Skip to main content

Command: plan

The tofu plan command creates an execution plan, which lets you preview the changes that OpenTofu plans to make to your infrastructure. By default, when OpenTofu creates a plan it:

  • Reads the current state of any already-existing remote objects to make sure that the OpenTofu state is up-to-date.
  • Compares the current configuration to the prior state and noting any differences.
  • Proposes a set of change actions that should, if applied, make the remote objects match the configuration.

The plan command alone does not actually carry out the proposed changes You can use this command to check whether the proposed changes match what you expected before you apply the changes or share your changes with your team for broader review.

If OpenTofu detects that no changes are needed to resource instances or to root module output values, tofu plan will report that no actions need to be taken.

If you are using OpenTofu directly in an interactive terminal and you expect to apply the changes OpenTofu proposes, you can alternatively run tofu apply directly. By default, the "apply" command automatically generates a new plan and prompts for you to approve it.

You can use the optional -out=FILE option to save the generated plan to a file on disk, which you can later execute by passing the file to tofu apply as an extra argument. This two-step workflow is primarily intended for when running OpenTofu in automation.

If you run tofu plan without the -out=FILE option then it will create a speculative plan, which is a description of the effect of the plan but without any intent to actually apply it.

In teams that use a version control and code review workflow for making changes to real infrastructure, developers can use speculative plans to verify the effect of their changes before submitting them for code review. However, it's important to consider that other changes made to the target system in the meantime might cause the final effect of a configuration change to be different than what an earlier speculative plan indicated, so you should always re-check the final non-speculative plan before applying to make sure that it still matches your intent.


Usage: tofu plan [options]

The plan subcommand looks in the current working directory for the root module configuration.

Because the plan command is one of the main commands of OpenTofu, it has a variety of different options, described in the following sections. However, most of the time you should not need to set any of these options, because a OpenTofu configuration should typically be designed to work with no special additional options for routine work.

The remaining sections on this page describe the various options:

  • Planning Modes: There are some special alternative planning modes that you can use for some special situations where your goal is not just to change the remote system to match your configuration.
  • Planning Options: Alongside the special planning modes, there are also some options you can set in order to customize the planning process for unusual needs.
    • Resource Targeting is one particular special planning option that has some important caveats associated with it.
  • Other Options: These change the behavior of the planning command itself, rather than customizing the content of the generated plan.

Planning Modes

The previous section describes OpenTofu's default planning behavior, which changes the remote system to match the changes you make to your configuration. OpenTofu has two alternative planning modes, each of which creates a plan with a different intended outcome. These options are available for both tofu plan and tofu apply.

  • Destroy mode: creates a plan whose goal is to destroy all remote objects that currently exist, leaving an empty OpenTofu state. It is the same as running tofu destroy. Destroy mode can be useful for situations like transient development environments, where the managed objects cease to be useful once the development task is complete.

    Activate destroy mode using the -destroy command line option.

  • Refresh-only mode: creates a plan whose goal is only to update the OpenTofu state and any root module output values to match changes made to remote objects outside of OpenTofu. This can be useful if you've intentionally changed one or more remote objects outside of the usual workflow (e.g. while responding to an incident) and you now need to reconcile OpenTofu's records with those changes.

    Activate refresh-only mode using the -refresh-only command line option.

In situations where we need to discuss the default planning mode that OpenTofu uses when none of the alternative modes are selected, we refer to it as "Normal mode". Because these alternative modes are for specialized situations only, some other OpenTofu documentation only discusses the normal planning mode.

The planning modes are all mutually-exclusive, so activating any non-default planning mode disables the "normal" planning mode, and you can't use more than one alternative mode at the same time.

Planning Options

In addition to alternate planning modes, there are several options that can modify planning behavior. These options are available for both tofu plan and tofu apply.

  • -refresh=false - Disables the default behavior of synchronizing the OpenTofu state with remote objects before checking for configuration changes. This can make the planning operation faster by reducing the number of remote API requests. However, setting refresh=false causes OpenTofu to ignore external changes, which could result in an incomplete or incorrect plan. You cannot use refresh=false in refresh-only planning mode because it would effectively disable the entirety of the planning operation.

  • -replace=ADDRESS - Instructs OpenTofu to plan to replace the resource instance with the given address. This is helpful when one or more remote objects have become degraded, and you can use replacement objects with the same configuration to align with immutable infrastructure patterns. OpenTofu will use a "replace" action if the specified resource would normally cause an "update" action or no action at all. Include this option multiple times to replace several objects at once. You cannot use -replace with the -destroy option.

  • -target=ADDRESS - Instructs OpenTofu to focus its planning efforts only on resource instances which match the given address and on any objects that those instances depend on.

  • -var 'NAME=VALUE' - Sets a value for a single input variable declared in the root module of the configuration. Use this option multiple times to set more than one variable. Refer to Input Variables on the Command Line for more information.

  • -var-file=FILENAME - Sets values for potentially many input variables declared in the root module of the configuration, using definitions from a "tfvars" file. Use this option multiple times to include values from more than one file.

There are several other ways to set values for input variables in the root module, aside from the -var and -var-file options. Refer to Assigning Values to Root Module Variables for more information.

Input Variables on the Command Line

You can use the -var command line option to specify values for input variables declared in your root module.

However, to do so will require writing a command line that is parsable both by your chosen command line shell and OpenTofu, which can be complicated for expressions involving lots of quotes and escape sequences. In most cases we recommend using the -var-file option instead, and write your actual values in a separate file so that OpenTofu can parse them directly, rather than interpreting the result of your shell's parsing.

To use -var on a Unix-style shell on a system like Linux or macOS we recommend writing the option argument in single quotes ' to ensure the shell will interpret the value literally:

Code Block
tofu plan -var 'name=value'

If your intended value also includes a single quote then you'll still need to escape that for correct interpretation by your shell, which also requires temporarily ending the quoted sequence so that the backslash escape character will be significant:

Code Block
tofu plan -var 'name=va'\''lue'

When using OpenTofu on Windows, we recommend using the Windows Command Prompt (cmd.exe). When you pass a variable value to OpenTofu from the Windows Command Prompt, use double quotes " around the argument:

Code Block
tofu plan -var "name=value"

If your intended value includes literal double quotes then you'll need to escape those with a backslash:

Code Block
tofu plan -var "name=va\"lue"

PowerShell on Windows cannot correctly pass literal quotes to external programs, so we do not recommend using OpenTofu with PowerShell when you are on Windows. Use Windows Command Prompt instead.

The appropriate syntax for writing the variable value is different depending on the variable's type constraint. The primitive types string, number, and bool all expect a direct string value with no special punctuation except that required by your shell, as shown in the above examples. For all other type constraints, including list, map, and set types and the special any keyword, you must write a valid OpenTofu language expression representing the value, and write any necessary quoting or escape characters to ensure it will pass through your shell literally to OpenTofu. For example, for a list(string) type constraint:

Code Block
# Unix-style shell
tofu plan -var 'name=["a", "b", "c"]'

# Windows Command Prompt (do not use PowerShell on Windows)
tofu plan -var "name=[\"a\", \"b\", \"c\"]"

Similar constraints apply when setting input variables using environment variables. For more information on the various methods for setting root module input variables, see Assigning Values to Root Module Variables.

Resource Targeting

You can use the -target option to focus OpenTofu's attention on only a subset of resources. You can use resource address syntax to specify the constraint. OpenTofu interprets the resource address as follows:

  • If the given address identifies one specific resource instance, OpenTofu will select that instance alone. For resources with either count or for_each set, a resource instance address must include the instance index part, like aws_instance.example[0].

  • If the given address identifies a resource as a whole, OpenTofu will select all of the instances of that resource. For resources with either count or for_each set, this means selecting all instance indexes currently associated with that resource. For single-instance resources (without either count or for_each), the resource address and the resource instance address are identical, so this possibility does not apply.

  • If the given address identifies an entire module instance, OpenTofu will select all instances of all resources that belong to that module instance and all of its child module instances.

Once OpenTofu has selected one or more resource instances that you've directly targeted, it will also then extend the selection to include all other objects that those selections depend on either directly or indirectly.

This targeting capability is provided for exceptional circumstances, such as recovering from mistakes or working around OpenTofu limitations. It is not recommended to use -target for routine operations, since this can lead to undetected configuration drift and confusion about how the true state of resources relates to configuration.

Instead of using -target as a means to operate on isolated portions of very large configurations, prefer instead to break large configurations into several smaller configurations that can each be independently applied. Data sources can be used to access information about resources created in other configurations, allowing a complex system architecture to be broken down into more manageable parts that can be updated independently.

Other Options

The tofu plan command also has some other options that are related to the input and output of the planning command, rather than customizing what sort of plan OpenTofu will create. These commands are not necessarily also available on tofu apply, unless otherwise stated in the documentation for that command.

The available options are:

  • -compact-warnings - Shows any warning messages in a compact form which includes only the summary messages, unless the warnings are accompanied by at least one error and thus the warning text might be useful context for the errors.

  • -detailed-exitcode - Returns a detailed exit code when the command exits. When provided, this argument changes the exit codes and their meanings to provide more granular information about what the resulting plan contains:

    • 0 = Succeeded with empty diff (no changes)
    • 1 = Error
    • 2 = Succeeded with non-empty diff (changes present)
  • -generate-config-out=PATH - (Experimental) If import blocks are present in configuration, instructs OpenTofu to generate HCL for any imported resources not already present. The configuration is written to a new file at PATH, which must not already exist, or OpenTofu will error. If the plan fails for another reason, OpenTofu may still attempt to write configuration.
  • -input=false - Disables OpenTofu's default behavior of prompting for input for root module input variables that have not otherwise been assigned a value. This option is particularly useful when running OpenTofu in non-interactive automation systems.

  • -json - Enables the machine readable JSON UI output. This implies -input=false, so the configuration must have no unassigned variable values to continue.

  • -lock=false - Don't hold a state lock during the operation. This is dangerous if others might concurrently run commands against the same workspace.

  • -lock-timeout=DURATION - Unless locking is disabled with -lock=false, instructs OpenTofu to retry acquiring a lock for a period of time before returning an error. The duration syntax is a number followed by a time unit letter, such as "3s" for three seconds.

  • -no-color - Disables terminal formatting sequences in the output. Use this if you are running OpenTofu in a context where its output will be rendered by a system that cannot interpret terminal formatting.

  • -concise - Displays plan output in a concise way. It skips showing the refreshing log lines.

  • -out=FILENAME - Writes the generated plan to the given filename in an opaque file format that you can later pass to tofu apply to execute the planned changes, and to some other OpenTofu commands that can work with saved plan files.

    OpenTofu will allow any filename for the plan file, but a typical convention is to name it tfplan. Do not name the file with a suffix that OpenTofu recognizes as another file format; if you use a .tf suffix then OpenTofu will try to interpret the file as a configuration source file, which will then cause syntax errors for subsequent commands.

    The generated file is not in any standard format intended for consumption by other software, but the file does contain your full configuration, all of the values associated with planned changes, and all of the plan options including the input variables. If your plan includes any sort of sensitive data, even if obscured in OpenTofu's terminal output, it will be saved in cleartext in the plan file. You should therefore treat any saved plan files as potentially-sensitive artifacts.

  • -parallelism=n - Limit the number of concurrent operations as OpenTofu walks the graph. Defaults to 10.

For configurations using the local backend only, tofu plan accepts the legacy command line option -state.

Passing a Different Configuration Directory

If your workflow relies on overriding the root module directory, use the -chdir global option instead, which works across all commands and makes OpenTofu consistently look in the given directory for all files it would normally read or write in the current working directory.