Merge branch 'jl-puma-add-healthcheck' into 'master'

Add recommendation to use the Puma health check

Closes #212850

See merge request gitlab-org/gitlab!32731

doc/administration/high_availability/load_balancer.md

@@ -109,6 +109,10 @@ Configure DNS for an alternate SSH hostname such as `altssh.gitlab.example.com`.
 | ------- | ------------ | -------- |
 | 443     | 22           | TCP      |
 
+## Readiness check
+
+It is strongly recommend that multi-node deployments configure load balancers to utilize the [readiness check](../../user/admin_area/monitoring/health_check.md#readiness) to ensure a node is ready to accept traffic, before routing traffic to it. This is especially important when utilizing Puma, as there is a brief period during a restart where Puma will not accept requests.
+
 ---
 
 Read more on high-availability configuration:

doc/administration/operations/puma.md

@@ -23,10 +23,11 @@ Beginning with GitLab 13.0, Puma is the default application server. We plan to r
 Unicorn in GitLab 14.0.
 
 When switching to Puma, Unicorn server configuration
-will _not_ carry over automatically, due to differences between the two application servers.
+will _not_ carry over automatically, due to differences between the two application servers. For Omnibus-based
+deployments, see [Configuring Puma Settings](https://docs.gitlab.com/omnibus/settings/puma.html#configuring-puma-settings).
+For Helm based deployments, see the [Webservice Chart documentation](https://docs.gitlab.com/charts/charts/gitlab/webservice/index.html).
 
-For Omnibus-based deployments, see [Configuring Puma Settings](https://docs.gitlab.com/omnibus/settings/puma.html#configuring-puma-settings). For Helm
-based deployments, see the [Webservice Chart documentation](https://docs.gitlab.com/charts/charts/gitlab/webservice/index.html).
+Additionally we strongly recommend that multi-node deployments [configure their load balancers to utilize the readiness check](../high_availability/load_balancer.md#readiness-check) due to a difference between Unicorn and Puma in how they handle connections during a restart of the service.
 
 ## Performance caveat when using Puma with Rugged