Vmkdump

From VI-Toolkit
Revision as of 11:26, 23 July 2009 by Wila (talk | contribs) (added category)
Jump to navigation Jump to search
NAME
     vmkdump - VMkernel dumper

SYNOPSIS
     vmkdump OPTIONS

COPYRIGHT
    VMware ESX Server is Copyright 2004-2007 VMware, Inc.  All rights reserved.

DESCRIPTION
     vmkdump manages the VMkernel dump partition.

Usage

     vmkdump -d --devname /dev/<Linux device name>
             -z --zdumpname outputName (optional)
             -n --newonly (optional)
             -s --slot slotnum (optional)
  Extract the log from a compressed dump:
     vmkdump -l --log zdumpname


OPTIONS

    -d, --devname deviceName
        Specify the Linux console device that should be used to read the  VMkernel dump.  
        Note: This option should not be used after the dump  partition has been set with the -p option.  
        If the partition has been set, you need to stop and reset the VMkernel before using this option.

    -n, --newonly
        For each VMkernel dump that is produced, a count is kept of the number of times the compressed 
        dump file has been read off the dump partition.  When this option is specified, the compressed 
        dump will only be read off the dump partition if the read count is zero.
    -z, --zdumpname compressedFileName
        When reading from the dump partition, this option specifies the name of the file to create with 
        the compressed dump.  By default the name is vmkernel-zdump.#, where # is a unique number.
    -s, --slot slotnum
        On dump partitions that reside on shared storage, multiple machines  can share the dump partition.  
        Because of this, the dump partition is split up into multiple slots.  Generally, when retrieving a 
        core dump, the physical machine's UUID is hashed into a slot number based on the total number 
        of slots, and the compressed dump is automatically read from that slot.  

        However, there may be situations where the machine that dumped core is no longer able to retrieve 
        core  dumps.  In those cases, this option can be used to override the normal UUID hash algorithm 
        for determining a slot.  Note that the slot used during a core dump will be displayed on the PSOD.
 
    -l, --log compressedFileName
       Extracts the vmkernel log from the specified compressed dump file.


Examples

  • Specify linux device to read the dump
#vmkdump -d /dev/sda5

Creates a file name vmkernel-zdump. # contained a compressed dump. # is 1 if no other compressed dumps are in the current directory, otherwise # is the lowest unused integer (ie, if up to vmkernel-zdump.5 exist, then it would create vmkernel-zdump.6).


  • Extract the zdump from the dump partition
#vmkdump -d /vmfs/devices/disks/mpx.vmhba1:C0:T0:L0:2

singleslot coredump


created file vmkernel-zdump.1