Start keeper asynchronously in some cases

[alesapin]

Changelog category (leave one):

• Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
Now Keeper (as part of clickhouse-server) will start asynchronously if it can connect to some other node.

[qoega]

And what is the use case?

[alesapin]

@Mergifyio update

[mergify]

Command update: success

Branch has been successfully updated

[alesapin]

?

When you start fresh empty replica it's better to connect to some other node and receive snapshot/logs from leader asynchronously.

[alesapin]

@Mergifyio update

[mergify]

Command update: success

Branch has been successfully updated

[tavplubix]

Integration tests (release) - test_keeper_two_nodes_cluster/test.py::test_read_write_two_nodes_with_blocade

[alesapin]

Integration tests (release) - test_keeper_two_nodes_cluster/test.py::test_read_write_two_nodes_with_blocade

Actually looks flaky and unrelated to changes.

[alesapin]

Checked logs from test_keeper_two_nodes_cluster/test.py::test_read_write_two_nodes_with_blocade. Looks like PartitionManager just didn't partition network for some reason: https://gist.github.com/alesapin/9105c15fcbdc79e50e0045487d7ac43e

Or it's a wrong log.