This issue is due to the TCP stream between the rhttpproxy and hostd. The rhttpproxy process resets the connection and RetrieveProperties code is called every 5 minutes, causing a log entry to appear every 5 minutes.
When a call is made to hostd to invoke cancelWaitForUpdates, the PendingRequest is not cleanly removed. When the next PendingRequest occurs after cancelWaitForUpdates, this results in a Connection reset by peer error. You can ignore this error.
Resolution
This is a known issue affecting VMware ESXi 5.0.x, 5.1.x, 5.5.x. and 6.x.
Currently, there is no resolution or workaround.
Additional Information
To be alerted when this article is update, click Subscribe to Article in the action box.