There are no specific log messages seen during this issue. This issue may be seen when a workflow is moving a file or directory to a child or ancestor directory and another workflow is trying to open or lookup the child directory simultaneously.
Example: The issue may be seen during log rotation.
This issue is caused due to race condition between open and rename operation when the lock ordering of open and rename are different.
Issue can be seen when a file is being moved from a child to a parent directory (or vice versa), while another thread is concurrently attempting to open the child directory. In addition to the simultaneous operation by two different work flow, the lock ordering in rename is important for the issue to occur. Hence the issue will not be seen every time a file movement and open happening in parallel.
This issue is resolved in VMware ESXi 7.0 Update 3q and VMware ESXi 8.0 Update 2b.
Workaround:
To workaround the issue, please reboot the host.