ESXi boxes upgraded from previous version might not have a big enough core dump partition, or a user error in the installation might have created that partition too small or none at all.

When ESXi host crashes, you end up with either an incomplete or no core dumps for analysis.

1. SSH into the ESXi host where were are going to perform this opertion
2. Add a new file to be used to core dumps:

   esxcli system coredump file add -d <DATASTORE_NAME> -f <FILENAME> -s <SIZE>

3. Get a list of all dump files that we have access to:

   esxcli system coredump file list

4. Select and set which dumpfile we will use for this host:

   esxcli system coredump file set -p /vmfs/volumes/<DATASTORE_NAME>/<FILENAME>

5. Activate/enable this file:

   esxcli system coredump file set -e=1

Example:

esxcli system coredump file add -d iSCSI -f `hostname -s`

esxcli system coredump file list

esxcli system coredump file set -p /vmfs/volumes/52bb4bb5-6a004c90-7581-005056032956/vmkdump/esxi-55-b.dumpfile

esxcli system coredump file set -e=1