Fix URL paths (#6189)

wiki/content/deploy/download.md

@@ -60,7 +60,7 @@ curl https://get.dgraph.io -sSf | VERSION=v2.0.0-beta1 bash
 ```
 
 {{% notice "note" %}}
-Be aware that using this script will overwrite the installed version and can lead to compatibility problems. For example, if you were using version v1.0.5 and forced the installation of v2.0.0-Beta, the existing data won't be compatible with the new version. The data must be [exported]({{< relref "deploy/index.md#exporting-database" >}}) before running this script and reimported to the new cluster running the updated version.
+Be aware that using this script will overwrite the installed version and can lead to compatibility problems. For example, if you were using version v1.0.5 and forced the installation of v2.0.0-Beta, the existing data won't be compatible with the new version. The data must be [exported]({{< relref "deploy/dgraph-administration.md#exporting-database" >}}) before running this script and reimported to the new cluster running the updated version.
 {{% /notice %}}
 
 ## Manual download [optional]

wiki/content/enterprise-features/encryption-at-rest.md

@@ -9,7 +9,7 @@ title = "Encryption at Rest"
 {{% notice "note" %}}
 This feature was introduced in [v1.1.1](https://github.com/dgraph-io/dgraph/releases/tag/v1.1.1).
 For migrating unencrypted data to a new Dgraph cluster with encryption enabled, you need to
-[export the database](https://dgraph.io/docs/deploy/#exporting-database) and [fast data load](https://dgraph.io/docs/deploy/#fast-data-loading),
+[export the database](https://dgraph.io/docs/deploy/dgraph-administration/#exporting-database) and [fast data load](https://dgraph.io/docs/deploy/#fast-data-loading),
 preferably using the [bulk loader](https://dgraph.io/docs/deploy/#bulk-loader).
 {{% /notice %}}
 
@@ -101,4 +101,3 @@ badger rotate --dir w --old-key-path enc_key_file --new-key-path new_enc_key_fil
 ```
 
 Then, you can start Alpha with the `new_enc_key_file` key file to use the new key.
-

wiki/content/faq/index.md

@@ -24,7 +24,7 @@ If you're running more than five tables in a traditional relational database man
 If your data doesn't have graph structure, i.e., there's only one predicate, then any graph database might not be a good fit for you. A NoSQL datastore is best for key-value type storage.
 
 ### Is Dgraph production ready?
-We recommend Dgraph to be used in production at companies. Minor releases at this stage might not be backward compatible; so we highly recommend using [frequent exports](/deploy#exporting-database).
+We recommend Dgraph to be used in production at companies. Minor releases at this stage might not be backward compatible; so we highly recommend using [frequent exports](/deploy/dgraph-administration/#exporting-database).
 
 ### Is Dgraph fast?
 Every other graph system that we've run it against, Dgraph has been at least a 10x factor faster. It only goes up from there. But, that's anecdotal observations.

wiki/content/howto/loading-csv-data.md

@@ -73,7 +73,7 @@ _:d,_:a
 ```
 
 {{% notice "note" %}}
-To reuse existing integer IDs from a CSV file as UIDs in Dgraph, use Dgraph Zero's [assign endpoint]({{< relref "deploy/index.md#more-about-dgraph-zero" >}}) before data loading to allocate a range of UIDs that can be safely assigned.
+To reuse existing integer IDs from a CSV file as UIDs in Dgraph, use Dgraph Zero's [assign endpoint]({{< relref "deploy/dgraph-zero" >}}) before data loading to allocate a range of UIDs that can be safely assigned.
 {{% /notice %}}
 
 To get the correct JSON format, you can convert the CSV into JSON and use `jq`