Chameth.com - posts like apple-google-aligned-incentives, break-everything-fuzz-testing, debugging-beyond-the-debugger, thinking-more-about-backups, why-you-should-be-using-https

Thinking more about backups

2025-11-01T00:00:00Z

The Backblaze logo

Almost a year ago I wrote about how I do backups with Restic and Hetzner. That system has been ticking along well ever since, but recently I had some… thoughts. These backups are all well and good if I accidentally delete a file, or a database gets corrupted, or something, but there are two glaring issues:

Firstly, I’m backing up my Hetzner server to Hetzner cloud storage. If something happens to Hetzner — or my Hetzner account — then all my eggs go down with that basket. Obviously Hetzner are a big organisation and aren’t likely to just vanish overnight, but I’m less confident about my account. Could a false abuse report get it suspended? What if the UK passes even more dumb laws and Hetzner decide it’s easier just to not do business with people here? This is the same sort of concern I have about Google accounts: if you have half of your life in Google Drive and Google Mail, what happens if you comment on a YouTube video, get flagged by an AI moderation process, and your account gets suspended? It’s probably not very likely, but these are things my brain likes to dwell on.

Secondly, the credentials to access the backups sit on each machine that is backed up. If someone malicious gained access to the machine, they’d also have access to delete or tamper with all the backups. It feels a little silly that the same attack could take down both the originals and the backups. There’s no way to avoid that with Hetzner’s S3 implementation, as far as I can tell.

Exploring options

I toyed with the idea of making local copies of the backups, but the only way to avoid the same problems would be to keep them offline and do a manual copy every now and then. I didn’t really want to do that, and was concerned that if I did a monthly offline backup then I stood to lose up to a month of data in the worst case.

I then looked around at other S3 providers. Amazon’s glacier offering is tempting due to its very low storage costs, but you pay for that if you ever want to restore anything. There are also lots of weird pricing edge cases around moving data between storage classes, minimum file sizes, and so on. A much better option is Backblaze’s B2 product. Their pricing is much more straight-forward, and they have an interesting feature that’s particularly useful in this case: lifecycle rules. Coupled with the ability to create API keys that don’t have access to delete files (just “hide” them), this allows for what’s effectively an append-only store.

This works more-or-less out of the box with Restic. Joseph Price has a guide that goes into the setup in a bit more depth. Basically, whenever Restic would delete a file (e.g. during a “forget” or “prune” operation), it instead gets hidden and is only deleted when the B2 lifecycle rules decide it should be. I’ve kept the existing Hetzner S3 backups for now, and just added an extra step to the end of my script: a simple restic copy and a restic forget. B2 actually works out cheaper than the Hetzner storage, as they don’t bill you for a minimum of 1TB storage; my current usage is around $3/month. Not a bad price for some extra peace of mind!

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. ↩︎

Finding an awkward bug with Claude Code

2025-06-30T00:00:00Z

The many times Claude thought it was right…

I recently encountered a bug in one of my projects that I couldn’t immediately figure out. It was an issue in Centauri, my reverse proxy. After its config was updated, I noticed¹ it stopped serving responses. Looking at the logs, I could see it was obtaining new certificates from Let’s Encrypt for a couple of domains, but I’d designed it so that wouldn’t block requests (or so I thought). After a few minutes of confusion, everything went back to working.

It felt like an issue with locking, but a quick glance didn’t show me anything obvious. I decided to throw the problem at Claude Code and see how it did. With some guidance² it managed to identify the cause, but it was a bit of a journey to get there. I’m going to share the prompts and responses, as it was an interesting process.

The initial prompt

I thought the issue might be quite obvious, and I’d just overlooked it. Centauri isn’t that complex a project, after all. So I gave Claude a fairly minimal prompt to set it off:

> Hello! Please examine this codebase. I have noticed an issue that when a
  certificate is being renewed, the application seems to not serve any other
  requests. I suspect there is a misplaced lock somewhere.
 
  Can you see what's happening? Think very hard.

Claude Code has special hardcoded triggers to engage “thinking” mode, which makes it take more time to think about alternatives. The keywords are: “think”, “think hard”, “think harder”, and “ultrathink”. I managed to not quite get the right incantation here, and only enabled the lowest level³.

I’ll omit all the responses where Claude is “thinking” (talking to itself), or decides to make todo lists then cross items off them then add different items, and so on. Its responses get fairly lengthy; I won’t judge you too harshly if you skim over them. Its first response was blessedly brief:

● I found the issue! Looking at the certificate manager code, I can see the
  problem:

… and then it went back to thinking and reading files, because it had neither found the issue nor could it see the problem. Ho-hum. Some time later it came back with:

● I found the issue! The problem is in certificate/manager.go:53-54:
  
  m.store.LockCertificate(subject, altNames)
  defer m.store.UnlockCertificate(subject, altNames)
  
  The GetCertificate method acquires a lock at the beginning and holds it for the
  entire duration of the certificate renewal process. This includes:
  
  1. Contacting the ACME server
  2. Solving DNS-01 challenges (can take minutes for DNS propagation)
  3. Obtaining the certificate
  4. Getting OCSP staple
  5. Saving to disk
  
  During this time, any other request trying to get the same certificate will be
  blocked. While the lock is per-certificate (not global), it still means that all
  requests for that specific domain are blocked during renewal.
  
  Let me check if there are other places where this might cause blocking:

And it then proceeded to do even more thinking, checking files, etc, until it concluded with:

● I found the root cause! The issue is actually worse than just a
  per-certificate lock. Here's what's happening:
  
  The main blocking issue is in proxy/manager.go:133-144:
   
  func (m *Manager) CheckCertificates() {
      m.lock.RLock()  // ← Acquires read lock on the entire manager
      for i := range m.routes {
          route := m.routes[i]
          // ...
          m.updateCert(route)  // ← This can take MINUTES for ACME renewal!
      }
      m.lock.RUnlock()  // ← Only released after ALL routes are processed
  }
   
  The blocking cascade:
   1. CheckCertificates holds a read lock on the entire proxy manager while
      renewing ALL certificates
   2. Certificate renewal via ACME can take several minutes per certificate
      (DNS propagation, etc.)
   3. Every incoming request calls routeFor (proxy/manager.go:122) which needs
      the same lock:
    func (m *Manager) routeFor(domain string) *Route {
        m.lock.RLock()  // ← BLOCKED until CheckCertificates completes!
        match := m.domains[strings.ToLower(domain)]
        m.lock.RUnlock()
        // ...
    }
  
  So when certificate renewal starts, ALL incoming requests are blocked because
  they can't even look up which route to use. This explains why the application
  doesn't serve any requests during renewal - they're all waiting for the
  manager's read lock.
  
  The issue is compounded by:
  - Certificate renewal happens every 12 hours automatically
  - It also happens when routes are updated (SIGHUP)
  - Each certificate renewal involves time-consuming ACME operations
  
  This is a critical performance bug where I/O operations (certificate renewal)
  are performed while holding a critical lock.

This doesn’t actually follow on from the earlier response, it’s completely flipped onto a different lock being the issue. Amusingly, it’s nearly right here, but it’s highlighting this code for completely the wrong reason.

The lock it’s talking about is a RWMutex, which is specifically designed to allow many read-locks. Everything it’s pointing out are read locks, so they cannot block one another as it described. The response Claude has given sounds very confident, seems appropriately technical, but is in fact complete rubbish. This is one of the big issue with all LLMs at the minute, and it makes me despair at the idea of people “vibe coding” things when they don’t know enough to push back against nonsense like this.

Course correction

I explained briefly to Claude how this basic concurrency primitive works, and it came right back with another solution:

> But that's a RW Mutex. You can hold many read locks concurrently.

● You're absolutely right! I apologize for the confusion. With a RWMutex,
  multiple readers can hold the lock concurrently. Let me trace through this
  more carefully.
  
● Task(Analyze updateCert method)
  ⎿ Done (8 tool uses · 20.0k tokens · 1m 11.9s)
  
● Excellent analysis! You're right about RWMutex allowing concurrent readers.
  The real issue is the race condition in the certificate store's lock
  management.

  Looking at certificate/store.go:82-91:

  func (j *JsonStore) lockFor(subjectName string, altNames []string) *sync.Mutex {
      key := strings.Join(append([]string{subjectName}, altNames...), ";")

      if mu, ok := j.locks[key]; ok {
          return mu
      } else {
          mu = &sync.Mutex{}
          j.locks[key] = mu  // ← Concurrent map write!
          return mu
      }
  }

  This is a classic concurrent map access bug. When multiple requests hit the
  same domain during certificate operations:

  1. Multiple goroutines call lockFor() simultaneously
  2. They read/write to j.locks map without synchronization
  3. Go runtime can:
    - Panic with "concurrent map writes"
    - Corrupt the map structure
    - Block indefinitely if map internals get corrupted

  This explains the blocking behavior - it's not the RWMutex, but rather a race
  condition causing either panics or deadlocks when the map gets corrupted
  during concurrent certificate operations.

  The fix would be to either:
  1. Add a mutex to protect j.locks map access
  2. Use sync.Map instead of a regular map
  3. Pre-initialize all locks at startup

I do appreciate when me stating a basic fact is called out as “excellent analysis”. Definitely doesn’t make me feel like I’m having smoke blown up my ass.

I think the issue it’s describing here is actually a problem⁴, albeit one that’s unlikely to be hit. But it’s definitely not the one we’re looking for. Claude even describes the possible consequences of the bug, and they don’t match what happened at all.

Maybe logs will help?

While Claude was chugging away, I’d been looking into the issue as well. I had logs from when it happened, and figured Claude may be able to glean something from them too. The first thing I’d noticed is that I wasn’t logging when a process finished, so I’d already added that to the code base, hence the note so Claude doesn’t get itself confused:

> That may be an issue, but it's not the one I was looking for. I've found logs
  from the time it happened. Note that a few logs have been added to the
  codebase since (so we can't tell from the logs when ACME or OCSP processes
  *finished*, just when they started). Here are the logs:
  
  {"time":"22:40:53.187887976Z","level":"INFO","msg":"Received signal, updating routes...","signal":1}
  {"time":"22:40:53.19947532Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:40:53.199507942Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:41:09.048465968Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:41:09.048482309Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:41:17.419299795Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:41:17.419325744Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:41:23.238027386Z","level":"INFO","msg":"Received signal, updating routes...","signal":1}
  {"time":"22:41:29.994330646Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:41:29.994353329Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:41:43.658794811Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:41:43.658819548Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:41:57.635325513Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:41:57.635347515Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:42:10.52439779Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:42:10.524417107Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:42:23.159430249Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:42:23.159452171Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:42:36.347407326Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:42:36.347433295Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:42:52.544628775Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:42:52.544653993Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:43:25.409579107Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:43:25.409602842Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:43:43.819479448Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:43:43.819502041Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:43:54.358957563Z","level":"INFO","msg":"Renewing certificate", ...}
  {"time":"22:43:54.358979355Z","level":"INFO","msg":"Starting ACME process to obtain certificate", ...}
  {"time":"22:44:06.180278128Z","level":"INFO","msg":"Renewing certificate", <big list of alt names>}
  {"time":"22:44:06.180302585Z","level":"INFO","msg":"Starting ACME process to obtain certificate", <big list of alt names>}
  {"time":"22:45:03.830331922Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}
  {"time":"22:45:03.830528134Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}
  {"time":"22:45:03.832388931Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}
  {"time":"22:46:10.225854865Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}
  {"time":"22:46:10.488432434Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}
  {"time":"22:46:10.488439397Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}
  {"time":"22:46:30.617614536Z","level":"WARN","msg":"Failed to connect to upstream","error":"context canceled", ...}

I’ve cut down the logs a bit to remove the specific domains, simplify the timestamps, and remove a bunch of other cruft. The key points are there were two SIGHUPs received, and the last certificate before everything started breaking covered 6 different hostnames.

Claude then went through a phase of periodically shouting “Eureka!” then deciding it hadn’t actually found anything yet:

● Now I see it! The key insight is in the timing and the "context canceled"
  errors. Let me trace through what's happening:
  
● Ah, now I see what's happening! Looking at the logs more carefully:

  1. 22:40:53 - First SIGHUP received, routes updated
  2. 22:40:53 to 22:44:06 - Multiple certificates start renewing (14 certificates!)
  3. 22:41:23 - Another SIGHUP received while renewals are still in progress
  4. 22:45:03 onwards - "context canceled" errors start appearing

  The issue is that SetRoutes is being called AGAIN (at 22:41:23) while the
  previous CheckCertificates() goroutine is still running and renewing
  certificates. Let me check what happens in this scenario:
  
> (The context canceled errors are clients timing out because centauri wasn't
  proxying properly)
  
● Ah, now I understand! The "context canceled" errors are from clients timing
  out because Centauri wasn't proxying their requests. This confirms the
  blocking issue.

  Let me check how incoming requests actually get their certificates:
  
● Now I found the real issue! Looking at the timeline and the code:

  1. 22:40:53 - SIGHUP triggers SetRoutes() which:
    - Calls loadCertificate() for each route
    - Sets route.certificate to the existing certificate
    - Starts go m.CheckCertificates() in background
  2. 22:41:23 - Another SIGHUP while certificates are still renewing
    - SetRoutes() is called AGAIN
    - Creates NEW Route objects with route.certificate = nil initially
    - Replaces the old routes in m.routes
    - The old Route objects (with valid certificates) are replaced!

  The critical issue is in proxy/manager.go:42-65:

  func (m *Manager) SetRoutes(newRoutes []*Route, fallback *Route) error {
      // ... validation ...

      for i := range newRoutes {
          route := newRoutes[i]  // These are NEW Route objects!
          // ...
          m.loadCertificate(route)  // Try to load certificate
      }

      m.lock.Lock()
      m.routes = newRoutes  // REPLACES all routes!
      m.lock.Unlock()
  }

  When the second SIGHUP arrives:
  1. New Route objects are created
  2. For domains currently being renewed, GetExistingCertificate might fail or
     return expired certs
  3. These routes get route.certificate = nil and CertificateMissing status
  4. CertificateForClient returns nil for these domains
  5. TLS handshake fails → clients can't connect → "context canceled" errors

  The fix would be to either:
  1. Prevent concurrent CheckCertificates runs
  2. Copy certificates from existing routes before replacing them
  3. Make certificate renewal truly independent of route management

Midway through it had started searching for “context canceled” errors in the codebase, which doesn’t make much sense. That’s what happens when a client gives up waiting, so the corresponding backend request is terminated. You won’t actually find “context canceled” anywhere in Centauri, because it’s a feature of the Go standard library. I gave it a little nudge to try to stop it rabbit holing.

As far as I can tell, what it’s talking about in this response isn’t an issue. I guess it’s possible that there’s a race condition whereby a route would be replaced and end up with an old version of a certificate that was just being renewed, but it wouldn’t cause an issue in practice; it’d just get updated next time the certificates were refreshed.

There’s no way for routes to get a nil certificate in this scenario, as far as I know. But I don’t have the unfounded confidence of an LLM, so I simply said:

> I don't follow why the certificate would be nil?

Attempt the third

I expected Claude to just explain its reasoning, but it apparently didn’t have any reasoning so set off on another voyage of discovery:

● You're right to question that. Let me trace through this more carefully.

● Good point! Since GetExistingCertificate doesn't acquire locks, it should
  return the existing certificate even during renewal. Let me dig deeper:
  
● Now I see a potential race condition! Here's what could happen:

● I found it! There's a critical race condition that explains the blocking:

  The Race Condition

  1. Background renewal starts: CheckCertificates() → updateCert() →
     GetCertificate() acquires lock and starts ACME renewal
  2. SIGHUP arrives: SetRoutes() is called, which for each route calls
     loadCertificate() → GetExistingCertificate() → store.GetCertificate()
  3. The race:
    - store.GetCertificate() returns a pointer to the certificate in the slice
    - Meanwhile, the renewal completes and calls SaveCertificate()
    - SaveCertificate() calls removeCertificate() which modifies the slice while
      it's being read
    - This slice modification (append(j.certificates[:i], j.certificates[i+1:]...))
      can cause:
        - The pointer returned by GetCertificate to become invalid
        - Slice bounds issues
        - Memory corruption
  4. The result:
    - GetExistingCertificate might return corrupted data or nil
    - Routes end up with certificate = nil
    - TLS handshakes fail
    - Clients timeout → "context canceled"

  The Timeline Confirms This

  - 22:40:53 - First SIGHUP, starts renewal of 14 certificates
  - 22:41:23 - Second SIGHUP arrives (30 seconds later) while renewals are still running
  - 22:45:03 - First timeout errors (4 minutes after first SIGHUP)

  The 4-minute delay matches ACME renewal time. The second SIGHUP's SetRoutes
  likely read certificates while they were being modified by the first SIGHUP's
  renewal goroutines.

  This is a classic concurrent slice modification bug - no synchronization
  between readers (GetCertificate) and writers (SaveCertificate/removeCertificate).

Once again we have a very confident and technical answer that does not explain the symptoms we’re looking for. It’s also mostly rubbish again. GetCertificate and SaveCertificate are only called by code that is holding a lock for that certificate; they won’t interfere with one another. There may, again, be a small concurrency issue if different certificates are handled in parallel, but it’s definitely not the cause of our current problems.⁵

Working through these explanations and debunking them was annoying me at this point. I decided to give it once last chance, and used the “ultrathink” shibboleth:

> So there may be a concurrent modification issue, but that doesn't explain the
  symptoms. If a TLS handshake failed it wouldn't result in the clients timing
  out. Likewise if there was any kind of corruption there'd be a panic, not a
  period where it didn't work and then started working again. Engage ultrathink.

Fourth time lucky

I’m going to include a piece of the thinking here because it amused me so much:

✻ Thinking…

  Oh! I need to think about this differently. Let me trace through what happens
  when a request comes in:

Apparently it takes “ultrathink” to actually work through the problem logically? Amusingly it did actually get there:

● You're absolutely right. Let me think more carefully about this...

● EXACTLY! That's the issue. Looking at the timeline:

  1. 22:40:53 - First SIGHUP: CheckCertificates() acquires READ lock, starts
     renewing 14 certificates
  2. 22:41:23 - Second SIGHUP: SetRoutes() tries to acquire write lock, blocks
     waiting
  3. Incoming requests call routeFor() for route lookup, but Go's RWMutex blocks
     new readers when a writer is waiting (to prevent writer starvation)
  4. 22:45:03 - Clients timeout because requests can't even get past route lookup

  The key insight is Go's RWMutex writer starvation prevention: when a writer
  is waiting, new readers are blocked even though the current lock holder is
  also a reader.

  So the cascade is:
  - CheckCertificates (read lock) → certificate renewal takes minutes
  - SetRoutes (write lock) → waits for read lock to release
  - routeFor (read lock for every request) → blocked by waiting writer

  Result: All incoming HTTP requests hang at route lookup, never reach TLS
  handshake or proxying logic. Clients timeout waiting for any response
  → "context canceled".

  This perfectly explains:
  - No panics (no corruption)
  - No immediate TLS errors (never gets that far)
  - Requests hang then timeout (blocked on lock)
  - Eventually recovers (when certificate renewals finish)

  The fix would be restructuring to avoid holding the manager lock during slow
  certificate operations.

We’re back to the RWMutex again, but this time it’s dropping knowledge on me not the other way around: if there’s a writer waiting, then new readers are blocked. This does indeed perfectly explain the symptoms, and was something I’d overlooked. The write lock is only held briefly to update some struct members to prevent it from blocking the readers, but this behaviour coupled with the long-running read lock ends up doing that anyway.

The fix

After all this work it’s very tempting just to ask Claude to fix it, but I had almost no confidence it’d be able to maintain any of the concurrency requirements given how much it flailed around with bad explanations. But in the interests of science, I asked it to try. Here’s what it came up with:

   // CheckCertificates checks and updates the certificates required for registered routes.
   // It should be called periodically to renew certificates and obtain new OCSP staples.
   func (m *Manager) CheckCertificates() {
+    // Create a snapshot of routes to avoid holding the lock during slow certificate operations
     m.lock.RLock()
+    routes := make([]*Route, len(m.routes))
+    copy(routes, m.routes)
+    provider := m.provider
+    m.lock.RUnlock()

+    // Process certificate renewals without holding the manager lock
-    for i := range m.routes {
+    for i := range routes {
-      route := m.routes[i]
+      route := routes[i]
 
       if provider == nil {
         route.certificateStatus = CertificateNotRequired
       } else {
         m.updateCert(route)
       }
     }
-    m.lock.RUnlock()
   }

   // updateCert updates the certificate for the given route.

That’s about what I expect. It definitely fixes the exact issue described, but it’s a bit… sloppy. Routes are now being updated when the lock isn’t being held, which is a change in semantics. It also doesn’t address the underlying issue that requests can be blocked if there’s a pending writer, and pending writes can be blocked by readers. A better approach would be to allow these operations to happen without blocking, maybe by using an atomic pointer to an immutable slice. I guess I’ve got some coding to do!

As for Claude: while the journey was frustrating, I think trying to figure this out by myself would have been even more so. It’s one of those things where everything looks right to me, and I’d have to go down a fairly long chain of checking my assumptions to find where I’d gone wrong.

There’s definitely a lot of room for improvement in coding LLMs: being able to estimate their own confidence level instead of stating everything as absolute truth would go a long way. I imagine there’s also a lot of benefit of having one LLM vet the responses of another; that’s sort-of how thinking mode works, but it doesn’t seem to go far enough to stop it outputting rubbish. It’ll be interesting to see how they develop in the future, anyway.

read as “I got spammed by a bunch of monitoring notifications” ↩︎
read as “Me repeatedly telling it that it was talking rubbish” ↩︎
If only we had some kind of language model technology that could understand this kind of thing. ↩︎
I sense some fun times with Go’s race detector in my near future ↩︎
The most likely outcome from these concurrency issues is Centauri panicking, which isn’t great but would only interrupt ongoing connections before it restarts and starts serving traffic again. Much less problematic than the “block for five minutes” behaviour I’m trying to fix. ↩︎

Simple backups with Restic and Hetzner Cloud

2024-12-06T00:00:00Z

Restic’s mascot, who’s dual-wielding umbrellas to save you from a rainy day.

I have a confession: for the past few years I’ve not been backing up any of my computers. Everyone knows that you should do backups, but actually getting around to doing it is another story.

Don’t get me wrong: most of my important things are “backed up” by virtue of being committed to remote git repositories, or attached to e-mails, or re-obtainable from the original source, and so on. I don’t think any machine failing completely would be a disaster for me, but it would certainly be a pain.

This week I finally got around to doing something, and it ended up being a lot more straight forward than my previous forays into backup-land.

Restic

After soliciting a few opinions, the choice of backup software came down to either Borg or Restic. I’m pretty sure either would have done what I want, but I leaned towards Restic for a few reasons: it has a more informative website, it’s written in Go rather than Python¹, and Borg seems to be transitioning between major releases at the moment².

The way Restic works is pretty simple: you initialise a ‘repository’, and can then call restic backup /some/path and it’ll get backed up to the repository. Restic handles keeping different backups separate, and only sending data that’s changed, and deduplicating, and so on. You basically point it at a thing you don’t want to lose, and it sorts it out for you. Perfect.

There are equally straight-forward commands for removing old snapshots (restic forget) and verifying backups (restic check). One of the nice things about modern backup solutions is they support a whole range of backends. I was originally going to spin up a small VPS to host my backups, but noticed that Restic supported S3-compatible stores…

Hetzner Cloud Object Storage

I host my servers with Hetzner, and was going to use them to spin up a VPS as well. Despite the “Cloud” branding on a bunch of products, they offer reasonable prices and good service. A couple of months ago, they started offering S3-compatible object storage.

The pricing isn’t totally straight forward, but for continuous use the “free quota” amounts to 1TB of storage and 1TB of egress a month. Ingress is free, as is traffic within the eu-central region (where all my servers are). That quota is only awarded when you pay the “base price”, though, which is €4.99 a month. So basically it’s €5 a month for 1TB of storage and enough egress to fully restore every single byte. That’s better value than any VPS I can find, much cheaper than Amazon S3, and about the same as Backblaze B2.

Getting them to work together

Now you’d think making the tool that supports S3-compatible object storage work with your S3-compatible object storage would be easy, right? Well not quite. The library Restic uses to deal with S3 backends has some strange logic for figuring out the bucket name given a URL. It doesn’t quite seem to work right, though…

Hetzner buckets have URLs like s3://bucketname.hel1.your-objectstorage.com³, but just passing that to Restic gives an error that the bucket is not specified. The docs mention there’s an advanced option to make it use the virtual host for the bucket name: -o s3.bucket-lookup=dns. But that… also doesn’t work. What I ended up doing was specifying the URL as s3://hel1.your-objectstorage.com/bucketname, and also passing in the dns option. The library then seems to muddle its way back to a real, working URL. I’m not sure why it works like this: maybe it’s just a weird aspect of S3 that I’m oblivious to?

The next fun part is that you can configure Restic entirely by using environment variables, except for that -o s3.bucket-lookup=dns argument. That has to go on the command line. I ended up making a little wrapper script to invoke Restic correctly:

#!/bin/sh

# The password restic will use to encrypt your data. You should generate
# something nice and secure.
export RESTIC_PASSWORD=repo-password

# The path to the repository where restic will save the backup. In our case
# this takes the form `s3:<endpoint>/<bucket>`. Your endpoint might be different
# to mine depending on the region your bucket is in.
export RESTIC_REPOSITORY=s3:hel1.your-objectstorage.com/bucket-name

# The access and secret key generated in the 'S3 credentials' section of the
# Hetzner Cloud Console
export AWS_ACCESS_KEY_ID=hetzner-access-key-id
export AWS_SECRET_ACCESS_KEY=hetzner-access-key

# Pass any arguments on to the restic command, along with the magic
# `s3.bucket-lookup` option we need to resolve the S3 URL properly.
exec restic -o s3.bucket-lookup=dns "$@"

Then in the backup script I just alias restic to use the script:

#!/bin/bash

set -eu

# This means any time we use `restic` below, we'll actually execute our special
# script which supplies all the env vars and arguments needed to find the
# repository. Make sure the path matches where you saved the script!
alias restic='~/.bin/restic'

# Actually do the backup. Each directory in the list below will be backed up
# separately.
dirs=(
	"/some/path/to/backup/"
	"/some/other/path/"
)
for i in "${dirs[@]}" 
do
	echo $i
	(cd $i && restic --verbose backup .)
done

# Prune our snapshots. You can tweak the numbers here. Run with `--dry-run` to
# see what effect any changes would have before actually committing to them. 
restic forget --keep-daily 7 --keep-weekly 10 --keep-monthly 24 --keep-yearly 10

The only other notable thing here is that the script changes into the directory to be backed up. If you give Restic an absolute path, it will create a new snapshot if the metadata of any folder in the path changes, which is not what I want.

If it wasn’t for the S3 URL issues, the whole thing would’ve probably taken me about half an hour. That’s including setting up the object storage, installing Restic, and so on. It’s painfully easy. Why didn’t I do this three years ago?!

Addendum: a step-by-step guide

If you want to do this yourself, here’s a quick step-by-step guide:

Log in to the Hetzner Cloud Console, and create a project.
On the “Object Storage” tab, create a new bucket. Note the name and the endpoint.
On the “Security” tab, go to “S3 Credentials” and generate new credentials. Note down the access key and the secret key.
Copy the first shell script above, fill in the password (you can pick!), endpoint, bucket name, access key and secret key.
Run the script with the init argument (e.g. ~/.bin/restic init). This will create a new repository, and only needs to be done once even if you backup multiple machines.
Copy the second shell script above, making sure the restic alias points at the script you saved in step 4. Change the list of directories to whatever you want to backup.
Schedule the script to be run automatically, using crontab or systemd timers or however you prefer.

To check everything is working, you can use the snapshots subcommand to see a list of saved snapshots. You might also want to try to restore a snapshot using the restore subcommand.

I’m not trying to be a language snob, but given the choice between two otherwise equal projects one in Go and one in Python, I’ll take the Go one. I know I’m not going to have weird library issues down the line, and I’m much more comfortable rummaging around the source. ↩︎
With a “Don’t use this in production!” notice on the shiny new version. ↩︎
Aside: I really hate URLs like that. What is with the trend for completely generic domains divorced from the service they’re a part of? ↩︎

Apple, Google and aligned incentives

2020-10-17T00:00:00Z

The HTC Dream, the first phone released running Android.

For the past decade I’ve exclusively used Android phones. I got the HTC Dream (aka the T-Mobile G1) shortly after it came out, and dutifully upgraded every 1-2 years. In that timespan I used Android as the basis for my Master’s Thesis, took a job on the Android team at Google, and eventually became a contractor specialising in Android app development. So when I switched to using an iPhone earlier this year a few people were surprised¹.

The good old days

When Android was announced in 2007 – alongside the formation of the Open Handset Alliance – it was positioned as a bastion of openness: it would be built on open standards and the operating system would be open source. At the time iPhones were strongly coupled to iTunes and Apple was exercising strict control over what app developers could do.

When the HTC Dream was released it lived up to expectations. You could write apps for it without shelling out for a Mac! You could get root access, and it was running Linux under the hood! A whole ecosystem of custom firmwares and bootloaders started to appear, thanks to the open source nature of the OS. It shipped with some Google apps, but they were just normal apps that served as examples of what could be done.

After the Dream came a line of Nexus devices. These were Android’s flagship devices, designed to show off what a good Android phone should look like. Both the hardware and software releases tended to feature interesting, useful upgrades. Some of the original open source apps were replaced with closed source, Google proprietary ones, but that was OK - the open source versions lived on in the open source project as examples of what you could do. The devices allowed flashing custom firmware, and the OS source was always released… eventually.

The downfall

The Pixel and Pixel XL

Over time, Android has got less and less free. The Nexus line of phones gave way to the Pixel range, which got rid of the clearly demarcated border where Android ended and Google began. The Android 11 highlights list is dotted with sections that say “On Pixel devices…”, and every single one is a software feature that could be implemented on any device, but Google have decided to keep it proprietary instead of releasing it as part of the open source platform.

At the same time more and more functionality has been added to Google Play Services. This was originally a shared location for Google specific services - in 2012 it merely handled some Google+ functionality and dealing with OAuth for Google accounts. These days it contains a huge swathe of Google services, as well as platform functionality such as push notifications, barcode scanning, geolocation, and so on. Google Play Services isn’t part of the Android Open Source Project, and is only available under license from Google. You can’t really have a device without these functions, so as a manufacturer you have the choice between agreeing to whatever terms Google requires² or spending an awful lot of development time creating an alternative.

While these are fairly abstract arguments about what a free platform should look like, during the same period there has been a marked restriction in how you can actually use Android devices – both as an end-user and as a developer. iOS has always had very stark restrictions on what apps can do in the background; Android started its life allowing pretty much anything, like any good general purpose computing device. This inevitably lead to lots of apps doing lots of things that ranged from stupid to mildly suboptimal, creating a tragedy of the commons amongst apps. The victim was the phone’s battery life, and Google’s solution was a progressive series of restrictions on what apps can do and when. This includes limits on when push notifications are delivered to a device, how frequently apps can wake up, and so forth. Some of these can be bypassed by the end-user, but not all, and the process is fairly cumbersome.

Aligned incentives

From my point of view, Android and iOS are now pretty much in a similar place. They’re not general purpose computers, but app consoles: much like games consoles they consist of hardware and software that the user lacks control over but accepts in order to access the library of apps/games.

Given there’s no clear winner between them in terms of hardware and software, my decision came down to a more holistic question: how aligned are their incentives to my own? Apple is a product company: they make their money by producing shiny things that people want to purchase; Google is an advertising company: they make their money by using my personal information to show me targeted adverts.

For Google, Android was originally a strategic move to ensure that Apple couldn’t dominate the mobile web, and by extension the revenue from ads. As a company, there’s nothing to really push them forward in any particular direction other than one that facilitates advertising. Of course, Google employs tonnes of good people who want to do good things which counterbalances this, but I’d still prefer to deal with a company that has an intrinsic motivation to do things I want them to do, rather than one forced to by regulation, custom, or the good intent of their employees.

So the decision became obvious: if I have no strong opinions about the software and hardware, Apple is the clear winner because their incentives are a lot better aligned to mine.

Image credits

Photo	Creator	Licence	Source
HTC Dream	Akela NDE	CC BY-SA 3.0	Wikimedia
Pixel and Pixel XL	Maurizio Pesce from Milan, Italia	CC BY 2.0	Wikimedia

Or, at least, politely feigned surprise. ↩︎
And Google’s terms were particularly onerous: as well as requiring Chrome and Google Search to be preinstalled, they prevented manufacturers from selling any devices powered by alternative versions of Android - e.g., Samsung wouldn’t be allowed to sell a refrigerator that ran Amazon’s FireOS. The European Union handed Google a $5,000,000,000 fine for this anti-competitive behaviour, which Google are still in the process of contesting. ↩︎