ESXi 5.x uses a new log infrastructure. If your host is deployed with Auto Deploy, or if you set up a log directory separate from the default location in a scratch directory on the VMFS volume, you might need to change your current log size and rotation settings to ensure that enough space for system logging exists.

All vSphere components use this infrastructure. The default values for log capacity in this infrastructure vary, depending on the amount of storage available and on how you have configured system logging. Hosts that are deployed with Auto Deploy store logs on a RAM disk, which means that the amount of space available for logs is small.

If your host is deployed with Auto Deploy, reconfigure your log storage in one of the following ways:

  • Redirect logs over the network to a remote collector.

  • Redirect logs to a NAS or NFS store.

You might also want to reconfigure log sizing and rotations for hosts that are installed to disk, if you redirect logs to nondefault storage, such as a NAS or NFS store.

You do not need to reconfigure log storage for ESXi hosts that use the default configuration, which stores logs in a scratch directory on the VMFS volume. For these hosts, ESXi 5.x autoconfigures logs to best suit your installation, and provides enough space to accommodate log messages.

Table 1. Recommended Minimum Size and Rotation Configuration for hostd, vpxa, and fdm Logs.


Maximum Log File Size

Number of Rotations to Preserve

Minimum Disk Space Required

Management Agent (hostd)




VirtualCenter Agent (vpxa)




vSphere HA agent (Fault Domain Manager, fdm)




For information about setting up and configuring syslog and a syslog server and installing vSphere Syslog Collector, see the vSphere Installation and Setup documentation.