How to use DB geo-replication in azure?

Question

Database

Hello everyone
After yesterday's problems with the availability of the database in openwork, it was decided to increase the fault tolerance of the database through geo-replication.
Geo-replication was done through the azure control panel, another database appeared with the "Secondary" replication role, the one that was before - "Primary", two connected points appeared on the map.
But, the load between the bases is not distributed. And the state of the secondary database is "Read-only".
In general, here are the questions:
1. Yesterday, in 90% of cases of accessing the database, the connection fell by timeout, sometimes the database was not available at all. Will geo-replication help in this case? At the same time, in the control panel for the first half hour it was written that there were no problems on the equipment. I suspect that there must be some kind of single point of access to the database, or the system working with the database must itself know that if this database responds more than the timeout specified, then you need to switch to the second database. How is it implemented? I haven't been able to find clear tutorials on this topic.
2. How to use all replicated databases to improve performance?