Documentation review questions

The following are my questions (some technical, some language/context-based) after going through the docs:

(The numbers are the chapters/pages in the docs)

• 1.3: TBH, these arch pictures doen't really make it much clearer and IMO should be re-designed. I'd prefer to see the higher-level architecture first, i.e. which components are there, without talking about the Kubernetes resources (pods, etc.), and then how it is translated to Kubernetes resources, with the stateful sets, pods, etc. Right now, this is a bit of a mixture and it took me a longer time to figure out with the diagram, what is contained and what the responsibilities are.
• 1.3 STACKGRES POD ARCHITECTURE DIAGRAM: it's not quite clear to me, does this show the containers in a pod? But then the PV should be outside, right?
• 1.4 "Connecting directly to PostgreSQL does not scale very well." -> I'd be much more specific, and point out that we're talking about individual concurrent connections. People who aren't that deep into the Postgres world might not get as to why or might misunderstand this.
• 1.4 Connection pooling is required in order to not saturate PostgreSQL processes by creating queue of sessions, transactions -> rephrase and maybe make it clearer or formulate it out a bit more in multiple sentences that are clearer.
• 1.4 PROXY -> I would mention that sidecars are used, and the whole setup more. Otherwise that might not be clear, even if people looked at page 1.3
• 1.4 "Envoy is an open source edge and service proxy,..." -> this whole paragraph is a bit weird to read with the examples, without more context of the setup
• 1.5 "We respect others who switch to or are directly built as source-available software" -> this is a bit hard to understand; I'd leave out the part with "respect others"
• 1.8 Make the first table col wider for readability
• 3.1 Kubernetes-compliant cluster from version 1.18 to 1.25 -> the website mentions 1.16+
• 3.1 Does everybody use eksctl or should aws be shown as alternative?
• 3.2 Is there a full list of additional parameters for the Helm chart?
• 4.1.1.2 Google setup: is --no-enable-ip-alias --release-channel "None" --metadata disable-legacy-endpoints=true still required? maybe remove this
• 8.1 "An attempt to update or delete a default configuration has been detected." -> is there a difference to blacklisted?
• 9.3 Maybe include how to create the backup in the first place?
• 9.4 Is the first Prom rule correct? the indentation looks weird

And the following questions appeared in the Slack channel, which might (or might not) be room for improvement, since apparently it wasn't clear:

• PosgreSQL cluster upgrade. I saw that a minor version upgrade is supported. What about major version upgrades?
• Is it possible to use my own distributed login solution? e.g. I need to send logs to Kafka.
• Is it possible to add my own sidecars?
• Is it possible to define users and DBs that must be created in some PostgreSQL cluster? Monitoring. It is good that we can see it in the SG UI, but do you have similar Grafana dashboards?
• Do you monitor Patroni somehow? e.g. data lag, PostgreSQL Instance status (Leader or Replica etc).