Prerequisites - Amazon Elastic Compute Cloud
Amazon Web Services RegionsWavelength Zones and Amazon OutpostsLocal ZonesInstance typesGrant accessSupport for browser-based clientInstance stateAmazon EC2 Systems Managersshd serverConfigure your chosen troubleshooting tool
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Prerequisites

Amazon Web Services Regions

Supported in all Amazon Web Services Regions except Canada West (Calgary).

Wavelength Zones and Amazon Outposts

Not supported.

Local Zones

Supported in all Local Zones.

Instance types

Supported instance types:

  • All virtualized instances built on the Nitro System.

  • All bare metal instances except:

    • General purpose: a1.metal, mac1.metal, mac2.metal

    • Accelerated computing: g5g.metal

    • Memory optimized: u-6tb1.metal, u-9tb1.metal, u-12tb1.metal, u-18tb1.metal, u-24tb1.metal

Grant access

You must complete the configuration tasks to grant access to the EC2 Serial Console. For more information, see Configure access to the EC2 Serial Console.

Support for browser-based client

To connect to the serial console using the browser-based client, your browser must support WebSocket. If your browser does not support WebSocket, connect to the serial console using your own key and an SSH client.

Instance state

Must be running.

You can't connect to the serial console if the instance is in the pending, stopping, stopped, shutting-down, or terminated state.

For more information about the instance states, see Instance lifecycle.

Amazon EC2 Systems Manager

If the instance uses Amazon EC2 Systems Manager, then SSM Agent version 3.0.854.0 or later must be installed on the instance. For information about SSM Agent, see Working with SSM Agent in the Amazon Systems Manager User Guide.

sshd server

You do not need an sshd server installed or running on your instance.

Configure your chosen troubleshooting tool

To troubleshoot your Linux instance via the serial console, you can use GRUB or SysRq. Before you can use these tools, you must first perform configuration steps on every instance on which you'll use them.

For the instructions to configure your chosen troubleshooting tool on Windows, see Configure your chosen troubleshooting tool in the Amazon EC2 User Guide for Windows Instances.

Configure GRUB

Before you can use GRUB via the serial console, you must configure your instance to use GRUB via the serial console.

To configure GRUB, choose one of the following procedures based on the AMI that was used to launch the instance.

Amazon Linux 2
To configure GRUB on an Amazon Linux 2 instance

  1. Connect to your Linux instance

  2. Add or change the following options in /etc/default/grub:

    • Set GRUB_TIMEOUT=1.

    • Add GRUB_TERMINAL="console serial".

    • Add GRUB_SERIAL_COMMAND="serial --speed=115200".

    The following is an example of /etc/default/grub. You might need to change the configuration based on your system setup.

    GRUB_CMDLINE_LINUX_DEFAULT="console=tty0 console=ttyS0,115200n8 net.ifnames=0 biosdevname=0 nvme_core.io_timeout=4294967295 rd.emergency=poweroff rd.shell=0"
GRUB_TIMEOUT=1
GRUB_DISABLE_RECOVERY="true"
GRUB_TERMINAL="console serial"
GRUB_SERIAL_COMMAND="serial --speed=115200"

  3. Apply the updated configuration by running the following command.

    [ec2-user ~]$ sudo grub2-mkconfig -o /boot/grub2/grub.cfg
Ubuntu
To configure GRUB on an Ubuntu instance

  1. Connect to your instance.

  2. Add or change the following options in /etc/default/grub.d/50-cloudimg-settings.cfg:

    • Set GRUB_TIMEOUT=1.

    • Add GRUB_TIMEOUT_STYLE=menu.

    • Add GRUB_TERMINAL="console serial".

    • Remove GRUB_HIDDEN_TIMEOUT.

    • Add GRUB_SERIAL_COMMAND="serial --speed=115200".

    The following is an example of /etc/default/grub.d/50-cloudimg-settings.cfg. You might need to change the configuration based on your system setup.

    # Cloud Image specific Grub settings for Generic Cloud Images
# CLOUD_IMG: This file was created/modified by the Cloud Image build process

# Set the recordfail timeout
GRUB_RECORDFAIL_TIMEOUT=0

# Do not wait on grub prompt
GRUB_TIMEOUT=1
GRUB_TIMEOUT_STYLE=menu

# Set the default commandline
GRUB_CMDLINE_LINUX_DEFAULT="console=tty1 console=ttyS0 nvme_core.io_timeout=4294967295"

# Set the grub console type
GRUB_TERMINAL="console serial"
GRUB_SERIAL_COMMAND="serial --speed 115200"

  3. Apply the updated configuration by running the following command.

    [ec2-user ~]$ sudo update-grub
RHEL
To configure GRUB on a RHEL instance

  1. Connect to your instance.

  2. Add or change the following options in /etc/default/grub:

    • Remove GRUB_TERMINAL_OUTPUT.

    • Add GRUB_TERMINAL="console serial".

    • Add GRUB_SERIAL_COMMAND="serial --speed=115200".

    The following is an example of /etc/default/grub. You might need to change the configuration based on your system setup.

    GRUB_TIMEOUT=1
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_CMDLINE_LINUX="console=tty0 console=ttyS0,115200n8 net.ifnames=0 rd.blacklist=nouveau nvme_core.io_timeout=4294967295 crashkernel=auto"
GRUB_DISABLE_RECOVERY="true"
GRUB_ENABLE_BLSCFG=true
GRUB_TERMINAL="console serial"
GRUB_SERIAL_COMMAND="serial --speed=115200"

  3. Apply the updated configuration by running the following command.

    [ec2-user ~]$ sudo grub2-mkconfig -o /boot/grub2/grub.cfg
CentOS

For instances that are launched using a CentOS AMI, GRUB is configured for the serial console by default.

The following is an example of /etc/default/grub. Your configuration might be different based on your system setup.

GRUB_TIMEOUT=1
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL="serial console"
GRUB_SERIAL_COMMAND="serial --speed=115200"
GRUB_CMDLINE_LINUX="console=tty0 crashkernel=auto console=ttyS0,115200"
GRUB_DISABLE_RECOVERY="true"

Configure SysRq

To configure SysRq, you enable the SysRq commands for the current boot cycle. To make the configuration persistent, you can also enable the SysRq commands for subsequent boots.

To enable all SysRq commands for the current boot cycle

  1. Connect to your instance.

  2. Run the following command.

    [ec2-user ~]$ sudo sysctl -w kernel.sysrq=1
    Note

    This setting will clear on the next reboot.

To enable all SysRq commands for subsequent boots

  1. Create the file /etc/sysctl.d/99-sysrq.conf and open it in your favorite editor.

    [ec2-user ~]$ sudo vi /etc/sysctl.d/99-sysrq.conf

  2. Add the following line.

    kernel.sysrq=1

  3. Reboot the instance to apply the changes.

    [ec2-user ~]$ sudo reboot

  4. At the login prompt, enter the user name of the password-based user that you set up previously, and then press Enter.

  5. At the Password prompt, enter the password, and then press Enter.