Posts published under the Open source Team account are collective contributions by the dbi services open infrastructure consultants. The practice covers open-source databases and platforms, PostgreSQL first of all, along with MySQL, MariaDB, and Linux, from architecture, replication, and high availability to performance tuning and operations under service level agreements, and contributes actively to the PostgreSQL community.
We’ve finished the last post with a working Compute (Nova) service on the controller and compute node. While only the compute(s) actually run compute resources, Nova is also running on the controller for the management tasks, while libvirt is only running on the compute node. Once more, looking back at what we need at a minimum, we’ve done most of…
We’re now coming closer to the final setup of the OpenStack test environment. Again, looking at the minimum services we need, there are only three of them left: In this post we’ll continue with the Compute Service, which is called Nova. As a short reminder, if you followed the last posts, this is what we have currently: In addition to…
While doing some research on OpenStack (see here for the introduction post), I thought blogging about a new feature for the next version of PostgreSQL might free my head a bit from all those OpenStack services, so here we go 🙂 What you can do in PostgreSQL since a very long time, is to return the values of an insert/update…
By the end of the last post we finally got the first OpenStack service up and running: Keystone, the Identity Service. Going back to the list of services we need at a minimum, this still leaves us with some more to setup: In this post we’ll setup the Image Service (Glance) and the Placement service. The Glance service is responsible…
By the end of the last post, we finished with preparing the controller and compute node for being ready to deploy the first OpenStack service: Keystone, the Identity Service. Before we dive into the details lets quickly talk about why we need such a service. In an OpenStack setup there needs to be some form of authentication and service discovery.…
In the introduction post we’ve briefly looked at some options you have available when you want to build your own private cloud. In this post we’re going to start our journey into OpenStack by preparing the two nodes we’re going to use for this setup: The controller node, and one compute node. Maybe you remember from the last post, this…
While public clouds are a trend since several years now, some companies are also looking into self hosted solutions to build a private cloud. Some do this because of costs, others do this because they don’t want to be dependent on one ore multiple public cloud providers, others do it because they want to keep their data locally. There are…
For PostgreSQL it is critical, that autovacuum is able to keep up with the changes to the instance. One of the parameters you can adapt for this is autovacuum_max_workers. This parameter controls how many worker process can be started in parallel by the autovacuum launcher process. By default, this is set to 3, which means that a maximum of three…
When people talk about containers in FreeBSD, they usually talk about Jails. We’ve already seen how this works in general here and here. An additional option you have, is to install Podman on FreeBSD. Podman is well known in the Linux world and can be used to build and run containers. With the release of FreeBSD 14.2, the FreeBSD project…
Now that we know how to work with the ports collection let’s have a look another option you have in FreeBSD: Sometimes it might be required to run unmodified Linux binaries on FreeBSD and to achieve this, FreeBSD comes with a Linux compatibility layer. This is optional and not enabled by default, but pretty easy to setup and use. Before…
If you followed the previous introduction posts about FreeBSD (here, here, here, here and here) you should already know how to install binary packages. On FreeBSD this is done using pkg, and if you are familiar with apt, dnf, or zypper (or any other package manager on Linux) this should not be a big deal. There is, however, another official…
This post builds on top of the previous post in which we’ve created a simple thin Jail. The goal of this post is, to put a bit of automation around this and to provide a simple PostgreSQL as a service platform. We want to do this by creating one PostgreSQL instance per Jail (without any restrictions on resource usage for…
