Update dns related processes

content/departments/product-engineering/engineering/cloud/devops/dns.md

@@ -1,17 +1,42 @@
 # Add or modify DNS records to Sourcegraph owned domains
 
 Sourcegraph follows an Infrastructure as Code (IaC) model. To make changes to Sourcegraph
-owned DNS records you should make the change to the code in Terraform in our infrastructure repo.
+owned DNS records you should make the change to the code in Terraform in our [sourcegraph/infrastructure] repo.
 
+1. Clone [sourcegraph/infrastructure]
 1. Navigate to the `dns` folder
 1. Create a file, add the record near the most relevant DNS entries, or modify the existing record
 
    1. For domains with a large number of TXT records like
       Sourcegraph.com, we typically create a file `sourcegraph.$TXT_PROVIDER.tf`
 
-1. Create a pull request in the infra repo with your changes
+1. Create a pull request in the repo with your changes
+1. Open [Buildkite](https://buildkite.com/sourcegraph/infrastructure)
+
+   1. Check status of `ci/terraform.sh dns` job and it should pass without error.
+   1. If there is any error, you should consult the logs or reach out to `#cloud-devops`
+
+1. Wait for Buildkite to finish your build
 1. Get approve from the cloud-devops team
 1. `terraform apply` the PR or ask a member of the devops team to do this
 1. Merge the PR **after** the apply succeeds
 
-For other
+## FAQ
+
+### Are you adding a DNS record for a Netlify site?
+
+You should create a CNAME record pointing to `sourcegraph.netlifyglobalcdn.com`. In terraform, it looks something like
+
+[example](https://github.com/sourcegraph/infrastructure/blob/621b30a24f448f78509b75600d4338491a7ec127/dns/sourcegraph.tf#L48-L54)
+
+```tf
+resource "cloudflare_record" "about-sourcegraph-com" {
+  zone_id = data.cloudflare_zones.sourcegraph_com.zones[0].id
+  name    = "about"
+  type    = "CNAME"
+  value   = var.netlify-enterprise-cdn-cname
+  proxied = false
+}
+```
+
+[sourcegraph/infrastructure]: https://github.com/sourcegraph/infrastructure

content/departments/product-engineering/engineering/process/deployments/playbooks.md

@@ -167,27 +167,23 @@ See an [example query](https://console.cloud.google.com/bigquery?sq=527047051561
 
 **Note**: This method only permits read-only access
 
-### Restarting about.sourcegraph.com and docs.sourcegraph.com
+### Restarting docs.sourcegraph.com
 
-To restart the services powering about.sourcegraph.com and docs.sourcegraph.com:
+To restart the services powering docs.sourcegraph.com:
 
-1. List the active pods with `kubectl get pods`. This should produce a list that includes items like the following:
+Configure kubectl context to the dotcom cluster
 
-   ```
-   NAME                                     READY   STATUS      RESTARTS   AGE
-   about-sourcegraph-com-74f96c659b-t6lqp   1/1     Running     0          7m49s
-   docs-sourcegraph-com-5f97dd5db-7wrxm     1/1     Running     0          7m49s
-   ```
-
-   The exact names will differ, but the general format will be the same.
+```sh
+gcloud container clusters get-credentials cloud --zone us-central1-f --project sourcegraph-dev
+```
 
-1. Delete the pods, being careful to copy the exact names from the output in the previous step. For example, with that output, you would run:
+Rollout a restart
 
-   ```
-   kubectl delete pod about-sourcegraph-com-74f96c659b-t6lqp docs-sourcegraph-com-5f97dd5db-7wrxm
-   ```
+```sh
+kubectl -n default rollout restart deploy docs-sourcegraph-com
+```
 
-1. Wait a moment, and check https://about.sourcegraph.com/ and https://docs.sourcegraph.com/.
+Wait a moment, and check https://docs.sourcegraph.com/.
 
 ### Creating banners for maintenance tasks