When attempting to restart the kdump service the following error is shown: "no record for 'mapper/mpath0p2' in database"

Solution Verified - Updated -

Environment

  • Red Hat Enterprise Linux 5
  • Multipathing

Issue

  • When attempting to restart the kdump service the following error is shown:
# service kdump restart
Stopping kdump:                                            [  OK  ]
No kdump initial ramdisk found.                            [WARNING]
Rebuilding /boot/initrd-2.6.18-274.12.1.el5kdump.img
no record for 'mapper/360000970000298700361533030374444p2' in database
Starting kdump:                                            [  OK  ]

Resolution

  • Adjust the lvm configuration to point to /dev/mpath instead of /dev/mapper

Root Cause

  • The warning is output by udevinfo when the given device is not in the udev database, meaning that it's not created by udevd directly. For such device, it cannot get any information with udevinfo. Kdump uses udevinfo to build the device paths, thus the reason for this error being thrown.

Diagnostic Steps

  • According to the filter defined in /etc/lvm/lvm.conf, the device file in /dev/mapper is preferable in output of the vgdisplay -v.
    # By default we accept every block device:
    #filter = [ "a/.*/" ]
    filter = [ "a|/dev/mapper/.*|", "r|.*|" ]

But, udevinfo works with the file (symlink) in /dev/mpath.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.

Comments