functional again. The three main goals here are: - Use the configuration models from the "configs" package instead of the older models in the "config" package, which is now deprecated and preserved only to help us write our migration tool. - Do expression inspection and evaluation using the functionality of the new "lang" package, instead of the Interpolator type and related functionality in the main "terraform" package. - Represent addresses of various objects using types in the addrs package, rather than hand-constructed strings. This is not critical to support the above, but was a big help during the implementation of these other points since it made it much more explicit what kind of address is expected in each context. Since our new packages are built to accommodate some future planned features that are not yet implemented (e.g. the "for_each" argument on resources, "count"/"for_each" on modules), and since there's still a fair amount of functionality still using old-style APIs, there is a moderate amount of shimming here to connect new assumptions with old, hopefully in a way that makes it easier to find and eliminate these shims later. I apologize in advance to the person who inevitably just found this huge commit while spelunking through the commit history.
2018-04-30 19:33:53 +02:00
var planDiags tfdiags.Diagnostics
doneCh := make(chan struct{})
go func() {
defer close(doneCh)
log.Printf("[INFO] backend/local: plan calling Plan")
terraform: ugly huge change to weave in new HCL2-oriented types Due to how deeply the configuration types go into Terraform Core, there isn't a great way to switch out to HCL2 gradually. As a consequence, this huge commit gets us from the old state to a _compilable_ new state, but does not yet attempt to fix any tests and has a number of known missing parts and bugs. We will continue to iterate on this in forthcoming commits, heading back towards passing tests and making Terraform fully-functional again. The three main goals here are: - Use the configuration models from the "configs" package instead of the older models in the "config" package, which is now deprecated and preserved only to help us write our migration tool. - Do expression inspection and evaluation using the functionality of the new "lang" package, instead of the Interpolator type and related functionality in the main "terraform" package. - Represent addresses of various objects using types in the addrs package, rather than hand-constructed strings. This is not critical to support the above, but was a big help during the implementation of these other points since it made it much more explicit what kind of address is expected in each context. Since our new packages are built to accommodate some future planned features that are not yet implemented (e.g. the "for_each" argument on resources, "count"/"for_each" on modules), and since there's still a fair amount of functionality still using old-style APIs, there is a moderate amount of shimming here to connect new assumptions with old, hopefully in a way that makes it easier to find and eliminate these shims later. I apologize in advance to the person who inevitably just found this huge commit while spelunking through the commit history.
2018-04-30 19:33:53 +02:00
plan, planDiags = tfCtx.Plan()
}()
if b.opWait(doneCh, stopCtx, cancelCtx, tfCtx, opState) {
// If we get in here then the operation was cancelled, which is always
// considered to be a failure.
log.Printf("[INFO] backend/local: plan operation was force-cancelled by interrupt")
runningOp.Result = backend.OperationFailure
return
}
log.Printf("[INFO] backend/local: plan operation completed")
terraform: ugly huge change to weave in new HCL2-oriented types Due to how deeply the configuration types go into Terraform Core, there isn't a great way to switch out to HCL2 gradually. As a consequence, this huge commit gets us from the old state to a _compilable_ new state, but does not yet attempt to fix any tests and has a number of known missing parts and bugs. We will continue to iterate on this in forthcoming commits, heading back towards passing tests and making Terraform fully-functional again. The three main goals here are: - Use the configuration models from the "configs" package instead of the older models in the "config" package, which is now deprecated and preserved only to help us write our migration tool. - Do expression inspection and evaluation using the functionality of the new "lang" package, instead of the Interpolator type and related functionality in the main "terraform" package. - Represent addresses of various objects using types in the addrs package, rather than hand-constructed strings. This is not critical to support the above, but was a big help during the implementation of these other points since it made it much more explicit what kind of address is expected in each context. Since our new packages are built to accommodate some future planned features that are not yet implemented (e.g. the "for_each" argument on resources, "count"/"for_each" on modules), and since there's still a fair amount of functionality still using old-style APIs, there is a moderate amount of shimming here to connect new assumptions with old, hopefully in a way that makes it easier to find and eliminate these shims later. I apologize in advance to the person who inevitably just found this huge commit while spelunking through the commit history.
2018-04-30 19:33:53 +02:00
diags = diags.Append(planDiags)
if planDiags.HasErrors() {
b.ReportResult(runningOp, diags)
return
}
backend/local: treat output changes as side-effects to be applied This is a baby-step towards an intended future where all Terraform actions which have side-effects in either remote objects or the Terraform state can go through the plan+apply workflow. This initial change is focused only on allowing plan+apply for changes to root module output values, so that these can be written into a new state snapshot (for consumption by terraform_remote_state elsewhere) without having to go outside of the primary workflow by running "terraform refresh". This is also better than "terraform refresh" because it gives an opportunity to review the proposed changes before applying them, as we're accustomed to with resource changes. The downside here is that Terraform Core was not designed to produce accurate changesets for root module outputs. Although we added a place for it in the plan model in Terraform 0.12, Terraform Core currently produces inaccurate changesets there which don't properly track the prior values. We're planning to rework Terraform Core's evaluation approach in a forthcoming release so it would itself be able to distinguish between the prior state and the planned new state to produce an accurate changeset, but this commit introduces a temporary stop-gap solution of implementing the logic up in the local backend code, where we can freeze a snapshot of the prior state before we take any other actions and then use that to produce an accurate output changeset to decide whether the plan has externally-visible side-effects and render any changes to output values. This temporary approach should be replaced by a more appropriately-placed solution in Terraform Core in a release, which should then allow further behaviors in similar vein, such as user-visible drift detection for resource instances.
2020-05-27 01:59:06 +02:00
// Record whether this plan includes any side-effects that could be applied.
runningOp.PlanEmpty = plan.Changes.Empty()
// Save the plan to disk
if path := op.PlanOutPath; path != "" {
if op.PlanOutBackend == nil {
// This is always a bug in the operation caller; it's not valid
// to set PlanOutPath without also setting PlanOutBackend.
diags = diags.Append(fmt.Errorf(
"PlanOutPath set without also setting PlanOutBackend (this is a bug in Terraform)"),
)
b.ReportResult(runningOp, diags)
return
}
plan.Backend = *op.PlanOutBackend
// We may have updated the state in the refresh step above, but we
// will freeze that updated state in the plan file for now and
// only write it if this plan is subsequently applied.
plannedStateFile := statemgr.PlannedStateUpdate(opState, plan.State)
log.Printf("[INFO] backend/local: writing plan output to: %s", path)
err := planfile.Create(path, configSnap, plannedStateFile, plan)
if err != nil {
diags = diags.Append(tfdiags.Sourceless(
tfdiags.Error,
"Failed to write plan file",
fmt.Sprintf("The plan file could not be written: %s.", err),
))
b.ReportResult(runningOp, diags)
return
}
}
// Perform some output tasks if we have a CLI to output to.
if b.CLI != nil {
schemas := tfCtx.Schemas()
backend/local: treat output changes as side-effects to be applied This is a baby-step towards an intended future where all Terraform actions which have side-effects in either remote objects or the Terraform state can go through the plan+apply workflow. This initial change is focused only on allowing plan+apply for changes to root module output values, so that these can be written into a new state snapshot (for consumption by terraform_remote_state elsewhere) without having to go outside of the primary workflow by running "terraform refresh". This is also better than "terraform refresh" because it gives an opportunity to review the proposed changes before applying them, as we're accustomed to with resource changes. The downside here is that Terraform Core was not designed to produce accurate changesets for root module outputs. Although we added a place for it in the plan model in Terraform 0.12, Terraform Core currently produces inaccurate changesets there which don't properly track the prior values. We're planning to rework Terraform Core's evaluation approach in a forthcoming release so it would itself be able to distinguish between the prior state and the planned new state to produce an accurate changeset, but this commit introduces a temporary stop-gap solution of implementing the logic up in the local backend code, where we can freeze a snapshot of the prior state before we take any other actions and then use that to produce an accurate output changeset to decide whether the plan has externally-visible side-effects and render any changes to output values. This temporary approach should be replaced by a more appropriately-placed solution in Terraform Core in a release, which should then allow further behaviors in similar vein, such as user-visible drift detection for resource instances.
2020-05-27 01:59:06 +02:00
if runningOp.PlanEmpty {
b.CLI.Output("\n" + b.Colorize().Color(strings.TrimSpace(planNoChanges)))
// Even if there are no changes, there still could be some warnings
b.ShowDiagnostics(diags)
return
}
b.renderPlan(plan, plan.State, schemas)
// If we've accumulated any warnings along the way then we'll show them
// here just before we show the summary and next steps. If we encountered
// errors then we would've returned early at some other point above.
b.ShowDiagnostics(diags)
cli: allow disabling "next steps" message in terraform plan In #15884 we adjusted the plan output to give an explicit command to run to apply a plan, whereas before this command was just alluded to in the prose. Since releasing that, we've got good feedback that it's confusing to include such instructions when Terraform is running in a workflow automation tool, because such tools usually abstract away exactly what commands are run and require users to take different actions to proceed through the workflow. To accommodate such environments while retaining helpful messages for normal CLI usage, here we introduce a new environment variable TF_IN_AUTOMATION which, when set to a non-empty value, is a hint to Terraform that it isn't being run in an interactive command shell and it should thus tone down the "next steps" messaging. The documentation for this setting is included as part of the "...in automation" guide since it's not generally useful in other cases. We also intentionally disclaim comprehensive support for this since we want to avoid creating an extreme number of "if running in automation..." codepaths that would increase the testing matrix and hurt maintainability. The focus is specifically on the output of the three commands we give in the automation guide, which at present means the following two situations: * "terraform init" does not include the final paragraphs that suggest running "terraform plan" and tell you in what situations you might need to re-run "terraform init". * "terraform plan" does not include the final paragraphs that either warn about not specifying "-out=..." or instruct to run "terraform apply" with the generated plan file.
2017-09-09 02:14:37 +02:00
// Give the user some next-steps, unless we're running in an automation
// tool which is presumed to provide its own UI for further actions.
if !b.RunningInAutomation {
b.CLI.Output("\n------------------------------------------------------------------------")
if path := op.PlanOutPath; path == "" {
b.CLI.Output(fmt.Sprintf(
"\n" + strings.TrimSpace(planHeaderNoOutput) + "\n",
))
} else {
b.CLI.Output(fmt.Sprintf(
"\n"+strings.TrimSpace(planHeaderYesOutput)+"\n",
path, path,
))
}
}
}
}
func (b *Local) renderPlan(plan *plans.Plan, baseState *states.State, schemas *terraform.Schemas) {
RenderPlan(plan, baseState, schemas, b.CLI, b.Colorize())
}
// RenderPlan renders the given plan to the given UI.
//
// This is exported only so that the "terraform show" command can re-use it.
// Ideally it would be somewhere outside of this backend code so that both
// can call into it, but we're leaving it here for now in order to avoid
// disruptive refactoring.
//
// If you find yourself wanting to call this function from a third callsite,
// please consider whether it's time to do the more disruptive refactoring
// so that something other than the local backend package is offering this
// functionality.
backend/local: treat output changes as side-effects to be applied This is a baby-step towards an intended future where all Terraform actions which have side-effects in either remote objects or the Terraform state can go through the plan+apply workflow. This initial change is focused only on allowing plan+apply for changes to root module output values, so that these can be written into a new state snapshot (for consumption by terraform_remote_state elsewhere) without having to go outside of the primary workflow by running "terraform refresh". This is also better than "terraform refresh" because it gives an opportunity to review the proposed changes before applying them, as we're accustomed to with resource changes. The downside here is that Terraform Core was not designed to produce accurate changesets for root module outputs. Although we added a place for it in the plan model in Terraform 0.12, Terraform Core currently produces inaccurate changesets there which don't properly track the prior values. We're planning to rework Terraform Core's evaluation approach in a forthcoming release so it would itself be able to distinguish between the prior state and the planned new state to produce an accurate changeset, but this commit introduces a temporary stop-gap solution of implementing the logic up in the local backend code, where we can freeze a snapshot of the prior state before we take any other actions and then use that to produce an accurate output changeset to decide whether the plan has externally-visible side-effects and render any changes to output values. This temporary approach should be replaced by a more appropriately-placed solution in Terraform Core in a release, which should then allow further behaviors in similar vein, such as user-visible drift detection for resource instances.
2020-05-27 01:59:06 +02:00
//
// The difference between baseState and priorState is that baseState is the
// result of implicitly running refresh (unless that was disabled) while
// priorState is a snapshot of the state as it was before we took any actions
// at all. priorState can optionally be nil if the caller has only a saved
// plan and not the prior state it was built from. In that case, changes to
// output values will not currently be rendered because their prior values
// are currently stored only in the prior state. (see the docstring for
// func planHasSideEffects for why this is and when that might change)
func RenderPlan(plan *plans.Plan, baseState *states.State, schemas *terraform.Schemas, ui cli.Ui, colorize *colorstring.Colorize) {
counts := map[plans.Action]int{}
var rChanges []*plans.ResourceInstanceChangeSrc
for _, change := range plan.Changes.Resources {
if change.Action == plans.Delete && change.Addr.Resource.Resource.Mode == addrs.DataResourceMode {
// Avoid rendering data sources on deletion
continue
}
rChanges = append(rChanges, change)
counts[change.Action]++
}
headerBuf := &bytes.Buffer{}
fmt.Fprintf(headerBuf, "\n%s\n", strings.TrimSpace(planHeaderIntro))
if counts[plans.Create] > 0 {
fmt.Fprintf(headerBuf, "%s create\n", format.DiffActionSymbol(plans.Create))
}
if counts[plans.Update] > 0 {
fmt.Fprintf(headerBuf, "%s update in-place\n", format.DiffActionSymbol(plans.Update))
}
if counts[plans.Delete] > 0 {
fmt.Fprintf(headerBuf, "%s destroy\n", format.DiffActionSymbol(plans.Delete))
}
if counts[plans.DeleteThenCreate] > 0 {
fmt.Fprintf(headerBuf, "%s destroy and then create replacement\n", format.DiffActionSymbol(plans.DeleteThenCreate))
}
if counts[plans.CreateThenDelete] > 0 {
fmt.Fprintf(headerBuf, "%s create replacement and then destroy\n", format.DiffActionSymbol(plans.CreateThenDelete))
}
if counts[plans.Read] > 0 {
fmt.Fprintf(headerBuf, "%s read (data resources)\n", format.DiffActionSymbol(plans.Read))
}
ui.Output(colorize.Color(headerBuf.String()))
ui.Output("Terraform will perform the following actions:\n")
// Note: we're modifying the backing slice of this plan object in-place
// here. The ordering of resource changes in a plan is not significant,
// but we can only do this safely here because we can assume that nobody
// is concurrently modifying our changes while we're trying to print it.
sort.Slice(rChanges, func(i, j int) bool {
iA := rChanges[i].Addr
jA := rChanges[j].Addr
if iA.String() == jA.String() {
return rChanges[i].DeposedKey < rChanges[j].DeposedKey
}
return iA.Less(jA)
})
for _, rcs := range rChanges {
if rcs.Action == plans.NoOp {
continue
}
providerSchema := schemas.ProviderSchema(rcs.ProviderAddr.Provider)
if providerSchema == nil {
// Should never happen
ui.Output(fmt.Sprintf("(schema missing for %s)\n", rcs.ProviderAddr))
continue
}
rSchema, _ := providerSchema.SchemaForResourceAddr(rcs.Addr.Resource.Resource)
if rSchema == nil {
// Should never happen
ui.Output(fmt.Sprintf("(schema missing for %s)\n", rcs.Addr))
continue
}
// check if the change is due to a tainted resource
tainted := false
backend/local: treat output changes as side-effects to be applied This is a baby-step towards an intended future where all Terraform actions which have side-effects in either remote objects or the Terraform state can go through the plan+apply workflow. This initial change is focused only on allowing plan+apply for changes to root module output values, so that these can be written into a new state snapshot (for consumption by terraform_remote_state elsewhere) without having to go outside of the primary workflow by running "terraform refresh". This is also better than "terraform refresh" because it gives an opportunity to review the proposed changes before applying them, as we're accustomed to with resource changes. The downside here is that Terraform Core was not designed to produce accurate changesets for root module outputs. Although we added a place for it in the plan model in Terraform 0.12, Terraform Core currently produces inaccurate changesets there which don't properly track the prior values. We're planning to rework Terraform Core's evaluation approach in a forthcoming release so it would itself be able to distinguish between the prior state and the planned new state to produce an accurate changeset, but this commit introduces a temporary stop-gap solution of implementing the logic up in the local backend code, where we can freeze a snapshot of the prior state before we take any other actions and then use that to produce an accurate output changeset to decide whether the plan has externally-visible side-effects and render any changes to output values. This temporary approach should be replaced by a more appropriately-placed solution in Terraform Core in a release, which should then allow further behaviors in similar vein, such as user-visible drift detection for resource instances.
2020-05-27 01:59:06 +02:00
if !baseState.Empty() {
if is := baseState.ResourceInstance(rcs.Addr); is != nil {
if obj := is.GetGeneration(rcs.DeposedKey.Generation()); obj != nil {
tainted = obj.Status == states.ObjectTainted
}
}
}
ui.Output(format.ResourceChange(
rcs,
tainted,
rSchema,
colorize,
))
}
// stats is similar to counts above, but:
// - it considers only resource changes
// - it simplifies "replace" into both a create and a delete
stats := map[plans.Action]int{}
for _, change := range rChanges {
switch change.Action {
case plans.CreateThenDelete, plans.DeleteThenCreate:
stats[plans.Create]++
stats[plans.Delete]++
default:
stats[change.Action]++
}
}
ui.Output(colorize.Color(fmt.Sprintf(
"[reset][bold]Plan:[reset] "+
"%d to add, %d to change, %d to destroy.",
stats[plans.Create], stats[plans.Update], stats[plans.Delete],
)))
backend/local: treat output changes as side-effects to be applied This is a baby-step towards an intended future where all Terraform actions which have side-effects in either remote objects or the Terraform state can go through the plan+apply workflow. This initial change is focused only on allowing plan+apply for changes to root module output values, so that these can be written into a new state snapshot (for consumption by terraform_remote_state elsewhere) without having to go outside of the primary workflow by running "terraform refresh". This is also better than "terraform refresh" because it gives an opportunity to review the proposed changes before applying them, as we're accustomed to with resource changes. The downside here is that Terraform Core was not designed to produce accurate changesets for root module outputs. Although we added a place for it in the plan model in Terraform 0.12, Terraform Core currently produces inaccurate changesets there which don't properly track the prior values. We're planning to rework Terraform Core's evaluation approach in a forthcoming release so it would itself be able to distinguish between the prior state and the planned new state to produce an accurate changeset, but this commit introduces a temporary stop-gap solution of implementing the logic up in the local backend code, where we can freeze a snapshot of the prior state before we take any other actions and then use that to produce an accurate output changeset to decide whether the plan has externally-visible side-effects and render any changes to output values. This temporary approach should be replaced by a more appropriately-placed solution in Terraform Core in a release, which should then allow further behaviors in similar vein, such as user-visible drift detection for resource instances.
2020-05-27 01:59:06 +02:00
// If there is at least one planned change to the root module outputs
// then we'll render a summary of those too.
var changedRootModuleOutputs []*plans.OutputChangeSrc
for _, output := range plan.Changes.Outputs {
if !output.Addr.Module.IsRoot() {
continue
}
if output.ChangeSrc.Action == plans.NoOp {
continue
}
changedRootModuleOutputs = append(changedRootModuleOutputs, output)
}
if len(changedRootModuleOutputs) > 0 {
ui.Output(colorize.Color("[reset]\n[bold]Changes to Outputs:[reset]" + format.OutputChanges(changedRootModuleOutputs, colorize)))
backend/local: treat output changes as side-effects to be applied This is a baby-step towards an intended future where all Terraform actions which have side-effects in either remote objects or the Terraform state can go through the plan+apply workflow. This initial change is focused only on allowing plan+apply for changes to root module output values, so that these can be written into a new state snapshot (for consumption by terraform_remote_state elsewhere) without having to go outside of the primary workflow by running "terraform refresh". This is also better than "terraform refresh" because it gives an opportunity to review the proposed changes before applying them, as we're accustomed to with resource changes. The downside here is that Terraform Core was not designed to produce accurate changesets for root module outputs. Although we added a place for it in the plan model in Terraform 0.12, Terraform Core currently produces inaccurate changesets there which don't properly track the prior values. We're planning to rework Terraform Core's evaluation approach in a forthcoming release so it would itself be able to distinguish between the prior state and the planned new state to produce an accurate changeset, but this commit introduces a temporary stop-gap solution of implementing the logic up in the local backend code, where we can freeze a snapshot of the prior state before we take any other actions and then use that to produce an accurate output changeset to decide whether the plan has externally-visible side-effects and render any changes to output values. This temporary approach should be replaced by a more appropriately-placed solution in Terraform Core in a release, which should then allow further behaviors in similar vein, such as user-visible drift detection for resource instances.
2020-05-27 01:59:06 +02:00
}
}
const planHeaderIntro = `
An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
`
const planHeaderNoOutput = `
Note: You didn't specify an "-out" parameter to save this plan, so Terraform
can't guarantee that exactly these actions will be performed if
"terraform apply" is subsequently run.
`
const planHeaderYesOutput = `
This plan was saved to: %s
To perform exactly these actions, run the following command to apply:
terraform apply %q
`
const planNoChanges = `
[reset][bold][green]No changes. Infrastructure is up-to-date.[reset][green]
This means that Terraform did not detect any differences between your
configuration and real physical resources that exist. As a result, no
actions need to be performed.
`