Seamless Logical Replication during Fail-over

Hi,

Is there any way to continue logical replication (either in-built or using
pglogical) between a promoted replica and existing subscribers without data
loss?

I could find some discussions about fail-over slots and other mechanisms
for addressing this as part of PostgreSQL 9.6 release. But I am not sure if
these changes made it to any of the releases.

The use-case is as follows:
* Cluster A with a master and two or more replicas using synchronized
streaming replication
* Cluster B containing a node which uses logical replication with Cluster A
master. There may be a lag in updating this node.
* When Cluster A master goes down, a replica is promoted but it does not
have any information about the logical replication slots created in the
master. So, even if a new slot is created on the replica, there will be
data loss on the subscriber side.

Any suggestions on how to address this using builtin mechanisms or using an
extension are welcome.

Thanks,
Murali