Monthly Archives: January 2021

The Hidden Drawbacks of P2P (And a Defense of Signal)

Not long ago, I posted a roundup of secure messengers with off-the-grid capabilities. Some conversation followed, which led me to consider some of the problems with P2P protocols.

P2P and Privacy

Brave adopting IPFS has driven a lot of buzz lately. IPFS is essentially a decentralized, distributed web. This concept has a lot of promise. But take a look at the IPFS privacy document. Some things to highlight:

  • “Nodes announce a variety of information essential to the DHT’s function — including their unique node identifiers (PeerIDs) and the CIDs of data that they’re providing — and because of this, information about which nodes are retrieving and/or reproviding which CIDs is publicly available.”
  • “those DHT queries happen in public. Because of this, it’s possible that third parties could be monitoring this traffic to determine what CIDs are being requested, when, and by whom.”
  • “nodes’ unique identifiers are themselves public…your PeerID is still a long-lived, unique identifier for your node. Keep in mind that it’s possible to do a DHT lookup on your PeerID and, particularly if your node is regularly running from the same location (like your home), find your IP address…Additionally, longer-term monitoring of the public IPFS network could yield information about what CIDs your node is requesting and/or reproviding and when.”

So in this case, you have traded giving information about what you request to specific sites to giving it to potentially hundreds of untrusted peers, some of which may be logging this for nefarious purposes. Worse, you have a durable PeerID that can be used for tracking and tied to your IP address — a data collector’s dream. This PeerID, combined with DHT requests and the CIDs (Content ID) of the things you host (implying you viewed them in the past), can be used to establish a picture of what you are requesting now and requested recently.

Similar can be said from everything like Scuttlebutt to GNU Jami; any service that operates on a P2P basis will likely reveal your IP, and tie your identity to it (and your IP address history). In some cases, as with Jami, this would be limited to friends you add; in others, as with Scuttlebutt and IPFS, it could be revealed to anyone.

The advantages of P2P are undeniable and profound, but few are effectively addressing the privacy implications. The one I know of that is, Briar, routes all traffic over Tor; every node is reached by a Tor onion service.

Federation: somewhat better

In a federated model, every client connects to a server, and there are many servers participating in a federation with each other. Matrix and Mastodon are examples of a federated model. In this scenario, only one server — your own homeserver — can track you by IP. End-to-end encryption is certainly possible in a federated model, and Matrix supports it. This does give a third party (the specific server you use) knowledge of your IP, but that knowledge can be significantly limited.

A downside of this approach is that if your particular homeserver is down, you are unable to communicate. Truly decentralized P2P solutions don’t have that problem — thought they do have a related one, which is that clients communicating with each other must both be online simultaneously in order for messages to be transmitted, and this can be a real challenge for mobile devices.

Centralization and Signal

Signal is centralized; it has one central server farm, and if it is down, you can’t communicate or choose any other server, either. We saw it go down recently after Elon Musk mentioned it.

Still, I recommend Signal for the general public. Here’s why.

Signal brings encryption and privacy to meet people where they’re at, not the other way around. People don’t have to choose a server, it can automatically recognize contacts that use Signal, it has emojis, attachments, secure voice and video calling, and (aside from the Musk incident), it all just works. It feels like, and is, a polished, modern experience with the bells and whistles people are used to.

I’m a huge fan of Matrix (aka Element) and even run my own instance. It has huge promise. But it is Not. There. Yet. Why do I saw this about Matrix?

  • Synapse, the only currently viable Matrix server, is not ready. My Matrix instance hosts ONE person, me. Synapse uses many GB of RAM and 10+GB of disk space. Despite extensive tuning, nothing helped much. It’s caused OOMs more than once. It can’t be hosted on a Raspberry Pi or even one of the cheaper VPSs.
  • Now then, how about choosing a Matrix instance? Well, you could just tell a person to use matrix.org. But then it spent a good portion of last year unable to federate with other popular nodes due to Synapse limitations. Or you could pick a random node, but will it be up when someone needs to say “my car broke down?” Some are run from a dorm computer, some by a team in a datacenter, some by one person with EC2, and you can’t really know. Will your homeserver be stable and long-lived? Hard to say.
  • Voice and video calling are not there yet in Matrix. Matrix has two incompatible video calling methods (Jitsi and built-in), neither work consistently well, both are hard to manage, and both have NAT challenges.
  • Matrix is so hard to set up on a server that there is matrix-docker-ansible-deploy. This makes it much better, but it is STILL terribly hard to deploy, and very simple things like “how do I delete a user” or “let me shrink down this 30GB database” are barely there yet, if at all.
  • Encryption isn’t mandatory in Matrix. E2EE has been getting dramatically better in the last few releases, but it is still optional, especially for what people would call “group chats” (rooms). Signal is ALWAYS encrypted. Always. (Unless, I guess, you set it as your SMS provider on Android). You’ve got to take the responsibility off the user to verify encryption status, and instead make it the one and only way to use the ecosystem.

Again, I love MAtrix. I use it every day to interact with Matrix, IRC, Slack, and Discord channels. It has a ton of promise. But would I count on it to carry a “my car’s broken down and I’m stranded” message? No.

How about some of the other options out there? I mentioned Briar above. It’s fantastic and its offline options are novel and promising. But in common usage, it can’t deliver a message unless both devices are online simultaneously, and doesn’t run on iOS (though both are being worked on). It also can’t send photos or do voice or video calling.

Some of these same limitations apply to most of the other Signal alternatives also. either that, or they are encryption-optional, or terribly hard to set up and use. I recently mentioned Status, which shows a ton of promise, but has no voice or video calling capabilities. Scuttlebutt is a fantastic protocol with extremely difficult onboarding (lengthy process, error-prone finding a pub, multi-GB initial download, etc.) And many of these leak IP addresses as discussed above.

So Signal gives people:

  • Dead-simple setup
  • Store-and-forward delivery (devices need not be online simultaneously)
  • Encrypted everything, including voice and video calls, and the ability to send photos and video encrypted

If you are going to tell someone, “it’s so EASY to get your texts away from Facebook and AT&T”, then Signal is the thing you’ve got to point them to. It may not be in two years, but for now, it is. Do not let the perfect be the enemy of the good. It advances the status quo without harming usability, which nothing else does yet.

I am aware of all of the very legitimate criticisms of Signal. They are real and they are why I am excited that there are so many alternatives with promise, some of which I use actively. Let us technical people use, debug, contribute to, and evangelize the alternatives.

And while we’re doing that, tell Grandma to contact us on Signal.

Roundup of Secure Messengers with Off-The-Grid Capabilities (Distributed/Mesh Messengers)

Amid all the conversation about Signal, and the debate over decentralization, one thing has often not been raised: all of these things require an Internet connection.

“Of course,” you might say. “Internet is everywhere these days.” Well, not so much, and it turns out there are some very good reasons that people might want messengers that work offline. Here are some examples:

  • Internet-using messengers leak certain metadata (eg, that a person is using it, or perhaps a sophisticated adversary could use timing analysis to determine that two people are talking using it)
  • Cell signal outages due to natural disaster, large influx of people (protests, unusual sporting events, festivals, etc), or other factors
  • Locations where cell signals are not available (rural areas, camping locations, wilderness areas, etc.)
  • Devices that don’t have cell data capability (many tablets, phones that have had service expire, etc.)

How do they work?

These all use some form of local radio signal. Some, such as Briar, may use short-range Bluetooth and Wifi, while others use radios such as LoRa that can reach several miles with low power. I’ve written quite a bit about LoRa before, and its unique low-speed but extreme-distance radio capabilities even on low power.

One common thread through these is that most of them are Android-only, though many are compatible with F-Droid and privacy-enhanced Android distributions.

Every item on this list uses full end-to-end encryption (E2EE).

Let’s dive on in.

Briar

Of all the options mentioned here, Briar is the one that bridges the traditional Internet-based approach with alternative options the best. It offers three ways for distributing data:

  • Over the Internet, via Tor onion services
  • Via Bluetooth to nearby devices
  • Via Wifi, to other devices connected to the same access point, even if Internet isn’t wokring on that AP

As far as I can tell, there is no centralized server in Briar at all. Your “account”, such as it is, lives entirely within your device; if you wipe your device, you will have to make a new account and re-establish contacts. The use of Tor is also neat to see; it ensures that an adversary can’t tell, just from that, that you’re using Briar at all, though of course timing analysis may still be possible (and Bluetooth and Wifi uses may reval some of who is communicating).

Briar features several types of messages (detailed in the manual), which really are just different spins on communication, which they liken to metaphors people are familiar with:

  • Basic 1-to-1 private messaging
  • “Private groups”, in which one particular person invites people to the chat group, and can dissolve it at any time
  • “Forums”, similar to private groups, but any existing member can invite more people to them, and they continue to exist until the last member leaves (founder isn’t special)
  • “Blogs”, messages that are automatically shared with all your contacts

By default, Briar raises an audible notification for incoming messages of all types. This is configurable for each type.

“Blogs” have a way to reblog (even a built-in RSS reader to facilitate that), but framed a different way, they are broadcast messages. They could, for instance, be useful for a “send help” message to everyone (assuming that people haven’t all shut off notifications of blogs due to others using them different ways).

Briar’s how it works page has an illustration specifically of how blogs are distributed. I’m unclear on some of the details, and to what extent this applies to other kinds of messages, but one thing that you can notice from this is that a person A could write a broadcast message without Internet access, person B could receive it via Bluetooth or whatever, and then when person B gets Internet access again, the post could be distributed more widely. However, it doesn’t appear that Briar is really a full mesh, since only known contacts in the distribution path for the message would repeat it.

There are some downsides to Briar. One is that, since an account is fully localized to a device, one must have a separate account for each device. That can lead to contacts having to pick a specific device to send a message to. There is an online indicator, which may help, but it’s definitely not the kind of seamless experience you get from Internet-only messengers. Also, it doesn’t support migrating to a new phone, live voice/video calls, or attachments, but attachments are in the works.

All in all, a solid communicator, and is the only one on this list that works 100% with the hardware everyone already has. While Bluetooth and Wifi have far more limited range than the other entries, there is undeniably convenience in not needing any additional hardware, and it may be particularly helpful when extra bags/pockets aren’t available. Also, Briar is fully Open Source.

Meshtastic

Meshtastic is a radio-first LoRa mesh project. What do I mean by radio-first? Well, basically cell phones are how you interact with Meshtastic, but they are optional. The hardware costs about $30 and the batteries last about 8 days. Range between nodes is a few miles in typical conditions (up to 11km / 7mi in ideal conditions), but nodes act as repeaters, so it is quite conceivable to just drop a node “in the middle” if you and contacts will be far apart. The project estimates that around 2000 nodes are in operation, and the network is stronger the more nodes are around.

The getting started site describes how to build one.

Most Meshtastic device builds have a screen and some buttons. They can be used independently from the Android app to display received messages, distance and bearing to other devices (assuming both have a GPS enabled), etc. This video is an introduction showing it off, this one goes over the hardware buttons. So even if your phone is dead, you can at least know where your friends are. Incidentally, the phone links up to the radio board using Bluetooth, and can provide a location source if you didn’t include one in your build. There are ideas about solar power for Meshtastic devices, too.

Meshtastic doesn’t, as far as I know, have an option for routing communication over the Internet, but the devices appear to be very thoughtfully-engineered and easy enough to put together. This one is definitely on my list to try.

Ripple-based devices

This is based on the LoRa Mesh Radio Instructables project, and is similar in concept to Meshtastic. It uses similar hardware, a similar app, but also has an option with a QWERTY hardware keyboard available, for those that want completely phone-free operation while still being able to send messages.

There are a number of related projects posted at Instructables: a GPS tracker, some sensors, etc. These are variations on the same basic concept.

These use the Ripple firmware, which is not open source, so I haven’t pursued it further.

GoTenna

For people that want less of a DIY model, and don’t mind proprietary solutions, there are two I’ll mention. The first is GoTenna Mesh, which is LoRa-based and sells units for $90 each. However, there are significant community concerns about the longevity of the project, as GoTenna has re-focused on government and corporate work. The Android app hasn’t been updated in 6 monnths despite a number of reviews citing issues, and the iOS app is also crusty.

Beartooth

Even more expensive at $125 each is the Beartooth. Also a proprietary option, I haven’t looked into it more, but they are specifically targetting backwoods types of markets.

Do not use: Bridgefy

Bridgefy was briefly prominent since it was used during the Hong Kong protests. However, numerous vulnerabilities have been demonstrated, and the developers have said they are re-working the app to address them. I wouldn’t recommend it for now.

Alternatives: GMRS handhelds

In the USA, GMRS voice handhelds are widely available. Although a license is required, it is simple (no exam) and cheap ($35) and extends to a whole family. GMRS radios also interoperate with FRS radios, which require no license and share some frequencies, but are limited to lower power (though are often sufficient).

Handheld GMRS radios that use up to 5W of power are readily available. A voice signal is a lot harder to carry for a long distance than a very low-bandwidth digital one, so even with much more power you will probably not get the same kind of range you will with something like Meshtastic, and they don’t come with any kind of security or encryption at all. However, for basic communication, they are often a useful tool.

Remote Directory Tree Comparison, Optionally Asynchronous and Airgapped

Note: this is another article in my series on asynchronous communication in Linux with UUCP and NNCP.

In the previous installment on store-and-forward backups, I mentioned how easy it is to do with ZFS, and some of the tools that can be used to do it without ZFS. A lot of those tools are a bit less robust, so we need some sort of store-and-forward mechanism to verify backups. To be sure, verifying backups is good with ANY scheme, and this could be used with ZFS backups also.

So let’s say you have a shiny new backup scheme in place, and you’d like to verify that it’s working correctly. To do that, you need to compare the source directory tree on machine A with the backed-up directory tree on machine B.

Assuming a conventional setup, here are some ways you might consider to do that:

  • Just copy everything from machine A to machine B and compare locally
  • Or copy everything from machine A to a USB drive, plug that into machine B, and compare locally
  • Use rsync in dry-run mode and see if it complains about anything

The first two options are not particularly practical for large datasets, though I note that the second is compatible with airgapping. Using rsync requires both systems to be online at the same time to perform the comparison.

What would be really nice here is a tool that would write out lots of information about the files on a system: their names, sizes, last modified dates, maybe even sha256sum and other data. This file would be far smaller than the directory tree itself, would compress nicely, and could be easily shipped to an airgapped system via NNCP, UUCP, a USB drive, or something similar.

Tool choices

It turns out there are already quite a few tools in Debian (and other Free operating systems) to do this, and half of them are named mtree (though, of course, not all mtrees are compatible with each other.) We’ll look at some of the options here.

I’ve made a simple test directory for illustration purposes with these commands:

mkdir test
cd test
echo hi > hi
ln -s hi there
ln hi foo
touch empty
mkdir emptydir
mkdir somethingdir
cd somethingdir
ln -s ../there

I then also used touch to set all files to a consistent timestamp for illustration purposes.

Tool option: getfacl (Debian package: acl)

This comes with the acl package, but can be used with other than ACL purposes. Unfortunately, it doesn’t come with a tool to directly compare its output with a filesystem (setfacl, for instance, can apply the permissions listed but won’t compare.) It ignores symlinks and doesn’t show sizes or dates, so is ineffective for our purposes.

Example output:

$ getfacl --numeric -R test
...
# file: test/hi
# owner: 1000
# group: 1000
user::rw-
group::r--
other::r--
...

Tool option: fmtree, the FreeBSD mtree (Debian package: freebsd-buildutils)

fmtree can prepare a “specification” based on a directory tree, and compare a directory tree to that specification. The comparison also is aware of files that exist in a directory tree but not in the specification. The specification format is a bit on the odd side, but works well enough with fmtree. Here’s a sample output with defaults:

$ fmtree -c -p test
...
# .
/set type=file uid=1000 gid=1000 mode=0644 nlink=1
.               type=dir mode=0755 nlink=4 time=1610421833.000000000
    empty       size=0 time=1610421833.000000000
    foo         nlink=2 size=3 time=1610421833.000000000
    hi          nlink=2 size=3 time=1610421833.000000000
    there       type=link mode=0777 time=1610421833.000000000 link=hi

... skipping ...

# ./somethingdir
/set type=file uid=1000 gid=1000 mode=0777 nlink=1
somethingdir    type=dir mode=0755 nlink=2 time=1610421833.000000000
    there       type=link time=1610421833.000000000 link=../there
# ./somethingdir
..

..

You might be wondering here what it does about special characters, and the answer is that it has octal escapes, so it is 8-bit clean.

To compare, you can save the output of fmtree to a file, then run like this:

cd test
fmtree < ../test.fmtree

If there is no output, then the trees are identical. Change something and you get a line of of output explaining each difference. You can also use fmtree -U to change things like modification dates to match the specification.

fmtree also supports quite a few optional keywords you can add with -K. They include things like file flags, user/group names, various tipes of hashes, and so forth. I'll note that none of the options can let you determine which files are hardlinked together.

Here's an excerpt with -K sha256digest added:

    empty       size=0 time=1610421833.000000000 \
                sha256digest=e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
    foo         nlink=2 size=3 time=1610421833.000000000 \
                sha256digest=98ea6e4f216f2fb4b69fff9b3a44842c38686ca685f3f55dc48c5d3fb1107be4

If you include a sha256digest in the spec, then when you verify it with fmtree, the verification will also include the sha256digest. Obviously fmtree -U can't correct a mismatch there, but of course it will detect and report it.

Tool option: mtree, the NetBSD mtree (Debian package: mtree-netbsd)

mtree produces (by default) output very similar to fmtree. With minor differences (such as the name of the sha256digest in the output), the discussion above about fmtree also applies to mtree.

There are some differences, and the most notable is that mtree adds a -C option which reads a spec and converts it to a "format that's easier to parse with various tools." Here's an example:

$ mtree -c -K sha256digest -p test | mtree -C
. type=dir uid=1000 gid=1000 mode=0755 nlink=4 time=1610421833.0 flags=none 
./empty type=file uid=1000 gid=1000 mode=0644 nlink=1 size=0 time=1610421833.0 flags=none 
./foo type=file uid=1000 gid=1000 mode=0644 nlink=2 size=3 time=1610421833.0 flags=none 
./hi type=file uid=1000 gid=1000 mode=0644 nlink=2 size=3 time=1610421833.0 flags=none 
./there type=link uid=1000 gid=1000 mode=0777 nlink=1 link=hi time=1610421833.0 flags=none 
./emptydir type=dir uid=1000 gid=1000 mode=0755 nlink=2 time=1610421833.0 flags=none 
./somethingdir type=dir uid=1000 gid=1000 mode=0755 nlink=2 time=1610421833.0 flags=none 
./somethingdir/there type=link uid=1000 gid=1000 mode=0777 nlink=1 link=../there time=1610421833.0 flags=none 

Most definitely an improvement in both space and convenience, while still retaining the relevant information. Note that if you want the sha256digest in the formatted output, you need to pass the -K to both mtree invocations. I could have done that here, but it is easier to read without it.

mtree can verify a specification in either format. Given what I'm about to show you about bsdtar, this should illustrate why I bothered to package mtree-netbsd for Debian.

Unlike fmtree, the mtree -U command will not adjust modification times based on the spec, but it will report on differences.

Tool option: bsdtar (Debian package: libarchive-tools)

bsdtar is a fascinating program that can work with many formats other than just tar files. Among the formats it supports is is the NetBSD mtree "pleasant" format (mtree -C compatible).

bsdtar can also convert between the formats it supports. So, put this together: bsdtar can convert a tar file to an mtree specification without extracting the tar file. bsdtar can also use an mtree specification to override the permissions on files going into tar -c, so it is a way to prepare a tar file with things owned by root without resorting to tools like fakeroot.

Let's look at how this can work:

$ cd test
$ bsdtar --numeric -cf - --format=mtree .

. time=1610472086.318593729 mode=755 gid=1000 uid=1000 type=dir
./empty time=1610421833.0 mode=644 gid=1000 uid=1000 type=file size=0
./foo nlink=2 time=1610421833.0 mode=644 gid=1000 uid=1000 type=file size=3
./hi nlink=2 time=1610421833.0 mode=644 gid=1000 uid=1000 type=file size=3
./ormat\075mtree time=1610472086.318593729 mode=644 gid=1000 uid=1000 type=file size=5632
./there time=1610421833.0 mode=777 gid=1000 uid=1000 type=link link=hi
./emptydir time=1610421833.0 mode=755 gid=1000 uid=1000 type=dir
./somethingdir time=1610421833.0 mode=755 gid=1000 uid=1000 type=dir
./somethingdir/there time=1610421833.0 mode=777 gid=1000 uid=1000 type=link link=../there

You can use mtree -U to verify that as before. With the --options mtree: set, you can also add hashes and similar to the bsdtar output. Since bsdtar can use input from tar, pax, cpio, zip, iso9660, 7z, etc., this capability can be used to create verification of the files inside quite a few different formats. You can convert with bsdtar -cf output.mtree --format=mtree @input.tar. There are some foibles with directly using these converted files with mtree -U, but usually minor changes will get it there.

Side mention: stat(1) (Debian package: coreutils)

This tool isn't included because it won't operate recursively, but is a tool in the similar toolbox.

Putting It Together

I will still be developing a complete non-ZFS backup system for NNCP (or UUCP) in a future post. But in the meantime, here are some ideas you can reflect on:

  • Let's say your backup scheme involves sending a full backup every night. On the source system, you could pipe the generated tar file through something like tee >(bsdtar -cf bcakup.mtree @-) to generate an mtree file in-band while generating the tar file. This mtree file could be shipped over for verification.
  • Perhaps your backup scheme involves sending incremental backup data via rdup or even ZFS, but you would like to periodically verify that everything is good -- that an incremental didn't miss something. Something like mtree -K sha256 -c -x -p / | mtree -C -K sha256 would let you accomplish that.

I will further develop at least one of these ideas in a future post.

Bonus: cross-tool comparisons

In my mtree-netbsd packaging, I added tests like this to compare between tools:

fmtree -c -K $(MTREE_KEYWORDS) | mtree
mtree -c -K $(MTREE_KEYWORDS) | sed -e 's/\(md5\|sha1\|sha256\|sha384\|sha512\)=/\1digest=/' -e 's/rmd160=/ripemd160digest=/' | fmtree
bsdtar -cf - --options 'mtree:uname,gname,md5,sha1,sha256,sha384,sha512,device,flags,gid,link,mode,nlink,size,time,uid,type,uname' --format mtree . | mtree

The Good, Bad, and Scary of the Banning of Donald Trump, and How Decentralization Makes It All Better

It is undeniable that banning Donald Trump from Facebook, Twitter, and similar sites is a benefit for the moment. It may well save lives, perhaps lots of lives. But it raises quite a few troubling issues.

First, as EFF points out, these platforms have privileged speakers with power, especially politicians, over regular users. For years now, it has been obvious to everyone that Donald Trump has been violating policies on both platforms, and yet they did little or nothing about it. The result we saw last week was entirely forseeable — and indeed, WAS forseen, including by elements in those companies themselves. (ACLU also raises some good points)

Contrast that with how others get treated. Facebook, two days after the coup attempt, banned Benjamin Wittes, apparently because he mentioned an Atlantic article opposed to nutcase conspiracy theories. The EFF has also documented many more egregious examples: taking down documentation of war crimes, childbirth images, black activists showing the racist messages they received, women discussing online harassment, etc. The list goes on; YouTube, for instance, has often been promoting far-right violent videos while removing peaceful LGBTQ ones.

In short, have we simply achieved legal censorship by outsourcing it to dominant corporations?

It is worth pausing at this point to recognize two important princples:

First, that we do not see it as right to compel speech.

Secondly, that there exist communications channels and other services that nobody is calling on to suspend Donald Trump.

Let’s dive into those a little bit.

There have been no prominent calls for AT&T, Verizon, Gmail, or whomever provides Trump and his campaign with cell phones or email to suspend their service to him. Moreover, the gas stations that fuel his vehicles and the airports that service his plane continue to provide those services, and nobody has seriously questioned that, either. Even his Apple phone that he uses to post to Twitter remains, as far as I know, fully active.

Secondly, imagine you were starting up a small web forum focused on raising tomato plants. It is, and should be, well within your rights to keep tomato-haters out, as well as people that have no interest in tomatoes but would rather talk about rutabagas, politics, or Mars. If you are going to host a forum about tomatoes, you have the right to keep it a forum about tomatoes; you cannot be forced to distribute someone else’s speech. Likewise in traditional media, a newspaper cannot be forced to print every letter to the editor in full.

In law, there is a notion of a common carrier, that provides services to the general public without discrimination. Phone companies and ISPs fall under this.

Facebook, Twitter, and tomato sites don’t. But consider what happens if Facebook bans you. You might be using Facebook-owned Whatsapp to communicate with family and friends, and suddenly find yourself unable to ask someone to pick you up. Or your treasured family photos might be in Facebook-owned Instagram, lost forever. It’s not just Facebook; similar things happen with Google, locking people out of their phones and laptops, their emails, even their photos.

Is it right that Facebook and Google aren’t regulated as common carriers? Perhaps, or perhaps we need some line of demarcation between their speech-to-the-public services (Facebook timeline posts, YouTube) and private communication (Whatsapp, Gmail). It’s a thorny issue; should government be regulating speech instead? That’s also fraught. So is corporate control.

Decentralization Helps Dramatically

With email, you get to pick your email provider (yes, there are two or three big ones, but still plenty of others). Each email provider will have its own set of things it considers acceptable, and its own set of other servers and accounts it’s willing to exchange mail with. (It is extremely common for mail providers to choose not to accept mail from various other mail servers based on ISP, IP address, reputation, and so forth.)

What if we could do something like that for Twitter and Facebook?

Let you join whatever instance you like. Maybe one instance is all about art and they don’t talk about politics. Or another is all about Free Software and they don’t have advertising. And then there are plenty of open instances that accept anything that’s respectful. And, like email, people of one server can interact with those using another just as easily as if they were using the same one.

Well, this isn’t hypothetical; it already exists in the Fediverse. The most common option is Mastodon, and it so happens that a month ago I wrote about its benefits for other reasons, and included some links on getting started.

There is no reason that we must all let our online speech be controlled by companies with a profit motive to keep hate speech on their platforms. There is no reason that we must all have a single set of rules, or accept strong corporate or government control, either. The quality of conversation on Mastodon is far higher than either Twitter or Facebook; decentralization works and it’s here today.

This Is How Tyrants Go: Alone

I remember reading an essay a month or so ago — sadly I forget where — talking about how things end for tyrants. If I were to sum it up, it would be with the word “alone.” Their power fading, they find that they had few true friends or believers; just others that were greedy for power or riches and, finding those no longer to be had, depart the sinking ship. The article looked back at examples like Nixon and examples from the 20th century in Europe and around the world.

Today we saw images of a failed coup attempt.

But we also saw hope.

Already senior staff in the White House are resigning. Ones that had been ardent supporters. In the end, just 6 senators supported the objection to the legitimate electors. Six. Lindsay Graham, Mike Pence, and Mitch McConnel all deserted Trump.

CNN reports that there are serious conversations about invoking the 25th amendment and removing him from office, because even Republicans are to the point of believing that America should not have two more weeks of this man.

Whether those efforts are successful or not, I don’t know. What I do know is that these actions have awakened many people, in a way that nothing else could for four years, to the dangers of Trump and, in the end, have bolstered the cause of democracy.

Hard work will remain but today, Donald Trump is in the White House alone, abandoned by allies and blocked by Twitter. And we know that within two weeks, he won’t be there at all.

We will get through this.

More Topics on Store-And-Forward (Possibly Airgapped) ZFS and Non-ZFS Backups with NNCP

Note: this is another article in my series on asynchronous communication in Linux with UUCP and NNCP.

In my previous post, I introduced a way to use ZFS backups over NNCP. In this post, I’ll expand on that and also explore non-ZFS backups.

Use of nncp-file instead of nncp-exec

The previous example used nncp-exec (like UUCP’s uux), which lets you pipe stdin in, then queues up a request to run a given command with that input on a remote. I discussed that NNCP doesn’t guarantee order of execution, but that for the ZFS use case, that was fine since zfs receive would just fail (causing NNCP to try again later).

At present, nncp-exec stores the data piped to it in RAM before generating the outbound packet (the author plans to fix this shortly) [Update: This is now fixed; use -use-tmp with nncp-exec!). That made it unusable for some of my backups, so I set it up another way: with nncp-file, the tool to transfer files to a remote machine. A cron job then picks them up and processes them.

On the machine being backed up, we have to find a way to encode the dataset to be received. I chose to do that as part of the filename, so the updated simplesnap-queue could look like this:

#!/bin/bash

set -e
set -o pipefail

DEST="`echo $1 | sed 's,^tank/simplesnap/,,'`"
FILE="bakfsfmt2-`date "+%s.%N".$$`_`echo "$DEST" | sed 's,/,@,g'`"

echo "Processing $DEST to $FILE" >&2
# stdin piped to this
zstd -8 - \
  | gpg --compress-algo none --cipher-algo AES256 -e -r 012345...  \
  | su nncp -c "/usr/local/nncp/bin/nncp-file -nice B -noprogress - 'backupsvr:$FILE'" >&2

echo "Queued $DEST to $FILE" >&2

I’ve added compression and encryption here as well; more on that below.

On the backup server, we would define a different incoming directory for each node in nncp.hjson. For instance:

host1: {
...
   incoming: "/var/local/nncp-bakcups-incoming/host1"
}

host2: {
...
   incoming: "/var/local/nncp-backups-incoming/host2"
}

I’ll present the scanning script in a bit.

Offsite Backup Rotation

Most of the time, you don’t want just a single drive to store the backups. You’d like to have a set. At minimum, one wouldn’t be plugged in so lightning wouldn’t ruin all your backups. But maybe you’d store a second drive at some other location you have access to (friend’s house, bank box, etc.)

There are several ways you could solve this:

  • If the remote machine is at a location with network access and you trust its physical security (remember that although it will store data encrypted at rest and will transport it encrypted, it will — in most cases — handle un-encrypted data during processing), you could of course send NNCP packets to it over the network at the same time you send them to your local backup system.
  • Alternatively, if the remote location doesn’t have network access or you want to keep it airgapped, you could transport the NNCP packets by USB drive to the remote end.
  • Or, if you don’t want to have any kind of processing capability remotely — probably a wise move — you could rotate the hard drives themselves, keeping one plugged in locally and unplugging the other to take it offsite.

The third option can be helped with NNCP, too. One way is to create separate NNCP installations for each of the drives that you store data on. Then, whenever one is plugged in, the appropriate NNCP config will be loaded and appropriate packets received and processed. The neighbor machine — the spooler — would just store up packets for the offsite drive until it comes back onsite (or, perhaps, your airgapped USB transport would do this). Then when it’s back onsite, all the queued up ZFS sends get replayed and the backups replicated.

Now, how might you handle this with NNCP?

The simple way would be to have each system generating backups send them to two destinations. For instance:

zstd -8 - | gpg --compress-algo none --cipher-algo AES256 -e -r 07D5794CD900FAF1D30B03AC3D13151E5039C9D5 \
  | tee >(su nncp -c "/usr/local/nncp/bin/nncp-file -nice B+5 -noprogress - 'backupdisk1:$FILE'") \
        >(su nncp -c "/usr/local/nncp/bin/nncp-file -nice B+5 -noprogress - 'backupdisk2:$FILE'") \
   > /dev/null

You could probably also more safely use pee(1) (from moreutils) to do this.

This has an unfortunate result of doubling the network traffic from every machine being backed up. So an alternative option would be to queue the packets to the spooling machine, and run a distribution script from it; something like this, in part:

INCOMINGDIR="/var/local/nncp-bakfs-incoming"
LOCKFILE="$INCOMINGDIR/.lock"
printf -v EVAL_SAFE_LOCKFILE '%q' "$LOCKFILE"
if dotlockfile -r 0 -l -p "${LOCKFILE}"; then
  logit "Lock obtained at ${LOCKFILE} with dotlockfile"
  trap 'ECODE=$?; dotlockfile -u '"${EVAL_SAFE_LOCKFILE}"'; exit $ECODE' EXIT INT TERM
else
  logit "Could not obtain lock at $LOCKFILE; $0 likely already running."
  exit 0
fi


logit "Scanning queue directory..."
cd "$INCOMINGDIR"
for HOST in *; do
   cd "$INCOMINGDIR/$HOST"
   for FILE in bakfsfmt2-*; do
           if [ -f "$FILE" ]; then
                   for BAKFS in backupdisk1 backupdisk2; do
                           runcommand nncp-file -nice B+5 -noprogress "$FILE" "$BAKFS:$HOST/$FILE"
                   done
                   runcommand rm "$FILE"
           else
                   logit "$HOST: Skipping $FILE since it doesn't exist"
           fi
   done

done
logit "Scan complete."

Security Considerations

You’ll notice that in my example above, the encryption happens as the root user, but nncp is called under su. This means that even if there is a vulnerability in NNCP, the data would still be protected by GPG. I’ll also note here that many sites run ssh as root unnecessarily; the same principles should apply there. (ssh has had vulnerabilities in the past as well). I could have used gpg’s built-in compression, but zstd is faster and better, so we can get good performance by using fast compression and piping that to an algorithm that can use hardware acceleration for encryption.

I strongly encourage considering transport, whether ssh or NNCP or UUCP, to be untrusted. Don’t run it as root if you can avoid it. In my example, the nncp user, which all NNCP commands are run as, has no access to the backup data at all. So even if NNCP were compromised, my backup data wouldn’t be. For even more security, I could also sign the backup stream with gpg and validate that on the receiving end.

I should note, however, that this conversation assumes that a network- or USB-facing ssh or NNCP is more likely to have an exploitable vulnerability than is gpg (which here is just processing a stream). This is probably a safe assumption in general. If you believe gpg is more likely to have an exploitable vulnerability than ssh or NNCP, then obviously you wouldn’t take this particular approach.

On the zfs side, the use of -F with zfs receive is avoided; this could lead to a compromised backed-up machine generating a malicious rollback on the destination. Backup zpools should be imported with -R or -N to ensure that a malicious mountpoint property couldn’t be used to cause an attack. I choose to use “zfs receive -u -o readonly=on” which is compatible with both unmounted backup datasets and zpools imported with -R (or both). To access the data in a backup dataset, you would normally clone it and access it there.

The processing script

So, put this all together and look at an example of a processing script that would run from cron as root and process the incoming ZFS data.

#!/bin/bash
set -e
set -o pipefail

# Log a message
logit () {
   logger -p info -t "`basename "$0"`[$$]" "$1"
}

# Log an error message
logerror () {
   logger -p err -t "`basename "$0"`[$$]" "$1"
}

# Log stdin with the given code.  Used normally to log stderr.
logstdin () {
   logger -p info -t "`basename "$0"`[$$/$1]"
}

# Run command, logging stderr and exit code
runcommand () {
   logit "Running $*"
   if "$@" 2> >(logstdin "$1") ; then
      logit "$1 exited successfully"
      return 0
   else
       RETVAL="$?"
       logerror "$1 exited with error $RETVAL"
       return "$RETVAL"
   fi
}

STORE=backups/simplesnap
INCOMINGDIR=/backups/nncp/incoming

if ! [ -d "$INCOMINGDIR" ]; then
        logerror "$INCOMINGDIR doesn't exist"
        exit 0
fi

LOCKFILE="/backups/nncp/.nncp-backups-zfs-scan.lock"
printf -v EVAL_SAFE_LOCKFILE '%q' "$LOCKFILE"
if dotlockfile -r 0 -l -p "${LOCKFILE}"; then
  logit "Lock obtained at ${LOCKFILE} with dotlockfile"
  trap 'ECODE=$?; dotlockfile -u '"${EVAL_SAFE_LOCKFILE}"'; exit $ECODE' EXIT INT TERM
else
  logit "Could not obtain lock at $LOCKFILE; $0 likely already running."
  exit 0
fi

EXITCODE=0


cd "$INCOMINGDIR"
logit "Scanning queue directory..."
for HOST in *; do
    HOSTPATH="$INCOMINGDIR/$HOST"
    # files like backupsfmt2-134.13134_dest
    for FILE in "$HOSTPATH"/backupsfmt2-[0-9]*_?*; do
        if [ ! -f "$FILE" ]; then
            logit "Skipping non-existent $FILE"
            continue
        fi

        # Now, $DEST will be HOST/DEST.  Strip off the @ also.
        DEST="`echo "$FILE" | sed -e 's/^.*backupsfmt2[^_]*_//' -e 's,@,/,g'`"

        if [ -z "$DEST" ]; then
            logerror "Malformed dest in $FILE"
            continue
        fi
        HOST2="`echo "$DEST" | sed 's,/.*,,g'`"
        if [ -z "$HOST2" ]; then
            logerror "Malformed DEST $DEST in $FILE"
            continue
        fi

        if [ ! "$HOST" = "$HOST2" ]; then
            logerror "$DIR: $HOST doesn't match $HOST2"
            continue
        fi

        logit "Processing $FILE to $STORE/$DEST"
            if runcommand gpg -q -d < "$FILE" | runcommand zstdcat | runcommand zfs receive -u -o readonly=on "$STORE/$DEST"; then
                logit "Successfully processed $FILE to $STORE/$DEST"
                runcommand rm "$FILE"
        else
                logerror "FAILED to process $FILE to $STORE/$DEST"
                EXITCODE=15
        fi

Applying These Ideas to Non-ZFS Backups

ZFS backups made our job easier in a lot of ways:

  • ZFS can calculate a diff based on an efficiently-stored previous local state (snapshot or bookmark), rather than a comparison to a remote state (rsync)
  • ZFS "incremental" sends, while less efficient than rsync, are reasonably efficient, sending only changed blocks
  • ZFS receive detects and enforces that the incremental source on the local machine must match the incremental source of the original stream, enforcing ordering
  • Datasets using ZFS encryption can be sent in their encrypted state
  • Incrementals can be done without a full scan of the filesystem

Some of these benefits you just won't get without ZFS (or something similar like btrfs), but let's see how we could apply these ideas to non-ZFS backups. I will explore the implementation of them in a future post.

When I say "non ZFS", I am being a bit vague as to whether the source, the destination, or both systems are running a non-ZFS filesystem. In general I'll assume that neither are ZFS.

The first and most obvious answer is to just tar up the whole system and send that every day. This is, of course, only suitable for small datasets on a fast network. These tarballs could be unpacked on the destination and stored more efficiently via any number of methods (hardlink trees, a block-level deduplicator like borg or rdedup, or even just simply compressed tarballs).

To make the network trip more efficient, something like rdiff or xdelta could be used. A signature file could be stored on the machine being backed up (generated via tee/pee at stream time), and the next run could simply send an rdiff delta over NNCP. This would be quite network-efficient, but still would require reading every byte of every file on every backup, and would also require quite a bit of temporary space on the receiving end (to apply the delta to the previous tarball and generate a new one).

Alternatively, a program that generates incremental backup files such as rdup could be used. These could be transmitted over NNCP to the backup server, and unpacked there. While perhaps less efficient on the network -- every file with at least one modified byte would be retransmitted in its entirety -- it avoids the need to read every byte of unmodified files or to have enormous temporary space. I should note here that GNU tar claims to have an incremental mode, but it has a potential data loss bug.

There are also some tools with algorithms that may apply well in this use care: syrep and fssync being the two most prominent examples, though rdedup (mentioned above) and the nascent asuran project may also be combinable with other tools to achieve this effect.

I should, of course, conclude this section by mentioning btrfs. Every time I've tried it, I've run into serious bugs, and its status page indicates that only some of them have been resolved. I would not consider using it for something as important as backups. However, if you are comfortable with it, it is likely to be able to run in more constrained environments than ZFS and could probably be processed in much the same way as zfs streams.