opentofu/command/state_rm.go

175 lines
4.8 KiB
Go
Raw Normal View History

package command
import (
2019-01-08 07:57:52 -06:00
"context"
"fmt"
"strings"
"github.com/hashicorp/terraform/addrs"
2019-01-08 07:57:52 -06:00
"github.com/hashicorp/terraform/command/clistate"
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
"github.com/hashicorp/terraform/tfdiags"
"github.com/mitchellh/cli"
)
// StateRmCommand is a Command implementation that shows a single resource.
type StateRmCommand struct {
StateMeta
}
func (c *StateRmCommand) Run(args []string) int {
args = c.Meta.process(args)
var dryRun bool
cmdFlags := c.Meta.defaultFlagSet("state rm")
cmdFlags.BoolVar(&dryRun, "dry-run", false, "dry run")
cmdFlags.StringVar(&c.backupPath, "backup", "-", "backup")
cmdFlags.BoolVar(&c.Meta.stateLock, "lock", true, "lock state")
cmdFlags.DurationVar(&c.Meta.stateLockTimeout, "lock-timeout", 0, "lock timeout")
cmdFlags.StringVar(&c.statePath, "state", "", "path")
if err := cmdFlags.Parse(args); err != nil {
c.Ui.Error(fmt.Sprintf("Error parsing command-line flags: %s\n", err.Error()))
return 1
}
args = cmdFlags.Args()
if len(args) < 1 {
c.Ui.Error("At least one address is required.\n")
return cli.RunResultHelp
}
// Get the state
stateMgr, err := c.State()
if err != nil {
c.Ui.Error(fmt.Sprintf(errStateLoadingState, err))
return 1
}
2019-01-08 07:57:52 -06:00
if c.stateLock {
stateLocker := clistate.NewLocker(context.Background(), c.stateLockTimeout, c.Ui, c.Colorize())
if err := stateLocker.Lock(stateMgr, "state-rm"); err != nil {
c.Ui.Error(fmt.Sprintf("Error locking state: %s", err))
return 1
}
defer stateLocker.Unlock(nil)
}
if err := stateMgr.RefreshState(); err != nil {
c.Ui.Error(fmt.Sprintf("Failed to refresh state: %s", err))
2017-02-21 22:35:43 -06:00
return 1
}
state := stateMgr.State()
if state == nil {
c.Ui.Error(fmt.Sprintf(errStateNotFound))
return 1
}
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
// This command primarily works with resource instances, though it will
// also clean up any modules and resources left empty by actions it takes.
var addrs []addrs.AbsResourceInstance
var diags tfdiags.Diagnostics
for _, addrStr := range args {
moreAddrs, moreDiags := c.lookupResourceInstanceAddr(state, true, addrStr)
addrs = append(addrs, moreAddrs...)
diags = diags.Append(moreDiags)
}
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
if diags.HasErrors() {
c.showDiagnostics(diags)
return 1
}
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
prefix := "Removed "
if dryRun {
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
prefix = "Would remove "
}
var isCount int
ss := state.SyncWrapper()
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
for _, addr := range addrs {
isCount++
c.Ui.Output(prefix + addr.String())
if !dryRun {
ss.ForgetResourceInstanceAll(addr)
ss.RemoveResourceIfEmpty(addr.ContainingResource())
}
}
if dryRun {
if isCount == 0 {
c.Ui.Output("Would have removed nothing.")
}
return 0 // This is as far as we go in dry-run mode
}
if err := stateMgr.WriteState(state); err != nil {
c.Ui.Error(fmt.Sprintf(errStateRmPersist, err))
return 1
}
if err := stateMgr.PersistState(); err != nil {
c.Ui.Error(fmt.Sprintf(errStateRmPersist, err))
return 1
}
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
if len(diags) > 0 {
c.showDiagnostics(diags)
}
if isCount == 0 {
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
c.Ui.Output("No matching resource instances found.")
} else {
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
c.Ui.Output(fmt.Sprintf("Successfully removed %d resource instance(s).", isCount))
}
return 0
}
func (c *StateRmCommand) Help() string {
helpText := `
Usage: terraform state rm [options] ADDRESS...
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
Remove one or more items from the Terraform state, causing Terraform to
"forget" those items without first destroying them in the remote system.
This command removes one or more resource instances from the Terraform state
based on the addresses given. You can view and list the available instances
with "terraform state list".
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
If you give the address of an entire module then all of the instances in
that module and any of its child modules will be removed from the state.
If you give the address of a resource that has "count" or "for_each" set,
all of the instances of that resource will be removed from the state.
Options:
-dry-run If set, prints out what would've been removed but
doesn't actually remove anything.
-backup=PATH Path where Terraform should write the backup
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
state.
-lock=true Lock the state file when locking is supported.
-lock-timeout=0s Duration to retry a state lock.
command: Fix various issues in the "terraform state ..." subcommands In earlier refactoring we updated these commands to support the new address and state types, but attempted to partially retain the old-style "StateFilter" abstraction that originally lived in the Terraform package, even though that was no longer being used for any other functionality. Unfortunately the adaptation of the existing filtering to the new types wasn't exact and so these commands ended up having a few bugs that were not covered by the existing tests. Since the old StateFilter behavior was the source of various misbehavior anyway, here it's removed altogether and replaced with some simpler functions in the state_meta.go file that are tailored to the use-cases of these sub-commands. As well as just generally behaving more consistently with the other parts of Terraform that use the new resource address types, this commit fixes the following bugs: - A resource address of aws_instance.foo would previously match an resource of that type and name in any module, which disagreed with the expected interpretation elsewhere of meaning a single resource in the root module. - The "terraform state mv" command was not supporting moves from a single resource address to an indexed address and vice-versa, because the old logic didn't need to make that distinction while they are two separate address types in the new logic. Now we allow resources that do not have count/for_each to be treated as if they are instances for the purposes of this command, which is a better match for likely user intent and for the old behavior. Finally, we also clean up a little some of the usage output from these commands, which hasn't been updated for some time and so had both some stale information and some inaccurate terminology.
2019-03-15 21:51:26 -05:00
-state=PATH Path to the state file to update. Defaults to the current
workspace state.
`
return strings.TrimSpace(helpText)
}
func (c *StateRmCommand) Synopsis() string {
return "Remove instances from the state"
}
const errStateRm = `Error removing items from the state: %s
The state was not saved. No items were removed from the persisted
state. No backup was created since no modification occurred. Please
resolve the issue above and try again.`
const errStateRmPersist = `Error saving the state: %s
The state was not saved. No items were removed from the persisted
state. No backup was created since no modification occurred. Please
resolve the issue above and try again.`