PowerFlex SDR failover mounts volume RO in destination Oracle linux server

SDR failover is successful and when the user observes the volume in PfxM at the destination it does show Read Only = No which means PowerFlex has the volume as RW but the multipath -ll command output shows RO in the linux OS. 

Oracle Linux OS has the correct settings in multipath.conf to exclude PowerFlex scini devices from the blacklist, but at the time of the failover the multipath service in the OS sees them as RO. User can see this in multipath -ll command output.

However, if the user check the RO status of the block device with the command 'cat /sys/block/sciniX/ro' where X is the device letter the user see it returns 0 for output meaning Read Only = falae. That means the block device in the linux OS is RW and it is a multipath that is out of sync.

To update the status, the multipathd must be restarted. Any further questions around this process must be addressed by Oracle support.

Now the multipath -ll command output shows updated state to match the scini block device in linux OS and the state of the volume on the PowerFlex array.

See Oracle KB https://support.oracle.com/knowledge/Oracle%20Linux%20and%20Virtualization/2660512_1.html However, the reload_readwrite yes option is multipatyh.conf is deprecated in OEL 8 and not needed.