The GOS timeout scripts are available to ensure that the disk timeout settings within a VM can "wait" the same amount of time for I/O to resume in the case of any break in communication on the underlying host - for example, during a takeover or giveback operation on the controller. While the NFS settings will ensure that the physical ESX server waits an appropriate amount of time before disconnecting the storage - the GOS scripts will do the same for the OS running inside of the VM, for each SCSI disk. The different times are available depending on which version of ESX/VSC/ONTAP/GOS is being run, and typically the 190s is only required for older combinations.
I also check this KB and this KB subject is "What are the guest OS tunings needed for a VMware SAN deployment?"
I think we have to face same scenario when use NFS datastore. so how 60-seconds GOS timeout setting work with ESXi 125 seconds NFS setting? the reason that I raised this case is for how to define cDOT volume move cut-over windows. by default, the volume move cut-over window is 45s, but in some case, 45 seconds is not enough. according the ESXi NFS setting, we could enlarger it to <125 second but not sure how it impact GOS.