KDUMP crashkernel set to 128 MB when Storage Foundation is installed on RHEL6 U1
Article ID: 100025507
Updated On:
Description
Error Message
/etc/init.d/kdump restart
Stopping kdump: [ OK ]
Detected change(s) the following file(s):
/etc/kdump.conf
Rebuilding /boot/initrd-2.6.32-131.0.15.el6.x86_64kdump.img
Your running kernel is using more than 70% of the amount of space you reserved for kdump, you should consider increasing your crash kernel r[WARNING]n
Starting kdump: [ OK ]
Cause
https://bugzilla.redhat.com/show_bug.cgi?id=722879
Kdump will create a ramdisk that will include all modules loaded on the system ( including VxVM and VxFS modules). Due to all modules being included in the ramdisk image and the size of crash kernel being 128 MB, will result in a warning during kdump startup|restart during the ramdisk build phase.
Kdump should not include modules that are not required to boot the kdump kernel. The only time Storage Foundation modules would be included in the kdump ramdisk is when the boot disk is encapsulated under VxVM control. Currently there is no way to omit VxVM|VxFS modules through command line arguments when running mkinitrd or mkdumprd.
Redhat bugzilla id 722879 has been opened for this issue.
Resolution
We recommend customer’s configure crash kernel parameter with 256 MB so that all modules will fit in the new ramdisk that kdump creates and so there is enough memory to allocate. This will also avoid the kdump warning when kdump is configured and failure when trying to boot into the kdump kernel.
Applies To
RHEL6 U1
Storage Foundation 5.1SP1PR2P1
Issue/Introduction
