Chameth.com - posts like apple-google-aligned-incentives, docker-automatic-nginx-proxy, g15-ram-upgrade, ipv6-docker-routing, why-you-should-be-using-https

An interesting Tailscale + Docker gotcha

2026-01-29T00:00:00Z

As I’ve written about before, I use Tailscale for a lot of things. I thought I had it set up in a reasonably secure manner, but I recently noticed a problem.

I use Tailscale’s ACLs to limit what each node can access, based on the tags I apply to it. So an app node can’t access anything via Tailscale, while an integration or server node can access things tagged with either app or integration. This is expressed pretty simply in the Tailscale ACL JSON:

{
	"tagOwners": {
		// Servers that can be SSH'd into
		"tag:server": [],
		// Applications that are exposed on tailscale but never connect out
		"tag:app": [],
		// Things which talk to other services over tailscale (connecting to apps etc)
		"tag:integration": [],
	},
	"grants": [
		{
			// Users can access everything
			"src": ["autogroup:member"],
			"dst": ["*"],
			"ip":  ["*"],
		},
		{
			// Servers and integrations can access integrations and apps
			"src": ["tag:server", "tag:integration"],
			"dst": ["tag:integration", "tag:app"],
			"ip":  ["*"],
		},
	],
}

Over the past week I’ve been setting up a private Forgejo instance behind Tailscale, complete with an actions runner that runs things using a docker-in-docker container. I didn’t want the runner knowing anything about Tailscale, so I had it configured to speak to Forgejo direct over HTTP (forgejo:3000) instead of using full Tailscale HTTPS URL that I use when accessing it (http://git.example-net.ts.net/)¹.

Everything was going fine, until I forgot to do that translation… and it worked. My Forgejo action runners could access anything on my tailnet. I run several things on Tailscale that just have authentication turned off, on the basis that only authorised devices can access them. Things like the admin interface for this website. I definitely didn’t intend for any workflow I run on my git server to have access to edit my website!

This wasn’t just limited to Forgejo, either. Any docker container I was running could access the tailnet. In hindsight it’s fairly obvious why: the host is running Tailscale, connected as a node tagged with my server tag. That creates a tailscale0 interface, and automagically sets up iptables rules to route Tailscale traffic over the interface. Docker also automagically sets up iptables rules to bridge traffic, and apparently these two sets of rules interact in such a way that traffic from Docker containers is allowed to route via the tailscale0 interface.

I say it’s fairly obvious in hindsight — there’s no reason why Docker would special case any particular host interface after all — but it still feels pretty surprising. Because both bits of software inject their own iptables rules, I never really had a good mental model for how they interact. The host Tailscale node was a completely separate building block to Docker. It would be a pain to use either of them if they didn’t do these rules, but it’s also one of the reasons I don’t really like “magical” things².

There are lots of ways to fix this, but none of them feel particularly great. You can configure both Tailscale and Docker to not automatically fiddle with iptables and handle the rules yourself, but I really hate dealing with iptables³. As a stopgap I did hold my nose and add some iptables rules to drop traffic to the tailscale0 interface if it originated from the IP ranges that Docker was configured to use:

iptables -I DOCKER-USER -s 192.168.0.0/16 -o tailscale0 -j DROP
iptables -I DOCKER-USER -s 172.17.0.0/12 -o tailscale0 -j DROP

The DOCKER-USER chain is a nice little escape hatch; it comes before the main auto-generated DOCKER chain, and Docker leaves the rules in it alone.

So I added these rules, and felt pretty good about myself, and then… everything started breaking in weird ways. After some debugging I realised the problem was DNS⁴. The server’s DNS resolver is 100.100.100.100, a special Tailscale address. This is so that it can resolve tailnet hostnames via MagicDNS (and so I don’t have to configure my custom DNS servers manually on each device; Tailscale does it for me). My new iptables rules inadvertently dropped all the DNS packets coming from docker containers. D’oh.

Obviously the solution here is to double down and add MORE iptables rules:

iptables -I DOCKER-USER -s 192.168.0.0/16 -d 100.100.100.100 -o tailscale0 -j ACCEPT
iptables -I DOCKER-USER -s 172.17.0.0/12 -d 100.100.100.100 -o tailscale0 -j ACCEPT

This got DNS working again, but it all feels a bit gross. Not least because to actually make these persist you need to use iptables-save, and then you also get all the rules that both Docker and Tailscale have inserted. I came across a script to filter out the Docker ones, but… yuck⁵.

I realised a cleaner idea might just be to lock down what server tagged nodes could access within Tailscale itself. That way I can avoid touching iptables at all. It’s handy to be able to curl services from the host when debugging, but it’s not really necessary. So I removed that access, and… stuff broke again. I’m using my Forgejo instance as a registry for some of the docker images I run, so the docker daemon needs to be able to reach it. I ended up making a new tag for infrastructure, which can be accessed from server devices. This does still allow all the docker containers to reach Forgejo, but I already have it set up with appropriate access controls and public/private repository splits. Forgejo is a service designed to run publicly, so this seems a reasonable trade-off for convenience. I used the tests feature of Tailscale’s ACL config to make sure I’d got the rules right:

{
	"tests": [
		{
			// Servers can only access infrastructure
			"src":   "tag:server",
			"proto": "tcp",
			"allow": [
				"tag:infrastructure:8080",
			],
			"deny": [
				"100.84.16.43:8080",
				"me@example.com:8080",
				"tag:server:8080",
				"tag:app:8080",
				"tag:integration:8080",
			],
		},
	],
}

The whole situation still feels a bit messy. If I ever get around to switching to nftables I might loop back and manually craft some rules for routing traffic, instead of leaving Tailscale and Docker to do their own thing.

I discovered later on that logging into the container registry at forgejo:3000 actually issued a redirect to https://git.example-net.ts.net/ so this was all basically for naught… ↩︎
I didn’t use Tailscale’s MagicDNS for a long time just because the word “magic” put me off. Only when I eventually got around to learning how it worked, and seeing that it wasn’t really that magical under the hood, did I change my mind. ↩︎
nftables seems far better in lots of ways, but I can’t really be bothered migrating. Maybe next time I reimage the server for whatever reason… ↩︎
It’s always DNS… ↩︎
Again, nftables would almost certainly help here. It actually has (gasp) configuration files. But again, I really didn’t want to spend the time migrating. ↩︎

How tech companies failed to build the Star Trek computer

2025-07-16T00:00:00Z

A computer core room on the Enterprise-D

In most Star Trek series, the ship or station computer is ever-present in the background, waiting to be called on by the main characters¹. It nearly always does exactly the right thing, and there’s little limit to the functions it can perform. Take this mundane example from DS9:

KIRA: Computer, establish link with the Bajoran Medical Index for the Northwestern District.
COMPUTER: Link established.
KIRA: Access all information on Doctor Surmak Ren.
COMPUTER: There are no records matching that name.
KIRA: Try the Northeastern District, same search.
COMPUTER: Doctor Surmak Ren, currently serving as Chief Administrator of the Ilvian Medical Complex.
KIRA: Computer, open a channel to the Ilvian Medical Complex. Administrator’s office.

The computer is doing some kind of networking to a database only identified by name. It does a search and summarises the lack of results. It then repeats the process with another database, and succinctly announces the results. Finally, it opens a communication channel to a specific room in a facility, based only on its name.

This whole interaction is remarkably boring². Kira doesn’t have to know any URLs or API endpoints, or what protocol she wants to use. She doesn’t have to open a specific app and then login and then try the query again. She just says what she wants and the computer does it.

It seems like this should be one of the most easily obtainable bits of sci-fi wizardry with our current technology. We have multiple massive companies throwing lots of money at digital assistants, LLMs that are improving at an insane rate, but we’re somehow not even close to the usability or usefulness of the Trek computers. What gives?

Boring is, well, boring.

Larry Page once said something that might help explain it:

The Star Trek computer doesn’t seem that interesting. They ask it random questions, it thinks for a while. I think we can do better than that.

This is the same Larry Page that founded Google, whose mission statement is “to organize the world’s information and make it universally accessible and useful”. Of all people, surely he should find an omnipresent computer that can answer ‘random questions’ interesting?! It seems like it should be the epitome of Google’s mission!

Google’s “better than that” seems to have been to stuff LLMs into every product they can, even when you don’t want them there. Even when they’re worse than the normal content they displace. These things look exciting when they’re part of a scripted demo at Google I/O, but they fall flat and just get in the way when they’re exposed to the reality of day-to-day use.

The Star Trek computer is the opposite: it isn’t snazzy, but it is genuinely useful. That means it’s not an attractive target for the company execs who want marketing opportunities, and it’s not appealing for engineers who need to demonstrate “impact”. But even if Google did try to make the Trek computer, there are other problems…

Assistants need to be free

A significant amount of tech companies’ business models currently revolves around trapping users in walled gardens. They want you using their ecosystem; that way they get more data from you, and you’re more likely to spend more money on their other offerings that work together. There’s barely any incentive to allow any kind of interoperability with other platforms outside carefully contracted integrations.

I remember trying to help a family member move their photos from iCloud to Google Photos. At one point they turned around and said, exasperated, “why is this so hard? Aren’t they both in the cloud?!”. It’s easy to dismiss that as someone who hasn’t quite grasped the fundamental idea that “the cloud” is just someone else’s computers, but that’s not the whole story. There’s no reason why there shouldn’t be a quick and easy transfer: both services already allow uploading and downloading, there’s just no incentive for the companies involved to make it so³.

These kinds of misaligned incentives and walled garden business models cause even more problems when it comes to digital assistants. Siri is basically never going to be able to interact with, say, your Google Drive; ~~Bard~~ Gemini is never going to be able to send a message via iMessage. Even when there are appropriately blessed interactions, they’re so clunky. Can you imagine Captain Picard saying “Computer, ask the turbolift skill to take me to deck 5”?

Someone else’s computer

Software issues aside, there’s still a key difference between the Star Trek computers and our current batch of digital assistants: where they run. The Trek computers are all housed within the ship or station they serve; they can connect elsewhere to gather information, but they run entirely independently. If they go wrong, a local engineer can go in and fix things. While some of our assistants may have physical hardware in your home, they don’t work without a vast cloud apparatus behind them. If your Internet connection fails, they become paperweights. If the company running them decide to remove some functionality you depend on, you have no recourse.

That kind of helplessness isn’t limited to assistants, either. There’s a rapidly growing trend of being unable to modify or repair hardware you fully own and control. Part of this is just that they’re becoming more complex: it’s a lot harder to replace a microchip than a gear, but companies are also going out of their way to make it more difficult for users through draconian DRM regimes⁴ and aggressive intellectual property enforcement. If the US Navy can’t repair their own equipment because a corporation says so, what hope do consumers have?

We’re approaching a point where you don’t actually own anything. Software is cloud and subscription based, hardware is unrepairable. Even cars can be remotely updated and have features added or removed. The Federation wouldn’t allow a third party control over their ships⁵, so why are we so happy to put up with it in everything we consume?

A small ray of hope?

The most promising way of tackling all of these problems is through legislation. The EU’s Digital Market Act is an attempt to force ‘gatekeepers’ like Google, Apple and Meta, to allow third-party access to their services. It seems like a pretty reasonable approach, but the tech companies are unsurprisingly resisting it. Apple in particular have gone out of their way to refuse to comply, and when forced to do so have limited the functionality to people in Europe. Still, the DMA is a promising start, and if similar legislation is introduced (and robustly enforced) elsewhere it might start forcing companies to behave a bit better.

There are also smaller companies that actually do the right thing. Framework make laptops that are user-serviceable; Fairphone do the same for mobile phones. Smaller software companies provide useful, open APIs. The average person on the street will probably have never heard of these, unfortunately, but they do still exist. Maybe as the bigger tech companies tighten the screws more, people will turn to alternatives like this? Or maybe we’ll just keep accepting that our computers work for everyone but us?

Unless, of course, the computer is playing the role of the episode’s MacGuffin and has contracted space-computer-COVID or something, then it’s a lot less in-the-background. ↩︎
It’s almost like it only exists to move the plot along. ↩︎
You can generally export your data, thanks to a combination of legislation and efforts like Google’s “Data Liberation Front”, but I’ve never seen an export format that could then just be imported into an equivalent commercial product. ↩︎
Oh, you’ve changed the screen on your iPhone? Better hope it can do the secret handshake with the Apple hardware. ↩︎
I think there might actually have been an episode where that did in fact happen. We’ll just ignore that as a plot contrivance. ↩︎

Upgrading the RAM in a Dell G15 laptop

2023-07-29T00:00:00Z

The Dell G15

I currently use a Dell G15 laptop for work. It has served me well for a little over a year, but recently it has been struggling a little with my day-to-day workload. It came with 32GB of RAM — the highest possible specification at the time¹ — but that is apparently no longer enough for me.

For a recent project, I was working on a Rust library used in an Android app. That meant running the usual glut of Android tools (Android Studio, an emulator and at least one Gradle daemon) alongside a normal IDE (IntelliJ IDEA). Throw in a web browser and a couple of electron apps, and I often managed to use all 32GB.

When you start swapping memory out to an encrypted disk — even an SSD — it doesn’t make for great performance. At first, I tried to work around this by enabling the Linux out-of-memory (OOM) killer, but it turns out that it’s not too good with Electron apps: it will kill the large browser process, but then the small Electron wrapper will just respawn it.

Can it be upgraded or not?

The obvious solution to not having enough RAM is to add more RAM. A quick look in the manual showed this might not be possible, though. The manual includes the following “Memory specifications” table:

Description	Values
Memory slots	Two SODIMM slots
Memory type	DDR5
Memory speed	4800
Maximum memory configuration	32GB
Minimum memory configuration	8GB
Memory size per slot	8GB or 16GB
Memory configurations supported	8 GB, 1 x 8 GB, DDR5, 4800 MHz 16 GB, 1 x 16 GB, DDR5, 4800 MHz 16 GB, 2 x 8 GB, DDR5, 4800 MHz, dual-channel 32 GB, 2 x 16 GB, DDR5, 4800 MHz, dual-channel

That unambiguously says that an upgrade from 32GB is not possible. I gave up.

Later though, I was complaining about memory issues to a friend, and he pointed out a Dell forum thread where a couple of people claim to have successfully installed dual-channel 32GB modules. Since the alternative was getting an entire new PC after only a year, I decided to give it a go.

The upgrade attempt

I ordered a pair of Crucial 32GB DDR5-4800 SODIMMs, and after they turned up dismantled the laptop. The G15 comes apart pretty normally: there are uncovered screws on the bottom holding the lower part of the case on. With those removed and some gentle prying, it pops off, and you get access to the battery, GPU and motherboard.

The first thing I saw was this:

The G15 motherboard adamantly proclaiming that it doesn’t want 32GB SODIMMs

Not one but two labels that indicate it will only accept 8GB or 16GB modules. Oh well, what’s the worst that can happen?

Oops?

I dutifully installed the new modules, reconnected the battery and put the case back together. I pressed the power button, and… nothing. None of the usual garish lights that immediately come on, no screen output, just a dead laptop. After reading some more of the user manual, I found that there is a tiny status LED on the side next to the ethernet port. Forcing the laptop to power off and back on again, the status LED blinked a distress code at me: 2 amber blinks, 4 white blinks. The manual says that is — unsurprisingly — a memory fault.

I figured at this point that the manual and labels on the motherboard were probably right. I took the laptop apart again, reinstalled the original 2x16GB modules, reassembled it, and pressed the power button. It didn’t boot. I don’t spend a lot of time fiddling inside computers, but I’ve done it enough that I’m reasonably confident I can’t entirely break a computer while swapping some RAM modules. I took to Googling², and found an interesting article that said Dell laptops don’t like to boot after RAM changes unless you clear the CMOS by popping out the battery for 15 minutes.

I opened the laptop up, and looked around for the CMOS battery. There wasn’t one. Turns out they don’t exist any more. I left the main battery disconnected for a while to see if it would help, and it didn’t.

Unexpected success

I started to get worried: if I couldn’t fix this, I wouldn’t be able to work until I got a new PC, and that wasn’t really in my budget at the minute. I sat reading old forum threads and help guides³, none of which were actually useful. Out of nowhere, though, the laptop booted up.

Naturally, I immediately shut the laptop down again, opened it up, and switched back to the new RAM modules. Then I turned it on again and sat waiting. After about 15 minutes of it looking totally dead, it turned on and showed a BIOS warning about the hardware configuration being changed. It then booted perfectly normally, and all 64GB of RAM was visible and usable.

My theory is that the forum threads were right: Dell laptops are funny about RAM upgrades. But somehow in removing the physical CMOS battery, they’ve kept the same “you have to wait 15 minutes” behaviour just without any indication that’s what’s happening. Regardless, I now have enough RAM even for the greediest of IDEs and Electron apps.

Bizarrely, the maximum spec has decreased to 16GB since then. ↩︎
In the generic sense. I use Kagi these days. ↩︎
On my phone because, y’know, the laptop was busted. ↩︎

Docker reverse proxying, redux

2022-12-06T00:00:00Z

Six years ago, I described my system for configuring a reverse proxy for docker containers. It involved six containers including a key-value store and a webserver. Nothing in that system has persisted to this day. Don’t get me wrong – it worked – but there were a lot of rough edges and areas for improvement.

Microservices and their limitations

My goal was to follow the UNIX philosophy of “do one thing and do it well”. Unfortunately, that doesn’t really work when applied to network services that have to interact with one and other. UNIX tools are built upon a common file system and simple data passed over STDIN. Microservices don’t have that shared foundation. You could make one: companies that use microservices in anger often have a team that deals with the “developer experience” of creating and using microservices. But as a solo developer that’s not something I wanted to spend my time doing.

This became very apparent when trying to debug issues. In the UNIX world, if your series of commands piped together does something strange you can simply echo the data at various stages. Not so much when that data is flying around on a network, going into and out of things like etcd. Trying to figure out why a certificate hadn’t been acquired was a case of searching through logs from four containers, none of which had particularly good logging. There are many ways to get insight into what’s happening with microservices but, again, that’s not something I wanted to do myself.

Over time, and with experience in other projects, I came to realise that microservices only really make sense if you’re unable to deploy a monolith. For tech companies this naturally happens when different teams contribute to the same service: splitting it out into smaller services that are wholly owned by individual teams makes sense. For solo developers, that never happens. You can still gain the other benefits of microservices – such as code separation and having clearly defined APIs – by sticking to certain coding standards.

Proxy inconveniences

As well as being unhappy with the microservice nature of the solution, I wasn’t pleased with nginx. If you requested an unknown domain, nginx would use the first server block in its config to serve a response, instead of sending an “unrecognised name” alert as I wanted. It was a minor issue, but it irked me.

So from nginx I switched to haproxy. It has a strict-sni option when configuring TLS connections which makes it behave properly. It also performs a lot better for this type of workload than nginx. All was well for a while, but then I started getting alerts that requests were occasionally failing. I couldn’t reproduce the issue, but my nightly jobs to build and push containers managed to hit it nearly every night, causing them to fail.

After some investigation, I found that the haproxy developers had refactored the header parsing code, and neglected to properly reset flags when multiple requests were sent over the same connection. There was a patch, but it wasn’t released. No problem, I thought, I’ll just cherry-pick it onto the last release… Except that haproxy use Git in the most convoluted manner I’ve ever seen – they have one repository per release. This makes it harder to patch, but it also made me question whether I trusted them to ship stable software: there were no tests for the header parsing code (which is both fundamental and finicky, the perfect target for tests), the source code management was weird, and they didn’t seem in any rush to patch this bug.

Not long after that issue, Greg managed to encounter another bug where haproxy returned a 500 error whenever the upstream server replied with a particular, perfectly valid, header. The die was cast – it was time to move to something else.

Not Invented Here syndrome

Looking for a new solution, there were many more options than back in 2016. I’m still convinced, however, that anything exposed to the Internet should not have access to run docker containers. It’s the modern equivalent of running a CGI script as root. That single requirement eliminates most off-the-shelf solutions. What do you do when nothing quite meets your specific requirements? You make something yourself! My new solution has two components: Dotege and Centauri.

Dotege is a replacement for the microservices that monitored containers and obtained certificates. It’s fundamentally a templating engine - whenever the containers change, it evaluates a template and saves the result to disk. The template has access to details about the containers, their labels, ports, and so on. Dotege can also obtain certificates from Let’s Encrypt, and raise a signal against another process whenever the template or certificates change. I used this to generate the configuration and certificates used by haproxy for a while, and more recently changed the template so that it works for Centauri.

Centauri is my own reverse proxy. It’s configured using a simple text file and can also obtain certificates from an ACME provider. It doesn’t serve static content, has no knowledge about docker, and avoids the other bells and whistles that adorn most reverse proxies. It also has good test coverage to ensure that I don’t, say, accidentally break header parsing.

As a software engineer I enjoy writing software, but I also enjoy running simple, easy to understand software. That’s what I’ve achieved here: it’s very easy to identify where the problem is if anything goes wrong, both are small Go programs rather than vast sprawling C monstrosities, and their interaction is primarily through a file written to disk that can be inspected or edited as needed.

Adventures in IPv6 routing in Docker

2022-10-24T00:00:00Z

One of the biggest flaws in Docker’s design is that it wasn’t created with IPv6 in mind. Out of the box Docker assigns each container a private IPv4 address, and they won’t be able to reach IPv6-only services. While incoming connections might work, the containers won’t know the correct remote IP address which can cause problems for some applications. This situation is obviously suboptimal in the current day and age. It’s a bit like not supporting HTTPS on a website — you might not have any issues because of it immediately, but you’re fighting against the currents of progress and are making life worse for your users.

Thankfully, it’s now relatively easy to make Docker behave a lot nicer. The docker-ipv6nat project has been around since 2016, and uses an IPv6 overlay network and some iptables magic to route traffic to and from containers in a sensible fashion. It uses NAT to emulate the behaviour Docker employs for IPv4 traffic; while using NAT with IPv6 is an anathema, I think it makes sense for containers. You could give each container a publicly routable IPv6 address, but that brings with it a lot of headaches: you’re basically going to be forced to implement service discovery and some kind of DNS management to deal with the fact that your containers will be popping up on randomly assigned IP addresses. That is completely overkill for people running a small number of services on one or two physical boxes; and if it’s not overkill for you then you’re probably already looking at more complicated orchestration solutions like Kubernetes.

More recently, similar functionality has been built into the Docker daemon itself. You can now edit the config file to enable ipv6 and each container will be assigned an address in the range specified when it uses the default bridge network. This gives more-or-less the same functionality of docker-ipv6nat — you lose a little flexibility as you can’t disable IPv6 on the default bridge, but that’s a very worthy trade for having the functionality built-in.

So far this all seems very simple. Hardly worthy of being called an “adventure”. Enter stage left: the wicked witch of destination address selection…

Destination address selection and you

When a computer program tries to connect to an address such as google.com, it first resolves it in to an IP address. That’s DNS 101, but what happens if the address resolves to multiple IP addresses? For example, google.com resolves to both 142.250.74.206 and 2a00:1450:4001:82b::200e. You might just assume there’s a simple “prefer IPv4 / prefer IPv6” toggle somewhere that decides, but it’s actually a lot more complicated. With IPv6, devices are likely to have many addresses - a link-local address, a unique local address, a normal public address, various privacy addresses, and so on. To deal with this, a number of RFCs define a series of rules that most implementations follow. These are called the destination address selection rules. RFC6724 gives the rules as:

Avoid unusable destinations
Prefer matching scope
Avoid deprecated addresses
Prefer home addresses
Prefer matching label
Prefer higher precedence
Prefer native transport
Prefer smaller scope
Use the longest matching prefix
Otherwise, leave the order unchanged

Applying these rules will re-order the list of IP addresses such that (in theory) the most likely one to work will be first. Most of them are fairly niche rules; the ones that do the heavy lifting are rules 5 and 6, which rely on a policy table to make their decisions. The policy table ‘SHOULD’ be configurable by system administrators to allow them to tweak how traffic is routed. In the absence of an admin-provided policy table, the RFC gives the following defaults:

Prefix	Precedence	Label	Notes
`::1/128`	50	0	Loopback address
`::/0`	40	1	Any IPv6 address
`::ffff:0:0/96`	35	4	IPv4 addresses mapped as v6 addresses
`2002::/16`	30	2	6-to-4 gateways
`2001::/32`	5	5	Toredo tunnels
`fc00::/7`	3	13	Unique Local Addresses (ULAs)
`::/96`	1	3	IPv4 compatible addresses (deprecated)
`fec0::/10`	1	11	Site-local addresses (deprecated)
`3ffe::/16`	1	12	6bone (deprecated)

The policy table is a bit complex, but you can see that normal IPv6 addresses are preferred (have a higher precedence) over IPv4 addresses, which are preferred over the various tunnels, local addresses, and various deprecated ranges.

We can see the result of this when I run ping google.com on a box that has a native IPv6 connection as well as a native IPv4 connection:

PING google.com(fra07s29-in-x200e.1e100.net (2a00:1450:4001:802::200e)) 56 data bytes
64 bytes from fra24s01-in-x0e.1e100.net (2a00:1450:4001:802::200e): icmp_seq=1 ttl=119 time=5.11 ms
64 bytes from fra24s01-in-x0e.1e100.net (2a00:1450:4001:802::200e): icmp_seq=2 ttl=119 time=5.28 ms
64 bytes from fra07s29-in-x200e.1e100.net (2a00:1450:4001:802::200e): icmp_seq=3 ttl=119 time=5.31 ms
^C
--- google.com ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2003ms
rtt min/avg/max/mdev = 5.114/5.235/5.311/0.086 ms

The address selection in this instance has been a result of rule 6: the IPv6 address has a higher precedence (40) than the IPv4 address (35).

However, if I run the same command in an IPv6-enabled Ubuntu container then it seems to prefer the IPv4 address:

PING google.com (172.217.16.206) 56(84) bytes of data.
64 bytes from fra16s08-in-f206.1e100.net (172.217.16.206): icmp_seq=1 ttl=59 time=4.86 ms
64 bytes from fra16s08-in-f14.1e100.net (172.217.16.206): icmp_seq=2 ttl=59 time=4.86 ms
64 bytes from fra16s08-in-f14.1e100.net (172.217.16.206): icmp_seq=3 ttl=59 time=5.03 ms
^C
--- google.com ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2003ms
rtt min/avg/max/mdev = 4.860/4.916/5.027/0.078 ms

If I force ping to use IPv6 with the -6 flag then it works the same as on the host, but when left to its own devices it prefers IPv4. What’s going on there? The key difference between the Docker container and the host is that the host’s network interfaces have public addresses, but the container has a private IPv4 address (172.19.0.7) and a private IPv6 address (fd00:dead:beef::7). My first reaction to this was to think “Ah yes, fc00::/7 has a lower precedence than native IPv4, that makes sense”, but that’s not quite right. These are the destination address selection rules; rule 6 doesn’t care about the source addresses. This is actually rule 5 at work: the label of source IPv6 address is 13, but the label of the destination address is 1; meanwhile both the source and destination IPv4 addresses are 4. This sorts the IPv4 address before the IPv6 one, and rule 6 becomes irrelevant.

Adjusting the policy table

Thankfully, the RFC says the policy table should be configurable by system administrators, so those of us who are doing unorthodox things like NAT’ing IPv6 can customise the behaviour to fit our weird environments. The configuration is done via the /etc/gai.conf file (‘gai’ standing for getaddrinfo, the function in the standard library responsible for dealing with all these rules). The gai.conf file by default (if it exists) will likely just contain comments and examples; if there are no uncommented “label” or “precedence” lines then the library will use its built-in defaults based on the RFC requirements.

To make our container happy, we need to cause fd00::/8 to have the same label as a public IPv6 address. To do this we can uncomment the default labels in gai.conf and then add a single extra line, like so:

 label ::1/128       0
 label ::/0          1
 label 2002::/16     2
 label ::/96         3
 label ::ffff:0:0/96 4
 label fec0::/10     5
 label fc00::/7      6
 label 2001:0::/32   7
+label fd00::/8      1

Because fd00::/8 has a longer prefix than fc00::/7 it will match our addresses and give them a label of 1, the same as a public IPv6 address would get. This makes rule 5 leave it alone, and the default precedence table used by the standard library will put the IPv6 address above the IPv4 address. (They don’t bother giving fc00::/7 a separate precedence as listed in the RFC because rule 5 would have already de-prioritised those addresses, as we discovered.)

Making this small change to the gai.conf file in our Ubuntu container makes it start preferring the IPv6 address for google.com:

PING google.com(fra16s65-in-x0e.1e100.net (2a00:1450:4001:806::200e)) 56 data bytes
64 bytes from fra15s29-in-x0e.1e100.net (2a00:1450:4001:806::200e): icmp_seq=1 ttl=118 time=5.00 ms
64 bytes from fra15s29-in-x0e.1e100.net (2a00:1450:4001:806::200e): icmp_seq=2 ttl=118 time=5.10 ms
64 bytes from fra16s65-in-x0e.1e100.net (2a00:1450:4001:806::200e): icmp_seq=3 ttl=118 time=5.08 ms
^C
--- google.com ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2004ms
rtt min/avg/max/mdev = 5.002/5.058/5.098/0.040 ms

A fairly complicated problem, but a simple fix!

But wait, there’s more…

Unfortunately, this fix won’t work on containers that use alpine. And that is a lot of containers. Alpine uses musl as its standard library, rather than the much more common glibc. There is very little documentation on the subject, but if you browse the source code for musl you will see that it doesn’t implement any way at all to configure the policy tables. In network/lookup_name.c we can see the sorting logic:

	/* The following implements a subset of RFC 3484/6724 destination
	 * address selection by generating a single 31-bit sort key for
	 * each address. Rules 3, 4, and 7 are omitted for having
	 * excessive runtime and code size cost and dubious benefit.
	 * So far the label/precedence table cannot be customized. */
	for (i=0; i<cnt; i++) {
		int family = buf[i].family;
		int key = 0;
		struct sockaddr_in6 sa6 = { 0 }, da6 = {
			.sin6_family = AF_INET6,
			.sin6_scope_id = buf[i].scopeid,
			.sin6_port = 65535
		};
		struct sockaddr_in sa4 = { 0 }, da4 = {
			.sin_family = AF_INET,
			.sin_port = 65535
		};
		void *sa, *da;
		socklen_t salen, dalen;
		if (family == AF_INET6) {
			memcpy(da6.sin6_addr.s6_addr, buf[i].addr, 16);
			da = &da6; dalen = sizeof da6;
			sa = &sa6; salen = sizeof sa6;
		} else {
			memcpy(sa6.sin6_addr.s6_addr,
				"\0\0\0\0\0\0\0\0\0\0\xff\xff", 12);
			memcpy(da6.sin6_addr.s6_addr+12, buf[i].addr, 4);
			memcpy(da6.sin6_addr.s6_addr,
				"\0\0\0\0\0\0\0\0\0\0\xff\xff", 12);
			memcpy(da6.sin6_addr.s6_addr+12, buf[i].addr, 4);
			memcpy(&da4.sin_addr, buf[i].addr, 4);
			da = &da4; dalen = sizeof da4;
			sa = &sa4; salen = sizeof sa4;
		}
		const struct policy *dpolicy = policyof(&da6.sin6_addr);
		int dscope = scopeof(&da6.sin6_addr);
		int dlabel = dpolicy->label;
		int dprec = dpolicy->prec;
		int prefixlen = 0;
		int fd = socket(family, SOCK_DGRAM|SOCK_CLOEXEC, IPPROTO_UDP);
		if (fd >= 0) {
			if (!connect(fd, da, dalen)) {
				key |= DAS_USABLE;
				if (!getsockname(fd, sa, &salen)) {
					if (family == AF_INET) memcpy(
						sa6.sin6_addr.s6_addr+12,
						&sa4.sin_addr, 4);
					if (dscope == scopeof(&sa6.sin6_addr))
						key |= DAS_MATCHINGSCOPE;
					if (dlabel == labelof(&sa6.sin6_addr))
						key |= DAS_MATCHINGLABEL;
					prefixlen = prefixmatch(&sa6.sin6_addr,
						&da6.sin6_addr);
				}
			}
			close(fd);
		}
		key |= dprec << DAS_PREC_SHIFT;
		key |= (15-dscope) << DAS_SCOPE_SHIFT;
		key |= prefixlen << DAS_PREFIX_SHIFT;
		key |= (MAXADDRS-i) << DAS_ORDER_SHIFT;
		buf[i].sortkey = key;
	}
	qsort(buf, cnt, sizeof *buf, addrcmp);

We can see if the address labels match, the sort key is adjusted using the DAS_MATCHINGLABEL constant. But where do the labels come from? For that we need to investigate the policyof func:

static const struct policy {
	unsigned char addr[16];
	unsigned char len, mask;
	unsigned char prec, label;
} defpolicy[] = {
	{ "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\1", 15, 0xff, 50, 0 },
	{ "\0\0\0\0\0\0\0\0\0\0\xff\xff", 11, 0xff, 35, 4 },
	{ "\x20\2", 1, 0xff, 30, 2 },
	{ "\x20\1", 3, 0xff, 5, 5 },
	{ "\xfc", 0, 0xfe, 3, 13 },
#if 0
	/* These are deprecated and/or returned to the address
	 * pool, so despite the RFC, treating them as special
	 * is probably wrong. */
	{ "", 11, 0xff, 1, 3 },
	{ "\xfe\xc0", 1, 0xc0, 1, 11 },
	{ "\x3f\xfe", 1, 0xff, 1, 12 },
#endif
	/* Last rule must match all addresses to stop loop. */
	{ "", 0, 0, 40, 1 },
};

static const struct policy *policyof(const struct in6_addr *a)
{
	int i;
	for (i=0; ; i++) {
		if (memcmp(a->s6_addr, defpolicy[i].addr, defpolicy[i].len))
			continue;
		if ((a->s6_addr[defpolicy[i].len] & defpolicy[i].mask)
		    != defpolicy[i].addr[defpolicy[i].len])
			continue;
		return defpolicy+i;
	}
}

So the policyof func simply uses the defpolicy array to find the policy that applies to an address. This defpolicy array contains a number of hardcoded entries which correspond exactly to the ones provided in the RFC. The one causing us problems is { "\xfc", 0, 0xfe, 3, 13 } which matches fc00::/7 addresses.

Unfortunately, without recompiling musl from source there’s not much we can do to address this directly. The only way we can make the containers work as desired is to give them a different address range. Ideally this would be a range that you control and that is otherwise not used, but there’s a good chance you won’t have such a range. One possible alternative is the 2001:db8::/32 range which is reserved for documentation purposes. This doesn’t feature in any of the policy tables, so is treated like a normal public IPv6 address. It doesn’t feel great to misuse a reserved range like that, but it’s probably the least of all evils, at least until musl allows configuring the policy table.