Add disclaimer about usage for pagination.

docs/development/core/server/kibana-plugin-core-server.savedobjectsclient.createpointintimefinder.md

@@ -4,7 +4,9 @@
 
 ## SavedObjectsClient.createPointInTimeFinder() method
 
-Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any find queries that might return more than 1000 saved objects.
+Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any `find` queries that might return more than 1000 saved objects, however this API is only intended for use in server-side "batch" processing of objects where you are collecting all objects in memory or streaming them back to the client.
+
+Do NOT use this API in a route handler to facilitate paging through saved objects on the client-side unless you are streaming all of the results back to the client at once. Because the returned generator is stateful, you cannot rely on subsequent http requests retrieving new pages from the same Kibana server in multi-instance deployments.
 
 The generator wraps calls to [SavedObjectsClient.find()](./kibana-plugin-core-server.savedobjectsclient.find.md) and iterates over multiple pages of results using `_pit` and `search_after`<!-- -->. This will open a new Point-In-Time (PIT), and continue paging until a set of results is received that's smaller than the designated `perPage`<!-- -->.
 

docs/development/core/server/kibana-plugin-core-server.savedobjectsclient.md

@@ -32,7 +32,7 @@ The constructor for this class is marked as internal. Third-party code should no
 |  [checkConflicts(objects, options)](./kibana-plugin-core-server.savedobjectsclient.checkconflicts.md) |  | Check what conflicts will result when creating a given array of saved objects. This includes "unresolvable conflicts", which are multi-namespace objects that exist in a different namespace; such conflicts cannot be resolved/overwritten. |
 |  [closePointInTime(id, options)](./kibana-plugin-core-server.savedobjectsclient.closepointintime.md) |  | Closes a Point In Time (PIT) by ID. This simply proxies the request to ES via the Elasticsearch client, and is included in the Saved Objects Client as a convenience for consumers who are using [SavedObjectsClient.openPointInTimeForType()](./kibana-plugin-core-server.savedobjectsclient.openpointintimefortype.md)<!-- -->.<!-- -->Only use this API if you have an advanced use case that's not solved by the [SavedObjectsClient.createPointInTimeFinder()](./kibana-plugin-core-server.savedobjectsclient.createpointintimefinder.md) method. |
 |  [create(type, attributes, options)](./kibana-plugin-core-server.savedobjectsclient.create.md) |  | Persists a SavedObject |
-|  [createPointInTimeFinder(findOptions, dependencies)](./kibana-plugin-core-server.savedobjectsclient.createpointintimefinder.md) |  | Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any find queries that might return more than 1000 saved objects.<!-- -->The generator wraps calls to [SavedObjectsClient.find()](./kibana-plugin-core-server.savedobjectsclient.find.md) and iterates over multiple pages of results using <code>_pit</code> and <code>search_after</code>. This will open a new Point-In-Time (PIT), and continue paging until a set of results is received that's smaller than the designated <code>perPage</code>.<!-- -->Once you have retrieved all of the results you need, it is recommended to call <code>close()</code> to clean up the PIT and prevent Elasticsearch from consuming resources unnecessarily. This is only required if you are done iterating and have not yet paged through all of the results: the PIT will automatically be closed for you once you reach the last page of results, or if the underlying call to <code>find</code> fails for any reason. |
+|  [createPointInTimeFinder(findOptions, dependencies)](./kibana-plugin-core-server.savedobjectsclient.createpointintimefinder.md) |  | Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any <code>find</code> queries that might return more than 1000 saved objects, however this API is only intended for use in server-side "batch" processing of objects where you are collecting all objects in memory or streaming them back to the client.<!-- -->Do NOT use this API in a route handler to facilitate paging through saved objects on the client-side unless you are streaming all of the results back to the client at once. Because the returned generator is stateful, you cannot rely on subsequent http requests retrieving new pages from the same Kibana server in multi-instance deployments.<!-- -->The generator wraps calls to [SavedObjectsClient.find()](./kibana-plugin-core-server.savedobjectsclient.find.md) and iterates over multiple pages of results using <code>_pit</code> and <code>search_after</code>. This will open a new Point-In-Time (PIT), and continue paging until a set of results is received that's smaller than the designated <code>perPage</code>.<!-- -->Once you have retrieved all of the results you need, it is recommended to call <code>close()</code> to clean up the PIT and prevent Elasticsearch from consuming resources unnecessarily. This is only required if you are done iterating and have not yet paged through all of the results: the PIT will automatically be closed for you once you reach the last page of results, or if the underlying call to <code>find</code> fails for any reason. |
 |  [delete(type, id, options)](./kibana-plugin-core-server.savedobjectsclient.delete.md) |  | Deletes a SavedObject |
 |  [deleteFromNamespaces(type, id, namespaces, options)](./kibana-plugin-core-server.savedobjectsclient.deletefromnamespaces.md) |  | Removes namespaces from a SavedObject |
 |  [find(options)](./kibana-plugin-core-server.savedobjectsclient.find.md) |  | Find all SavedObjects matching the search query |

docs/development/core/server/kibana-plugin-core-server.savedobjectsrepository.createpointintimefinder.md

@@ -4,7 +4,9 @@
 
 ## SavedObjectsRepository.createPointInTimeFinder() method
 
-Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any find queries that might return more than 1000 saved objects.
+Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any `find` queries that might return more than 1000 saved objects, however this API is only intended for use in server-side "batch" processing of objects where you are collecting all objects in memory or streaming them back to the client.
+
+Do NOT use this API in a route handler to facilitate paging through saved objects on the client-side unless you are streaming all of the results back to the client at once. Because the returned generator is stateful, you cannot rely on subsequent http requests retrieving new pages from the same Kibana server in multi-instance deployments.
 
 This generator wraps calls to [SavedObjectsRepository.find()](./kibana-plugin-core-server.savedobjectsrepository.find.md) and iterates over multiple pages of results using `_pit` and `search_after`<!-- -->. This will open a new Point-In-Time (PIT), and continue paging until a set of results is received that's smaller than the designated `perPage`<!-- -->.
 

docs/development/core/server/kibana-plugin-core-server.savedobjectsrepository.md

@@ -22,7 +22,7 @@ export declare class SavedObjectsRepository
 |  [checkConflicts(objects, options)](./kibana-plugin-core-server.savedobjectsrepository.checkconflicts.md) |  | Check what conflicts will result when creating a given array of saved objects. This includes "unresolvable conflicts", which are multi-namespace objects that exist in a different namespace; such conflicts cannot be resolved/overwritten. |
 |  [closePointInTime(id, options)](./kibana-plugin-core-server.savedobjectsrepository.closepointintime.md) |  | Closes a Point In Time (PIT) by ID. This simply proxies the request to ES via the Elasticsearch client, and is included in the Saved Objects Client as a convenience for consumers who are using <code>openPointInTimeForType</code>.<!-- -->Only use this API if you have an advanced use case that's not solved by the [SavedObjectsRepository.createPointInTimeFinder()](./kibana-plugin-core-server.savedobjectsrepository.createpointintimefinder.md) method. |
 |  [create(type, attributes, options)](./kibana-plugin-core-server.savedobjectsrepository.create.md) |  | Persists an object |
-|  [createPointInTimeFinder(findOptions, dependencies)](./kibana-plugin-core-server.savedobjectsrepository.createpointintimefinder.md) |  | Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any find queries that might return more than 1000 saved objects.<!-- -->This generator wraps calls to [SavedObjectsRepository.find()](./kibana-plugin-core-server.savedobjectsrepository.find.md) and iterates over multiple pages of results using <code>_pit</code> and <code>search_after</code>. This will open a new Point-In-Time (PIT), and continue paging until a set of results is received that's smaller than the designated <code>perPage</code>.<!-- -->Once you have retrieved all of the results you need, it is recommended to call <code>close()</code> to clean up the PIT and prevent Elasticsearch from consuming resources unnecessarily. This is only required if you are done iterating and have not yet paged through all of the results: the PIT will automatically be closed for you once you reach the last page of results, or if the underlying call to <code>find</code> fails for any reason. |
+|  [createPointInTimeFinder(findOptions, dependencies)](./kibana-plugin-core-server.savedobjectsrepository.createpointintimefinder.md) |  | Returns a [ISavedObjectsPointInTimeFinder](./kibana-plugin-core-server.isavedobjectspointintimefinder.md) to help page through large sets of saved objects. We strongly recommend using this API for any <code>find</code> queries that might return more than 1000 saved objects, however this API is only intended for use in server-side "batch" processing of objects where you are collecting all objects in memory or streaming them back to the client.<!-- -->Do NOT use this API in a route handler to facilitate paging through saved objects on the client-side unless you are streaming all of the results back to the client at once. Because the returned generator is stateful, you cannot rely on subsequent http requests retrieving new pages from the same Kibana server in multi-instance deployments.<!-- -->This generator wraps calls to [SavedObjectsRepository.find()](./kibana-plugin-core-server.savedobjectsrepository.find.md) and iterates over multiple pages of results using <code>_pit</code> and <code>search_after</code>. This will open a new Point-In-Time (PIT), and continue paging until a set of results is received that's smaller than the designated <code>perPage</code>.<!-- -->Once you have retrieved all of the results you need, it is recommended to call <code>close()</code> to clean up the PIT and prevent Elasticsearch from consuming resources unnecessarily. This is only required if you are done iterating and have not yet paged through all of the results: the PIT will automatically be closed for you once you reach the last page of results, or if the underlying call to <code>find</code> fails for any reason. |
 |  [delete(type, id, options)](./kibana-plugin-core-server.savedobjectsrepository.delete.md) |  | Deletes an object |
 |  [deleteByNamespace(namespace, options)](./kibana-plugin-core-server.savedobjectsrepository.deletebynamespace.md) |  | Deletes all objects from the provided namespace. |
 |  [deleteFromNamespaces(type, id, namespaces, options)](./kibana-plugin-core-server.savedobjectsrepository.deletefromnamespaces.md) |  | Removes one or more namespaces from a given multi-namespace saved object. If no namespaces remain, the saved object is deleted entirely. This method and \[<code>addToNamespaces</code>\][SavedObjectsRepository.addToNamespaces()](./kibana-plugin-core-server.savedobjectsrepository.addtonamespaces.md) are the only ways to change which Spaces a multi-namespace saved object is shared to. |

docs/development/plugins/data/public/kibana-plugin-plugins-data-public.querystringinputprops.submitonblur.md

@@ -0,0 +1,11 @@
+<!-- Do not edit this file. It is automatically generated by API Documenter. -->
+
+[Home](./index.md) &gt; [kibana-plugin-plugins-data-public](./kibana-plugin-plugins-data-public.md) &gt; [QueryStringInputProps](./kibana-plugin-plugins-data-public.querystringinputprops.md) &gt; [submitOnBlur](./kibana-plugin-plugins-data-public.querystringinputprops.submitonblur.md)
+
+## QueryStringInputProps.submitOnBlur property
+
+<b>Signature:</b>
+
+```typescript
+submitOnBlur?: boolean;
+```

src/core/server/saved_objects/service/lib/repository.ts

@@ -1918,7 +1918,16 @@ export class SavedObjectsRepository {
   /**
    * Returns a {@link ISavedObjectsPointInTimeFinder} to help page through
    * large sets of saved objects. We strongly recommend using this API for
-   * any find queries that might return more than 1000 saved objects.
+   * any `find` queries that might return more than 1000 saved objects,
+   * however this API is only intended for use in server-side "batch"
+   * processing of objects where you are collecting all objects in memory
+   * or streaming them back to the client.
+   *
+   * Do NOT use this API in a route handler to facilitate paging through
+   * saved objects on the client-side unless you are streaming all of the
+   * results back to the client at once. Because the returned generator is
+   * stateful, you cannot rely on subsequent http requests retrieving new
+   * pages from the same Kibana server in multi-instance deployments.
    *
    * This generator wraps calls to {@link SavedObjectsRepository.find} and
    * iterates over multiple pages of results using `_pit` and `search_after`.

src/core/server/saved_objects/service/saved_objects_client.ts

@@ -618,7 +618,16 @@ export class SavedObjectsClient {
   /**
    * Returns a {@link ISavedObjectsPointInTimeFinder} to help page through
    * large sets of saved objects. We strongly recommend using this API for
-   * any find queries that might return more than 1000 saved objects.
+   * any `find` queries that might return more than 1000 saved objects,
+   * however this API is only intended for use in server-side "batch"
+   * processing of objects where you are collecting all objects in memory
+   * or streaming them back to the client.
+   *
+   * Do NOT use this API in a route handler to facilitate paging through
+   * saved objects on the client-side unless you are streaming all of the
+   * results back to the client at once. Because the returned generator is
+   * stateful, you cannot rely on subsequent http requests retrieving new
+   * pages from the same Kibana server in multi-instance deployments.
    *
    * The generator wraps calls to {@link SavedObjectsClient.find} and iterates
    * over multiple pages of results using `_pit` and `search_after`. This will