During a host failure with a purple diagnostic screen, VMware ESXi/ESX attempts to save diagnostic information to one or more pre-configured locations. The available target locations and the methods to configure the host differ between versions of ESXi/ESX.
If the host is configured with a location for saving a coredump, but does not successfully record a coredump during a purple diagnostic screen, there are issues with either the storage medium or the connection to the storage medium. Configure an alternate target for saving the coredump to successfully capture diagnostic information from a subsequent outage.
- Network Dump Collection
ESXi 5.x/6.0 hosts dump diagnostic information to a Network Dump Collector service through the network. For more information, see:
- Disk Dump Collection
ESXi/ESX 3.0-6.0 hosts dump diagnostic information to a partition on disk, including local or remote block devices. For more information, see:
- ESX 3.0-4.1 also generates a coredump from the service console Linux kernel. The VMKcore diagnostic coredump partition is used to store the VMkernel coredump, but the service console coredump is placed on a VMFS datastore. Depending on the type of failure, a coredump is generated from one or both components. For more information, see Configuring an ESX host to capture a Service Console coredump (1032962).
- ESXi 5.5 introduces dump to file. For more information, see Configuring ESXi coredump to file instead of partition (2077516). This feature is introduced to remove the 100MB limitation which most of the time causes the dump to be truncated. Using the automatic sizing is the recommended way to configure dump to file and never results in a truncated dump.
For translated versions of this article, see: