VMs residing on the affected datastore may experience hung I/O, causing applications, databases, and operating systems to freeze or crash.
If storage saturation is causing the timeouts, throttle the execution queues on your hosts. Lowering the maximum number of concurrent commands allowed per LUN prevents the array from becoming overwhelmed during high-load events. Fall Back to Standard Locking (If Necessary) VMs residing on the affected datastore may experience
When a host wants to lock a metadata block on a shared datastore, it sends an ATS command (specifically the SCSI COMPARE AND WRITE command) to the storage array. Fall Back to Standard Locking (If Necessary) When
The primary source of information is the vmkernel.log file on the ESXi host. Look for messages containing ATS Miscompare , Cmd 0x89 , or the exact error phrase. Pay attention to the timing and the specific device ( naa.xxx ) involved. Pay attention to the timing and the specific device ( naa