You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
[Cherry-pick][main] Clarify how to recreate a db using the empty list of seeding servers (#2425) (#2431)
# Backport
This will backport the following commits from `dev` to `main`:
- [Clarify how to recreate a db using the empty list of seeding servers
(#2425)](#2425)
<!--- Backport version: 9.5.1 -->
### Questions ?
Please refer to the [Backport tool
documentation](https://github.com/sorenlouv/backport)
Co-authored-by: NataliaIvakina <[email protected]>
Co-authored-by: Anna Sjerling <[email protected]>
If you provide an empty list of seeding servers and do not specify a `seedURI`, Neo4j automatically selects all available allocations of the database as seeders.
144
-
The store will be replaced by the most up-to-date seeder available in the cluster.
143
+
If you provide an empty list of seeding servers and do not specify a `seedURI`, Neo4j will select all allocations of the database (primaries and secondaries) on currently enabled and non-cordoned servers to act as seeders.
144
+
+
145
+
Before running the procedure, ensure that all unavailable servers are xref:clustering/servers.adoc#_cordoned_state[cordoned]; otherwise, the procedure will fail.
146
+
+
147
+
To determine where the database is allocated, use the `SHOW DATABASES` command.
148
+
To identify all enabled and available servers hosting the required database, run the `SHOW SERVERS` command.
149
+
Servers have to show `health = Available` and `status = Enabled`.
150
+
Cordon all unreachable servers by running the `dbms.cluster.cordonServer()` procedure.
151
+
+
152
+
Then Neo4j will select the most up-to-date seeder available in the cluster to recreate the database.
0 commit comments