Chapter 19. Supported kdump configurations and targets
The kdump mechanism generates crash dump files when kernel crashes occur, providing critical information for root cause analysis. Identify supported configurations and targets, configure kdump, and verify its operation on Red Hat Enterprise Linux systems.
19.1. Memory requirements for kdump Copy linkLink copied to clipboard!
kdump requires permanently reserved system memory for a capture kernel to store crash dumps for analysis. The main kernel cannot access this memory.
The memory requirements vary based on certain system parameters. One of the major factors is the system’s hardware architecture. To identify the exact machine architecture, such as Intel 64 and AMD64, also known as x86_64, and print it to standard output, use the following command:
$ uname -m
With the stated list of minimum memory requirements, you can set the appropriate memory size to automatically reserve a memory for kdump on the latest available versions. The memory size depends on the system’s architecture and total available physical memory.
| Architecture | Available Memory | Minimum Reserved Memory |
|---|---|---|
|
AMD64 and Intel 64 ( | 2 GB to 64 GB | 256 MB of RAM |
| 64 GB and more | 512 MB of RAM | |
| 64-bit ARM (4k pages) | 1 GB to 4 GB | 256 MB of RAM |
| 4 GB to 64 GB | 320 MB of RAM | |
| 64 GB and more | 576 MB of RAM | |
| 64-bit ARM (64k pages) | 1 GB to 4 GB | 356 MB of RAM |
| 4 GB to 64 GB | 420 MB of RAM | |
| 64 GB and more | 676 MB of RAM | |
|
IBM Power Systems ( | 2 GB to 4 GB | 384 MB of RAM |
| 4 GB to 16 GB | 512 MB of RAM | |
| 16 GB to 64 GB | 1 GB of RAM | |
| 64 GB to 128 GB | 2 GB of RAM | |
| 128 GB and more | 4 GB of RAM | |
|
IBM Z ( | 2 GB to 64 GB | 256 MB of RAM |
| 64 GB and more | 512 MB of RAM |
On many systems, kdump is able to estimate the amount of required memory and reserve it automatically. This behavior is enabled by default, but only works on systems that have more than a certain amount of total available memory, which varies based on the system architecture.
The automatic configuration of reserved memory based on the total amount of memory in the system is a best effort estimation. The actual required memory might vary due to other factors such as I/O devices. Not using enough memory might cause debug kernel unable to boot as a capture kernel in the case of kernel panic. To avoid this problem, increase the crash kernel memory sufficiently.
19.2. Minimum threshold for automatic memory reservation Copy linkLink copied to clipboard!
The kdump-utils utility automatically reserves memory for kdump by using the crashkernel parameter. However, this requires a minimum amount of system memory, which varies by architecture. If your system’s memory is less than the threshold, you must configure the reservation manually.
| Architecture | Required Memory |
|---|---|
|
AMD64 and Intel 64 ( | 2 GB |
|
IBM Power Systems ( | 2 GB |
|
IBM Z ( | 2 GB |
| 64-bit ARM | 2 GB |
The crashkernel=1G-4G:192M,4G-64G:256M,64G:512M option in the boot command line is no longer supported from RHEL 10.
19.3. Supported kdump targets Copy linkLink copied to clipboard!
When a kernel crash occurs, the operating system saves the dump file on the configured or default target location. You can save the dump to a device, as a file on a local file system, or over a network. The following list shows which kdump targets Red Hat Enterprise Linux currently supports or does not support.
| Target type | Supported Targets | Unsupported Targets |
|---|---|---|
| Physical Storage |
|
|
| Network |
|
|
| Hypervisor |
| |
| Filesystem |
The |
The |
| Firmware |
|
19.4. Supported kdump filtering levels Copy linkLink copied to clipboard!
To reduce dump file size, kdump uses makedumpfile to compress data and exclude unwanted information. See the table for supported filtering levels, such as excluding hugepages.
| Option | Description |
|---|---|
|
| Zero pages |
|
| Cache pages |
|
| Cache private |
|
| User pages |
|
| Free pages |
19.5. Supported default failure responses Copy linkLink copied to clipboard!
By default, when kdump fails to create a core dump, the operating system reboots. However, you can configure kdump to perform a different operation in case it fails to save the core dump to the primary target.
dump_to_rootfs- Attempt to save the core dump to the root file system. This option is especially useful in combination with a network target: if the network target is unreachable, this option configures kdump to save the core dump locally. The system is rebooted afterwards.
reboot- Reboot the system, losing the core dump in the process.
halt- Halt the system, losing the core dump in the process.
poweroff- Power off the system, losing the core dump in the process.
shell- Run a shell session from within the initramfs, allowing the user to record the core dump manually.
final_action-
Enable additional operations such as
reboot,halt, andpoweroffactions after a successfulkdumpor whenshellordump_to_rootfsfailure action completes. The defaultfinal_actionoption isreboot. failure_action-
Specifies the action to perform when a dump might fail in case of a kernel crash. The default
failure_actionoption isreboot.
19.6. Using final_action parameter Copy linkLink copied to clipboard!
To perform operations such as reboot, halt, or poweroff after kdump succeeds or fails, use the final_action parameter. The default action is reboot.
Procedure
To configure
final_action, edit the/etc/kdump.conffile and add one of the following options:-
final_action reboot -
final_action halt -
final_action poweroff
-
Restart the
kdumpservice for the changes to take effect.# kdumpctl restart
19.7. Using failure_action parameter Copy linkLink copied to clipboard!
To specify the action when a crash dump fails, use the failure_action parameter. The default action is reboot.
The parameter recognizes the following actions to take:
reboot- Reboots the system after a dump failure.
dump_to_rootfs- Saves the dump file on a root file system when a non-root dump target is configured.
halt- Halts the system.
poweroff- Stops the running operations on the system.
shell-
Starts a shell session inside
initramfs, from which you can manually perform additional recovery actions.
Procedure
To configure an action to take if the dump fails, edit the
/etc/kdump.conffile and specify one of thefailure_actionoptions:-
failure_action reboot -
failure_action halt -
failure_action poweroff -
failure_action shell -
failure_action dump_to_rootfs
-
Restart the
kdumpservice for the changes to take effect:# kdumpctl restart