kexec -p does not work consistently

[Sushmita Ghan]

Hi,

I am trying to setup kdump on my Linux(x86 machine),
where I am using kexec-tools-2.0.22.
Reserving 512M of memory of production kernel for crash dump kernel.

# dmesg | grep crash
[    0.000000] Command line: BOOT_IMAGE=/vmlinuz-4.19.213-pc64
console=tty0 console=ttyS0,115200 acpi_osi=Linux intel_iommu=on
iommu=pt crashkernel=512M root=/dev/VG_System/sysimg
[    0.006893] Reserving 512MB of memory at 352MB for crashkernel
(System RAM: 16352MB)
[    0.319548] Kernel command line: BOOT_IMAGE=/vmlinuz-4.19.213-pc64
console=tty0 console=ttyS0,115200 acpi_osi=Linux intel_iommu=on
iommu=pt crashkernel=512M root=/dev/VG_System/sysimg


Using below options with kexec -p:

kexec -p /mnt/images/boot$BOOT_IMAGE --reuse-cmdline
--append="maxcpus=1 reset_devices irqpoll".

Here in my case sometimes it successfully loads second kernel, but
sometimes it gets stuck while loading second kernel(crash dump
kernel).

According to below document,
https://www.kernel.org/doc/html/latest/admin-guide/kdump/kdump.html

Following are the arch specific command line options to be used while
loading dump-capture kernel.

For i386, x86_64 and ia64:

“1 irqpoll nr_cpus=1 reset_devices”

With above parameters system gets stuck while loading crash dump
kernel.This behavior is inconsistent.
But when tested without "irqpoll" parametere, it seems working consistently.

Is this sysctl parameter changes required with new kexec version
or any pointers regarding this issue?

Can I use only provided below options for crash dump kernel?

# kexec -p /boot/vmlinuz-4.19.213-pc64 --reuse-cmdline
--append="maxcpus=1 reset_devices"

Any clue would be a great help for me.

Thanks & regards,
Sushmita