June 2025 – Made by Mikal

Getting Digitech / Bresser 6in1 weather stations working with rtl_433, MQTT, and Home Assistant

Post author:mikal
Post published:June 24, 2025
Post category:HomeAssistant

Please note: this post is a re-write of a previous post about my Vevor weather station that failed after a few months because of ineffective weather seals on the outdoor unit. Given the poor warranty service from Vevor, I instead went and bought a weather station from a local retailer. In my case a Jaycar Digitech XC-0434, which also appears to be sold online as a Bressser 6in1 weather station.

Now that we’ve defeated QNAP’s slightly broken udev, we can run a Docker container with rtl_433 in it to wire up our Digitech / Bresser 6in1 weather station to Home Assistant via MQTT. First off, we need a Docker container running rtl_433, which assumes you’ve already setup the udev rule mentioned in the previous post, even if you’re not using a QNAP!

(more…)

The Complete Robot (again)

Post author:mikal
Post published:June 23, 2025
Post category:Book

I’ve read this book a few times, but honestly the Foundation TV series has left me with a new enthusiasm to re-read some Asimov stuff. I have previously read the entire extended multi-author series, and honestly a fair few of them sucked — especially the ones by other authors — so this time I have the luxury of being a bit more picky. Worse, Asimov remixed the robot stories several times into various volumes, and it can be quite confusing. The Complete Robot contains all the robot stories, and replaces I, Robot (reading one, reading two), The Rest of the Robots; Robot Dreams; and Robot Visions. It also contains a couple of previously unpublished stories.

(more…)

Title: The Complete Robot
Author: Isaac Asimov
Genre: Robots
Release Date: 1982
Pages: 688

Training isn’t a work perk, its essential operational risk mitigation

Post author:mikal
Post published:June 13, 2025
Post category:Work

I’ve been thinking a bit about training at work recently, largely in the context of having spent the last twenty years working for US technology companies. I think effectively all of these companies made a pretty big mistake — they viewed training of employees as a perk much like vacation, book budgets, or t-shirts. They advertise their training programs as part of their recruitment process, and just like other perks they’re cut when times get a bit grim. However, that’s not actually why employers should train their people. We train people so that they have the skills they need to do their jobs — especially when things get real and aren’t working out to plan.

There are definitely industries who have good examples of this sort of risk reduction training done well — airlines and the military both engage in regular training activities that ensure that when things get exciting the people know what to do. This includes leaders being trained on how to make decisions that are likely to result in the desired outcomes.

(more…)

I really wanted to like etcd, but Andy Pavlo was right

Post author:mikal
Post published:June 9, 2025
Post category:Databases etcd Shaken Fist

Andy Pavlo of the CMU Database Group is well known for saying that while NoSQL databases acquire cyclical popularity, all databases eventually iterate back to a SQL interface — it happened with MongoDB and Google’s BigTable for example.

I think I have hit that point with etcd. Initially I ported from MySQL to etcd because I really wanted the inexpensive distributed locking and being able to watch values. However, I never actually watch values in my code any more, and I now spend a huge amount of my time maintaining what my code calls “caches”, but which I can now see are just poorly implemented secondary indexes. The straw that broke the camel’s back was https://github.com/etcd-io/etcd/issues/9043, which changed etcd’s defaults to only being able to return 1.5mb in a RPC request.

I therefore think it might be time for me to port back to a real SQL database, perhaps keeping etcd to manage distributed locks. Perhaps.

(more…)

Databases built on object stores are officially interesting

Post author:mikal
Post published:June 4, 2025
Post category:Databases

…even if none of my friends seem to think so.

I’ve been off on a bit of a tangent recently. Its a slow burn tangent, that I am pretty sure was kicked off by this Geek Narrator podcast episode about the design of Turbo Puffer with Simon Eskildsen:

The basic idea is that you can build very large scale database systems using only the primitives provided by an object store such as Amazon S3. Now, the performance might also suck, but you can alleviate some of that with a good caching layer and in return you get massive scale. This first video caused me to discover the work of Andy Pavlo, who was interviewed by the same podcast:

(more…)

The simplest boot target for the Kerbside SPICE VDI proxy CI

Post author:mikal
Post published:June 1, 2025
Post category:Kerbside OpenStack Shaken Fist

For the last couple of years I have been working on a SPICE protocol native proxy called Kerbside. The basic idea is to be able to provide SPICE Virtual Desktop Interface (VDI) consoles to users from cloud platforms such as Shaken Fist, OpenStack, or oVirt. Think Citrix, but for Open Source cloud platforms. SPICE is attractive here because it has some features that other more common VDI protocols like VNC don’t have — good cut and paste support, USB device pass-through, multiple monitor support, and so on. RDP has these, but RDP was not a supported VDI protocol when using qemu on Linux with KVM until incredibly recently — literally the last couple of months.

(In terms of clouds that Kerbside supports, I think it would be relatively trivial to also support Proxmox, KubeVirt, or a list of static manually created virtual machines, but there’s only so many things one Mikal can do at once…)

Some of these cloud platforms have supported SPICE consoles for a while, but generally with warts. OpenStack for example only exposes them as HTML5 transcoded sessions with reduced functionality. oVirt exposes them via a “proxy” which is just squid (or equivalent), but its fairly dumb — it exposes the underlying hypervisor details to the client for example. I thought I could do better than that.

(more…)