Hacker News Reader

Comment by apavlo | Hacker News Reader

We have deployed OtterTune for on-prem databases (baremetal, containers). But we are not offering that to everyone right now because organizations manage their configurations for these deployments in a bunch of different ways (Github actions, Chef/Puppet, Terraform). The nice thing about RDS is that they have a single API for updating configurations (https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/USER_...).

As for the Postgres-derivates that you mentioned (Timescale, Citus), we have not tested OtterTune for them yet. But there is no reason they shouldn't also work because they expose the same metrics API (pg_stat_database) and config knobs. There are just way more people using Postgres (RDS, Aurora), so we are focusing on them right now.

View on HN

Hi Andy.

First of all, congrats for the project.

As you probably know, you and any other user can indeed manage Postgres configurations, in any environment, via an API. See the "MANAGE" section in https://postgresqlco.nf/ Essentially, a hosted "Git repository" for Postgres configurations.

Alvaro! Good to hear from you. We really like the Ongres conf management tool. We just haven't come across anybody using it for on-prem DBs in the wild.

As I said above, everyone that we've talked to rolls their conf management tools because they are doing more than just DBMS confs (proxies, networking, kernel params, middleware, etc). Lots of Terraform.

Glad to hear that.

Yes, indeed we have considered doing a TF module for the configuration management API :)

What about OpenStack Trove DBaaS? OpenStack Trove is like an open source self-hosted Amazon RDS or Google CloudSQL. https://docs.openstack.org/trove/latest/

FWIU, Trove supports 10+ databases including MySQL and PostgreSQL.

AFAIU, there are sound reasons to host containers with e.g. OpenStack VMs instead of a k8s scheduler with a proper SAN and just figure out how to redundantly and resiliently sync - replicate, synchronize, primary/secondary, nodeprocd - and tune given the CAP theorem and the given DB implementation(s)?

Here's the official Ansible role for Trove, which provisions various e.g. SQL databases on an OpenStack cloud: https://github.com/openstack/openstack-ansible-os_trove

Makes sense! thanks for the reply Andy.

Reply on news.ycombinator.com