Your software release might not support all the features documented in this module. For the latest caveats and feature information, see the Bug Search Tool at https://tools.cisco.com/bugsearch/ and the release notes for your software release. To find information about the features documented in this module, and to see a list of the releases in which each feature is supported, see the "New and Changed Information"chapter or the Feature History table in this chapter.

A rollback allows you to take a snapshot, or user checkpoint, of the Cisco NX-OS configuration and then reapply that configuration to your device at any point without having to reload the device. A rollback allows any authorized administrator to apply this checkpoint configuration without requiring expert knowledge of the features configured in the checkpoint.

Cisco NX-OS automatically creates system checkpoints. You can use either a user or system checkpoint to perform a rollback.

You can create a checkpoint copy of the current running configuration at any time. Cisco NX-OS saves this checkpoint as an ASCII file which you can use to roll back the running configuration to the checkpoint configuration at a future time. You can create multiple checkpoints to save different versions of your running configuration.

When you roll back the running configuration, you can trigger the following rollback types:

• atomic—Implement a rollback only if no errors occur.
• best-effort—Implement a rollback and skip any errors.
• stop-at-first-failure—Implement a rollback that stops if an error occurs.

The default rollback type is atomic.

When you are ready to roll back to a checkpoint configuration, you can view the changes that will be applied to your current running configuration before committing to the rollback operation. If an error occurs during the rollback operation, you can choose to cancel the operation, or ignore the error and proceed with the rollback. If you cancel the operation, Cisco NX-OS provides a list of changes already applied before the error occurred. You need to clean up these changes manually.

To configure rollback, you must have network-admin user privileges.

Rollbacks have the following configuration guidelines and limitations:

• You can create up to ten checkpoint copies.

• Your checkpoint filenames must be 80 characters or less.

• You cannot apply a checkpoint configuration in a nondefault VDC if there is a change in the global configuration portion of the running configuration compared to the checkpoint configuration.

• Your checkpoint filenames must be 80 characters or less.

• You cannot start a checkpoint filename with the word system .

• Beginning in Cisco NX-OS Release 4.2(1), you can start a checkpoint filename with the word auto .

• Beginning in Cisco NX-OS Release 4.2(1), you can name a checkpoint file summary or any abbreviation of the word summary .

• Only one user can perform a checkpoint, rollback, or copy the running configuration to the startup configuration at the same time.

• After the system executes the write erase or reload command, checkpoints are deleted. You can use the clear checkpoint database command to clear out all checkpoint files.

• A rollback fails for NetFlow if during a rollback, you try to modify a record that is programmed in the hardware.

• Although a rollback is not supported for checkpoints across software versions, users can perform a rollback at their own discretion and can use the best-effort mode to recover from errors.

• When checkpoints are created on bootflash, differences with the running-system configuration cannot be performed before performing the rollback, and the system reports “No Changes.”

• Checkpoints are local to a virtual device context (VDC).

• Checkpoints created using the checkpoint and checkpoint checkpoint_name commands are present upon a switchover.

• Checkpoints created in the default VDC are present upon reload unless a write-erase command is issued before a reload.

• Checkpoints created in nondefault VDCs are present upon reload only if a copy running-config startup-config command is issued in the applicable VDC and the default VDC.

• A rollback to files on bootflash is supported only on files created using the checkpoint checkpoint_name command and not on any other type of ASCII file.

• Checkpoint names must be unique. You cannot overwrite previously saved checkpoints with the same name.

• Rollback is not supported in the storage VDC.

• Rollback is not supported in the Admin virtual device context (VDC) feature.

• Configure the terminal dont-ask command before executing the rollback command to a checkpoint. In a rollback patch, the rollback process does not pause for user interaction and takes the default values for interactive commands. Configuring the terminal dont-ask command before executing the rollback command helps in resolving this issue.

• Rollback is not supported in the context of auto configurations. Checkpoints do not store auto configurations. Therefore, after a rollback is performed, the corresponding auto configurations will not be present.

• When you perform rollback, if the patch contains the reload command for the corresponding module along with the configuration commands for that module, rollback fails. This is because the rollback action does not wait for the module to come online; it starts executing the configuration commands on the module even as the reload process is in progress. To resolve this issue, manually execute the configuration commands for the module after the module is online.

  Examples:

  • A rollback fails when you execute the bfd hw-offload-module command or the no form of this command. In this instance, failure is because rollback cannot execute these commands when the switch interfaces that are a part of the BFD sessions are powered up. To resolve this issue, shut down all the interfaces that are a part of the BFD sessions using the shutdown command before executing the bfd hw-offload-module command or the no form of this command.

  • The following BFD command configurations are not supported during a rollback configuration:

    • bfd {ipv4 | ipv6} echo

    • bfd {ipv4 | ipv6} per-link

    • bfd hw-offload-module module-number

    • port-channel bfd track-member-link

    • port-channel bfd destination destination-ip-address

  • When an FEX is being configured while a rollback vPC is applied to an interface, the FEX goes offline momentarily. When this occurs, rollback does not wait for the FEX to come online, and executes the configuration commands for the interface, resulting in failure because the corresponding FEX is not yet provisioned. To resolve this issue, manually execute the FEX-related configuration commands after the FEX is online.

• Checkpoint descriptions are not persistent across switch reloads. When a description for a checkpoint is created by using the checkpoint description command, the description is not visible in the output of the show checkpoint summary command after the switch is reloaded. If the checkpoint description can be qualified as a checkpoint name, we recommend using the same alphanumeric string for both the checkpoint name and description. The checkpoint name is visible in the output of the show checkpoint summary command even after the switch is reloaded

This table lists the default settings for rollback parameters.

To display the rollback configuration information, perform one of the following tasks:

Use the clear checkpoint database command to delete all checkpoint files.

Note	When a checkpoint is created, you can view the default configuration priority-flow-control mode auto using the show run all command. You cannot view the configuration priority-flow-control mode auto using the show run command for the interface.

This example shows how to create a checkpoint file and then implements a best-effort rollback to a user checkpoint name:

checkpoint stable
rollback running-config checkpoint stable best-effort