Wednesday, April 1, 2015

Breaking news

It was brought to my attention recently by reputable sources that the recent announcement of increased usage in recent years produced an internal firestorm within Mozilla. Key figures raised alarm that some of the tech press had interpreted the blog post as a sign that Thunderbird was not, in fact, dead. As a result, they asked Thunderbird community members to make corrections to emphasize that Mozilla was trying to kill Thunderbird.

The primary fear, it seems, is that knowledge that the largest open-source email client was still receiving regular updates would impel its userbase to agitate for increased funding and maintenance of the client to help forestall potential threats to the open nature of email as well as to innovate in the space of providing usable and private communication channels. Such funding, however, would be an unaffordable luxury and would only distract Mozilla from its central goal of building developer productivity tooling. Persistent rumors that Mozilla would be willing to fund Thunderbird were it renamed Firefox Email were finally addressed with the comment, "such a renaming would violate our current policy that all projects be named Persona."

Monday, January 12, 2015

Why email is hard, part 8: why email security failed

This post is part 8 of an intermittent series exploring the difficulties of writing an email client. Part 1 describes a brief history of the infrastructure. Part 2 discusses internationalization. Part 3 discusses MIME. Part 4 discusses email addresses. Part 5 discusses the more general problem of email headers. Part 6 discusses how email security works in practice. Part 7 discusses the problem of trust. This part discusses why email security has largely failed.

At the end of the last part in this series, I posed the question, "Which email security protocol is most popular?" The answer to the question is actually neither S/MIME nor PGP, but a third protocol, DKIM. I haven't brought up DKIM until now because DKIM doesn't try to secure email in the same vein as S/MIME or PGP, but I still consider it relevant to discussing email security.

Unquestionably, DKIM is the only security protocol for email that can be considered successful. There are perhaps 4 billion active email addresses [1]. Of these, about 1-2 billion use DKIM. In contrast, S/MIME can count a few million users, and PGP at best a few hundred thousand. No other security protocols have really caught on past these three. Why did DKIM succeed where the others fail?

DKIM's success stems from its relatively narrow focus. It is nothing more than a cryptographic signature of the message body and a smattering of headers, and is itself stuck in the DKIM-Signature header. It is meant to be applied to messages only on outgoing servers and read and processed at the recipient mail server—it completely bypasses clients. That it bypasses clients allows it to solve the problem of key discovery and key management very easily (public keys are stored in DNS, which is already a key part of mail delivery), and its role in spam filtering is strong motivation to get it implemented quickly (it is 7 years old as of this writing). It's also simple: this one paragraph description is basically all you need to know [2].

The failure of S/MIME and PGP to see large deployment is certainly a large topic of discussion on myriads of cryptography enthusiast mailing lists, which often like to partake in propositions of new end-to-end encryption of email paradigms, such as the recent DIME proposal. Quite frankly, all of these solutions suffer broadly from at least the same 5 fundamental weaknesses, and I see it unlikely that a protocol will come about that can fix these weaknesses well enough to become successful.

The first weakness, and one I've harped about many times already, is UI. Most email security UI is abysmal and generally at best usable only by enthusiasts. At least some of this is endemic to security: while it mean seem obvious how to convey what an email signature or an encrypted email signifies, how do you convey the distinctions between sign-and-encrypt, encrypt-and-sign, or an S/MIME triple wrap? The Web of Trust model used by PGP (and many other proposals) is even worse, in that inherently requires users to do other actions out-of-band of email to work properly.

Trust is the second weakness. Consider that, for all intents and purposes, the email address is the unique identifier on the Internet. By extension, that implies that a lot of services are ultimately predicated on the notion that the ability to receive and respond to an email is a sufficient means to identify an individual. However, the entire purpose of secure email, or at least of end-to-end encryption, is subtly based on the fact that other people in fact have access to your mailbox, thus destroying the most natural ways to build trust models on the Internet. The quest for anonymity or privacy also renders untenable many other plausible ways to establish trust (e.g., phone verification or government-issued ID cards).

Key discovery is another weakness, although it's arguably the easiest one to solve. If you try to keep discovery independent of trust, the problem of key discovery is merely picking a protocol to publish and another one to find keys. Some of these already exist: PGP key servers, for example, or using DANE to publish S/MIME or PGP keys.

Key management, on the other hand, is a more troubling weakness. S/MIME, for example, basically works without issue if you have a certificate, but managing to get an S/MIME certificate is a daunting task (necessitated, in part, by its trust model—see how these issues all intertwine?). This is also where it's easy to say that webmail is an unsolvable problem, but on further reflection, I'm not sure I agree with that statement anymore. One solution is just storing the private key with the webmail provider (you're trusting them as an email client, after all), but it's also not impossible to imagine using phones or flash drives as keystores. Other key management factors are more difficult to solve: people who lose their private keys or key rollover create thorny issues. There is also the difficulty of managing user expectations: if I forget my password to most sites (even my email provider), I can usually get it reset somehow, but when a private key is lost, the user is totally and completely out of luck.

Of course, there is one glaring and almost completely insurmountable problem. Encrypted email fundamentally precludes certain features that we have come to take for granted. The lesser known is server-side search and filtration. While there exist some mechanisms to do search on encrypted text, those mechanisms rely on the fact that you can manipulate the text to change the message, destroying the integrity feature of secure email. They also tend to be fairly expensive. It's easy to just say "who needs server-side stuff?", but the contingent of people who do email on smartphones would not be happy to have to pay the transfer rates to download all the messages in their folder just to find one little email, nor the energy costs of doing it on the phone. And those who have really large folders—Fastmail has a design point of 1,000,000 in a single folder—would still prefer to not have to transfer all their mail even on desktops.

The more well-known feature that would disappear is spam filtration. Consider that 90% of all email is spam, and if you think your spam folder is too slim for that to be true, it's because your spam folder only contains messages that your email provider wasn't sure were spam. The loss of server-side spam filtering would dramatically increase the cost of spam (a 10% reduction in efficiency would double the amount of server storage, per my calculations), and client-side spam filtering is quite literally too slow [3] and too costly (remember smartphones? Imagine having your email take 10 times as much energy and bandwidth) to be a tenable option. And privacy or anonymity tends to be an invitation to abuse (cf. Tor and Wikipedia). Proposed solutions to the spam problem are so common that there is a checklist containing most of the objections.

When you consider all of those weaknesses, it is easy to be pessimistic about the possibility of wide deployment of powerful email security solutions. The strongest future—all email is encrypted, including metadata—is probably impossible or at least woefully impractical. That said, if you weaken some of the assumptions (say, don't desire all or most traffic to be encrypted), then solutions seem possible if difficult.

This concludes my discussion of email security, at least until things change for the better. I don't have a topic for the next part in this series picked out (this part actually concludes the set I knew I wanted to discuss when I started), although OAuth and DMARC are two topics that have been bugging me enough recently to consider writing about. They also have the unfortunate side effect of being things likely to see changes in the near future, unlike most of the topics I've discussed so far. But rest assured that I will find more difficulties in the email infrastructure to write about before long!

[1] All of these numbers are crude estimates and are accurate to only an order of magnitude. To justify my choices: I assume 1 email address per Internet user (this overestimates the developing world and underestimates the developed world). The largest webmail providers have given numbers that claim to be 1 billion active accounts between them, and all of them use DKIM. S/MIME is guessed by assuming that any smartcard deployment supports S/MIME, and noting that the US Department of Defense and Estonia's digital ID project are both heavy users of such smartcards. PGP is estimated from the size of the strong set and old numbers on the reachable set from the core Web of Trust.
[2] Ever since last April, it's become impossible to mention DKIM without referring to DMARC, as a result of Yahoo's controversial DMARC policy. A proper discussion of DMARC (and why what Yahoo did was controversial) requires explaining the mail transmission architecture and spam, however, so I'll defer that to a later post. It's also possible that changes in this space could happen within the next year.
[3] According to a former GMail spam employee, if it takes you as long as three minutes to calculate reputation, the spammer wins.

Saturday, January 10, 2015

A unified history for comm-central

Several years back, Ehsan and Jeff Muizelaar attempted to build a unified history of mozilla-central across the Mercurial era and the CVS era. Their result is now used in the gecko-dev repository. While being distracted on yet another side project, I thought that I might want to do the same for comm-central. It turns out that building a unified history for comm-central makes mozilla-central look easy: mozilla-central merely had one import from CVS. In contrast, comm-central imported twice from CVS (the calendar code came later), four times from mozilla-central (once with converted history), and imported twice from Instantbird's repository (once with converted history). Three of those conversions also involved moving paths. But I've worked through all of those issues to provide a nice snapshot of the repository [1]. And since I've been frustrated by failing to find good documentation on how this sort of process went for mozilla-central, I'll provide details on the process for comm-central.

The first step and probably the hardest is getting the CVS history in DVCS form (I use hg because I'm more comfortable it, but there's effectively no difference between hg, git, or bzr here). There is a git version of mozilla's CVS tree available, but I've noticed after doing research that its last revision is about a month before the revision I need for Calendar's import. The documentation for how that repo was built is no longer on the web, although we eventually found a copy after I wrote this post on I tried doing another conversion using hg convert to get CVS tags, but that rudely blew up in my face. For now, I've filed a bug on getting an official, branchy-and-tag-filled version of this repository, while using the current lack of history as a base. Calendar people will have to suffer missing a month of history.

CVS is famously hard to convert to more modern repositories, and, as I've done my research, Mozilla's CVS looks like it uses those features which make it difficult. In particular, both the calendar CVS import and the comm-central initial CVS import used a CVS tag HG_COMM_INITIAL_IMPORT. That tagging was done, on only a small portion of the tree, twice, about two months apart. Fortunately, mailnews code was never touched on CVS trunk after the import (there appears to be one commit on calendar after the tagging), so it is probably possible to salvage a repository-wide consistent tag.

The start of my script for conversion looks like this:


set -e


# Bug 445146: m-c/editor/ui -> c-c/editor/ui

# Bug 669040: m-c/db/mork -> c-c/db/mork

# Bug 1027241, bug 611752 m-c/security/manager/ssl/** -> c-c/mailnews/mime/src/*

# Step 0: Grab the mozilla CVS history.
if [ ! -e $HGCVS ]; then
  hg clone git+ $HGCVS

Since I don't want to include the changesets useless to comm-central history, I trimmed the history by using hg convert to eliminate changesets that don't change the necessary files. Most of the files are simple directory-wide changes, but S/MIME only moved a few files over, so it requires a more complex way to grab the file list. In addition, I also replaced the % in the usernames with @ that they are used to appearing in hg. The relevant code is here:

# Step 1: Trim mozilla CVS history to include only the files we are ultimately
# interested in.
cat >$WORKDIR/convert-filemap.txt <<EOF
# Revision e4f4569d451a
include directory/xpcom
include mail
include mailnews
include other-licenses/branding/thunderbird
include suite
# Revision 7c0bfdcda673
include calendar
include other-licenses/branding/sunbird
# Revision ee719a0502491fc663bda942dcfc52c0825938d3
include editor/ui
# Revision 52efa9789800829c6f0ee6a005f83ed45a250396
include db/mork/
include db/mdb/

# Add the S/MIME import files
hg -R $MC log -r "children($MC_SMIME_IMPORT)" \
  --template "{file_dels % 'include {file}\n'}" >>$WORKDIR/convert-filemap.txt

if [ ! -e $WORKDIR/convert-authormap.txt ]; then
hg -R $HGCVS log --template "{email(author)}={sub('%', '@', email(author))}\n" \
  | sort -u > $WORKDIR/convert-authormap.txt

hg convert $HGCVS $OUTPUT --filemap convert-filemap.txt -A convert-authormap.txt

That last command provides us the subset of the CVS history that we need for unified history. Strictly speaking, I should be pulling a specific revision, but I happen to know that there's no need to (we're cloning the only head) in this case. At this point, we now need to pull in the mozilla-central changes before we pull in comm-central. Order is key; hg convert will only apply the graft points when converting the child changeset (which it does but once), and it needs the parents to exist before it can do that. We also need to ensure that the mozilla-central graft point is included before continuing, so we do that, and then pull mozilla-central:

CC_CVS_BASE=$(hg log -R $HGCVS -r 'tip' --template '{node}')
CC_CVS_BASE=$(grep $CC_CVS_BASE $OUTPUT/.hg/shamap | cut -d' ' -f2)
MC_CVS_BASE=$(hg log -R $HGCVS -r 'gitnode(215f52d06f4260fdcca797eebd78266524ea3d2c)' --template '{node}')
MC_CVS_BASE=$(grep $MC_CVS_BASE $OUTPUT/.hg/shamap | cut -d' ' -f2)

# Okay, now we need to build the map of revisions.
cat >$WORKDIR/convert-revmap.txt <<EOF
e4f4569d451a5e0d12a6aa33ebd916f979dd8faa $CC_CVS_BASE # Thunderbird / Suite
7c0bfdcda6731e77303f3c47b01736aaa93d5534 d4b728dc9da418f8d5601ed6735e9a00ac963c4e, $CC_CVS_BASE # Calendar
9b2a99adc05e53cd4010de512f50118594756650 $MC_CVS_BASE # Mozilla graft point
ee719a0502491fc663bda942dcfc52c0825938d3 78b3d6c649f71eff41fe3f486c6cc4f4b899fd35, $MC_EDITOR_IMPORT # Editor
8cdfed92867f885fda98664395236b7829947a1d 4b5da7e5d0680c6617ec743109e6efc88ca413da, e4e612fcae9d0e5181a5543ed17f705a83a3de71 # Chat

# Next, import mozilla-central revisions
  hg convert $MC $OUTPUT -r $rev --splicemap $WORKDIR/convert-revmap.txt \
    --filemap $WORKDIR/convert-filemap.txt

Some notes about all of the revision ids in the script. The splicemap requires the full 40-character SHA ids; anything less and the thing complains. I also need to specify the parents of the revisions that deleted the code for the mozilla-central import, so if you go hunting for those revisions and are surprised that they don't remove the code in question, that's why.

I mentioned complications about the merges earlier. The Mork and S/MIME import codes here moved files, so that what was db/mdb in mozilla-central became db/mork. There's no support for causing the generated splice to record these as a move, so I have to manually construct those renamings:

# We need to execute a few hg move commands due to renamings.
pushd $OUTPUT
hg update -r $(grep $MC_MORK_IMPORT .hg/shamap | cut -d' ' -f2)
(hg -R $MC log -r "children($MC_MORK_IMPORT)" \
  --template "{file_dels % 'hg mv {file} {sub(\"db/mdb\", \"db/mork\", file)}\n'}") | bash
hg commit -m 'Pseudo-changeset to move Mork files' -d '2011-08-06 17:25:21 +0200'
MC_MORK_IMPORT=$(hg log -r tip --template '{node}')

hg update -r $(grep $MC_SMIME_IMPORT .hg/shamap | cut -d' ' -f2)
(hg -R $MC log -r "children($MC_SMIME_IMPORT)" \
  --template "{file_dels % 'hg mv {file} {sub(\"security/manager/ssl\", \"mailnews/mime\", file)}\n'}") | bash
hg commit -m 'Pseudo-changeset to move S/MIME files' -d '2014-06-15 20:51:51 -0700'
MC_SMIME_IMPORT=$(hg log -r tip --template '{node}')

# Echo the new move commands to the changeset conversion map.
cat >>$WORKDIR/convert-revmap.txt <<EOF
52efa9789800829c6f0ee6a005f83ed45a250396 abfd23d7c5042bc87502506c9f34c965fb9a09d1, $MC_MORK_IMPORT # Mork
50f5b5fc3f53c680dba4f237856e530e2097adfd 97253b3cca68f1c287eb5729647ba6f9a5dab08a, $MC_SMIME_IMPORT # S/MIME

Now that we have all of the graft points defined, and all of the external code ready, we can pull comm-central and do the conversion. That's not quite it, though—when we graft the S/MIME history to the original mozilla-central history, we have a small segment of abandoned converted history. A call to hg strip removes that.

# Now, import comm-central revisions that we need
hg convert $CC $OUTPUT --splicemap $WORKDIR/convert-revmap.txt
hg strip 2f69e0a3a05a

[1] I left out one of the graft points because I just didn't want to deal with it. I'll leave it as an exercise to the reader to figure out which one it was. Hint: it's the only one I didn't know about before I searched for the archive points [2].
[2] Since I wasn't sure I knew all of the graft points, I decided to try to comb through all of the changesets to figure out who imported code. It turns out that hg log -r 'adds("**")' narrows it down nicely (1667 changesets to look at instead of 17547), and using the {file_adds} template helps winnow it down more easily.

Tuesday, August 5, 2014

Why email is hard, part 7: email security and trust

This post is part 7 of an intermittent series exploring the difficulties of writing an email client. Part 1 describes a brief history of the infrastructure. Part 2 discusses internationalization. Part 3 discusses MIME. Part 4 discusses email addresses. Part 5 discusses the more general problem of email headers. Part 6 discusses how email security works in practice. This part discusses the problem of trust.

At a technical level, S/MIME and PGP (or at least PGP/MIME) use cryptography essentially identically. Yet the two are treated as radically different models of email security because they diverge on the most important question of public key cryptography: how do you trust the identity of a public key? Trust is critical, as it is the only way to stop an active, man-in-the-middle (MITM) attack. MITM attacks are actually easier to pull off in email, since all email messages effectively have to pass through both the sender's and the recipients' email servers [1], allowing attackers to be able to pull off permanent, long-lasting MITM attacks [2].

S/MIME uses the same trust model that SSL uses, based on X.509 certificates and certificate authorities. X.509 certificates effectively work by providing a certificate that says who you are which is signed by another authority. In the original concept (as you might guess from the name "X.509"), the trusted authority was your telecom provider, and the certificates were furthermore intended to be a part of the global X.500 directory—a natural extension of the OSI internet model. The OSI model of the internet never gained traction, and the trusted telecom providers were replaced with trusted root CAs.

PGP, by contrast, uses a trust model that's generally known as the Web of Trust. Every user has a PGP key (containing their identity and their public key), and users can sign others' public keys. Trust generally flows from these signatures: if you trust a user, you know the keys that they sign are correct. The name "Web of Trust" comes from the vision that trust flows along the paths of signatures, building a tight web of trust.

And now for the controversial part of the post, the comparisons and critiques of these trust models. A disclaimer: I am not a security expert, although I am a programmer who revels in dreaming up arcane edge cases. I also don't use PGP at all, and use S/MIME to a very limited extent for some Mozilla work [3], although I did try a few abortive attempts to dogfood it in the past. I've attempted to replace personal experience with comprehensive research [4], but most existing critiques and comparisons of these two trust models are about 10-15 years old and predate several changes to CA certificate practices.

A basic tenet of development that I have found is that the average user is fairly ignorant. At the same time, a lot of the defense of trust models, both CAs and Web of Trust, tends to hinge on configurability. How many people, for example, know how to add or remove a CA root from Firefox, Windows, or Android? Even among the subgroup of Mozilla developers, I suspect the number of people who know how to do so are rather few. Or in the case of PGP, how many people know how to change the maximum path length? Or even understand the security implications of doing so?

Seen in the light of ignorant users, the Web of Trust is a UX disaster. Its entire security model is predicated on having users precisely specify how much they trust other people to trust others (ultimate, full, marginal, none, unknown) and also on having them continually do out-of-band verification procedures and publicly reporting those steps. In 1998, a seminal paper on the usability of a GUI for PGP encryption came to the conclusion that the UI was effectively unusable for users, to the point that only a third of the users were able to send an encrypted email (and even then, only with significant help from the test administrators), and a quarter managed to publicly announce their private keys at some point, which is pretty much the worst thing you can do. They also noted that the complex trust UI was never used by participants, although the failure of many users to get that far makes generalization dangerous [5]. While newer versions of security UI have undoubtedly fixed many of the original issues found (in no small part due to the paper, one of the first to argue that usability is integral, not orthogonal, to security), I have yet to find an actual study on the usability of the trust model itself.

The Web of Trust has other faults. The notion of "marginal" trust it turns out is rather broken: if you marginally trust a user who has two keys who both signed another person's key, that's the same as fully trusting a user with one key who signed that key. There are several proposals for different trust formulas [6], but none of them have caught on in practice to my knowledge.

A hidden fault is associated with its manner of presentation: in sharp contrast to CAs, the Web of Trust appears to not delegate trust, but any practical widespread deployment needs to solve the problem of contacting people who have had no prior contact. Combined with the need to bootstrap new users, this implies that there needs to be some keys that have signed a lot of other keys that are essentially default-trusted—in other words, a CA, a fact sometimes lost on advocates of the Web of Trust.

That said, a valid point in favor of the Web of Trust is that it more easily allows people to distrust CAs if they wish to. While I'm skeptical of its utility to a broader audience, the ability to do so for is crucial for a not-insignificant portion of the population, and it's important enough to be explicitly called out.

X.509 certificates are most commonly discussed in the context of SSL/TLS connections, so I'll discuss them in that context as well, as the implications for S/MIME are mostly the same. Almost all criticism of this trust model essentially boils down to a single complaint: certificate authorities aren't trustworthy. A historical criticism is that the addition of CAs to the main root trust stores was ad-hoc. Since then, however, the main oligopoly of these root stores (Microsoft, Apple, Google, and Mozilla) have made their policies public and clear [7]. The introduction of the CA/Browser Forum in 2005, with a collection of major CAs and the major browsers as members, and several [8] helps in articulating common policies. These policies, simplified immensely, boil down to:

  1. You must verify information (depending on certificate type). This information must be relatively recent.
  2. You must not use weak algorithms in your certificates (e.g., no MD5).
  3. You must not make certificates that are valid for too long.
  4. You must maintain revocation checking services.
  5. You must have fairly stringent physical and digital security practices and intrusion detection mechanisms.
  6. You must be [externally] audited every year that you follow the above rules.
  7. If you screw up, we can kick you out.

I'm not going to claim that this is necessarily the best policy or even that any policy can feasibly stop intrusions from happening. But it's a policy, so CAs must abide by some set of rules.

Another CA criticism is the fear that they may be suborned by national government spy agencies. I find this claim underwhelming, considering that the number of certificates acquired by intrusions that were used in the wild is larger than the number of certificates acquired by national governments that were used in the wild: 1 and 0, respectively. Yet no one complains about the untrustworthiness of CAs due to their ability to be hacked by outsiders. Another attack is that CAs are controlled by profit-seeking corporations, which misses the point because the business of CAs is not selling certificates but selling their access to the root databases. As we will see shortly, jeopardizing that access is a great way for a CA to go out of business.

To understand issues involving CAs in greater detail, there are two CAs that are particularly useful to look at. The first is CACert. CACert is favored by many by its attempt to handle X.509 certificates in a Web of Trust model, so invariably every public discussion about CACert ends up devolving into an attack on other CAs for their perceived capture by national governments or corporate interests. Yet what many of the proponents for inclusion of CACert miss (or dismiss) is the fact that CACert actually failed the required audit, and it is unlikely to ever pass an audit. This shows a central failure of both CAs and Web of Trust: different people have different definitions of "trust," and in the case of CACert, some people are favoring a subjective definition (I trust their owners because they're not evil) when an objective definition fails (in this case, that the root signing key is securely kept).

The other CA of note here is DigiNotar. In July 2011, some hackers managed to acquire a few fraudulent certificates by hacking into DigiNotar's systems. By late August, people had become aware of these certificates being used in practice [9] to intercept communications, mostly in Iran. The use appears to have been caught after Chromium updates failed due to invalid certificate fingerprints. After it became clear that the fraudulent certificates were not limited to a single fake Google certificate, and that DigiNotar had failed to notify potentially affected companies of its breach, DigiNotar was swiftly removed from all of the trust databases. It ended up declaring bankruptcy within two weeks.

DigiNotar indicates several things. One, SSL MITM attacks are not theoretical (I have seen at least two or three security experts advising pre-DigiNotar that SSL MITM attacks are "theoretical" and therefore the wrong target for security mechanisms). Two, keeping the trust of browsers is necessary for commercial operation of CAs. Three, the notion that a CA is "too big to fail" is false: DigiNotar played an important role in the Dutch community as a major CA and the operator of Staat der Nederlanden. Yet when DigiNotar screwed up and lost its trust, it was swiftly kicked out despite this role. I suspect that even Verisign could be kicked out if it manages to screw up badly enough.

This isn't to say that the CA model isn't problematic. But the source of its problems is that delegating trust isn't a feasible model in the first place, a problem that it shares with the Web of Trust as well. Different notions of what "trust" actually means and the uncertainty that gets introduced as chains of trust get longer both make delegating trust weak to both social engineering and technical engineering attacks. There appears to be an increasing consensus that the best way forward is some variant of key pinning, much akin to how SSH works: once you know someone's public key, you complain if that public key appears to change, even if it appears to be "trusted." This does leave people open to attacks on first use, and the question of what to do when you need to legitimately re-key is not easy to solve.

In short, both CAs and the Web of Trust have issues. Whether or not you should prefer S/MIME or PGP ultimately comes down to the very conscious question of how you want to deal with trust—a question without a clear, obvious answer. If I appear to be painting CAs and S/MIME in a positive light and the Web of Trust and PGP in a negative one in this post, it is more because I am trying to focus on the positions less commonly taken to balance perspective on the internet. In my next post, I'll round out the discussion on email security by explaining why email security has seen poor uptake and answering the question as to which email security protocol is most popular. The answer may surprise you!

[1] Strictly speaking, you can bypass the sender's SMTP server. In practice, this is considered a hole in the SMTP system that email providers are trying to plug.
[2] I've had 13 different connections to the internet in the same time as I've had my main email address, not counting all the public wifis that I have used. Whereas an attacker would find it extraordinarily difficult to intercept all of my SSH sessions for a MITM attack, intercepting all of my email sessions is clearly far easier if the attacker were my email provider.
[3] Before you read too much into this personal choice of S/MIME over PGP, it's entirely motivated by a simple concern: S/MIME is built into Thunderbird; PGP is not. As someone who does a lot of Thunderbird development work that could easily break the Enigmail extension locally, needing to use an extension would be disruptive to workflow.
[4] This is not to say that I don't heavily research many of my other posts, but I did go so far for this one as to actually start going through a lot of published journals in an attempt to find information.
[5] It's questionable how well the usability of a trust model UI can be measured in a lab setting, since the observer effect is particularly strong for all metrics of trust.
[6] The web of trust makes a nice graph, and graphs invite lots of interesting mathematical metrics. I've always been partial to eigenvectors of the graph, myself.
[7] Mozilla's policy for addition to NSS is basically the standard policy adopted by all open-source Linux or BSD distributions, seeing as OpenSSL never attempted to produce a root database.
[8] It looks to me that it's the browsers who are more in charge in this forum than the CAs.
[9] To my knowledge, this is the first—and so far only—attempt to actively MITM an SSL connection.

Monday, May 26, 2014

Why email is hard, part 6: today's email security

This post is part 6 of an intermittent series exploring the difficulties of writing an email client. Part 1 describes a brief history of the infrastructure. Part 2 discusses internationalization. Part 3 discusses MIME. Part 4 discusses email addresses. Part 5 discusses the more general problem of email headers. This part discusses how email security works in practice.

Email security is a rather wide-ranging topic, and one that I've wanted to cover for some time, well before several recent events that have made it come up in the wider public knowledge. There is no way I can hope to cover it in a single post (I think it would outpace even the length of my internationalization discussion), and there are definitely parts for which I am underqualified, as I am by no means an expert in cryptography. Instead, I will be discussing this over the course of several posts of which this is but the first; to ease up on the amount of background explanation, I will assume passing familiarity with cryptographic concepts like public keys, hash functions, as well as knowing what SSL and SSH are (though not necessarily how they work). If you don't have that knowledge, ask Wikipedia.

Before discussing how email security works, it is first necessary to ask what email security actually means. Unfortunately, the layman's interpretation is likely going to differ from the actual precise definition. Security is often treated by laymen as a boolean interpretation: something is either secure or insecure. The most prevalent model of security to people is SSL connections: these allow the establishment of a communication channel whose contents are secret to outside observers while also guaranteeing to the client the authenticity of the server. The server often then gets authenticity of the client via a more normal authentication scheme (i.e., the client sends a username and password). Thus there is, at the end, a channel that has both secrecy and authenticity [1]: channels with both of these are considered secure and channels without these are considered insecure [2].

In email, the situation becomes more difficult. Whereas an SSL connection is between a client and a server, the architecture of email is such that email providers must be considered as distinct entities from end users. In addition, messages can be sent from one person to multiple parties. Thus secure email is a more complex undertaking than just porting relevant details of SSL. There are two major cryptographic implementations of secure email [3]: S/MIME and PGP. In terms of implementation, they are basically the same [4], although PGP has an extra mode which wraps general ASCII (known as "ASCII-armor"), which I have been led to believe is less recommended these days. Since I know the S/MIME specifications better, I'll refer specifically to how S/MIME works.

S/MIME defines two main MIME types: multipart/signed, which contains the message text as a subpart followed by data indicating the cryptographic signature, and application/pkcs7-mime, which contains an encrypted MIME part. The important things to note about this delineation are that only the body data is encrypted [5], that it's theoretically possible to encrypt only part of a message's body, and that the signing and encryption constitute different steps. These factors combine to make for a potentially infuriating UI setup.

How does S/MIME tackle the challenges of encrypting email? First, rather than encrypting using recipients' public keys, the message is encrypted with a symmetric key. This symmetric key is then encrypted with each of the recipients' keys and then attached to the message. Second, by only signing or encrypting the body of the message, the transit headers are kept intact for the mail system to retain its ability to route, process, and deliver the message. The body is supposed to be prepared in the "safest" form before transit to avoid intermediate routers munging the contents. Finally, to actually ascertain what the recipients' public keys are, clients typically passively pull the information from signed emails. LDAP, unsurprisingly, contains an entry for a user's public key certificate, which could be useful in large enterprise deployments. There is also work ongoing right now to publish keys via DNS and DANE.

I mentioned before that S/MIME's use can present some interesting UI design decisions. I ended up actually testing some common email clients on how they handled S/MIME messages: Thunderbird, Apple Mail, Outlook [6], and Evolution. In my attempts to create a surreptitious signed part to confuse the UI, Outlook decided that the message had no body at all, and Thunderbird decided to ignore all indication of the existence of said part. Apple Mail managed to claim the message was signed in one of these scenarios, and Evolution took the cake by always agreeing that the message was signed [7]. It didn't even bother questioning the signature if the certificate's identity disagreed with the easily-spoofable From address. I was actually surprised by how well people did in my tests—I expected far more confusion among clients, particularly since the will to maintain S/MIME has clearly been relatively low, judging by poor support for "new" features such as triple-wrapping or header protection.

Another fault of S/MIME's design is that it makes the mistaken belief that composing a signing step and an encryption step is equivalent in strength to a simultaneous sign-and-encrypt. Another page describes this in far better detail than I have room to; note that this flaw is fixed via triple-wrapping (which has relatively poor support). This creates yet more UI burden into how to adequately describe in UI all the various minutiae in differing security guarantees. Considering that users already have a hard time even understanding that just because a message says it's from example@isp.invalid doesn't actually mean it's from example@isp.invalid, trying to develop UI that both adequately expresses the security issues and is understandable to end-users is an extreme challenge.

What we have in S/MIME (and PGP) is a system that allows for strong guarantees, if certain conditions are met, yet is also vulnerable to breaches of security if the message handling subsystems are poorly designed. Hopefully this is a sufficient guide to the technical impacts of secure email in the email world. My next post will discuss the most critical component of secure email: the trust model. After that, I will discuss why secure email has seen poor uptake and other relevant concerns on the future of email security.

[1] This is a bit of a lie: a channel that does secrecy and authentication at different times isn't as secure as one that does them at the same time.
[2] It is worth noting that authenticity is, in many respects, necessary to achieve secrecy.
[3] This, too, is a bit of a lie. More on this in a subsequent post.
[4] I'm very aware that S/MIME and PGP use radically different trust models. Trust models will be covered later.
[5] S/MIME 3.0 did add a provision stating that if the signed/encrypted part is a message/rfc822 part, the headers of that part should override the outer message's headers. However, I am not aware of a major email client that actually handles these kind of messages gracefully.
[6] Actually, I tested Windows Live Mail instead of Outlook, but given the presence of an official MIME-to-Microsoft's-internal-message-format document which seems to agree with what Windows Live Mail was doing, I figure their output would be identical.
[7] On a more careful examination after the fact, it appears that Evolution may have tried to indicate signedness on a part-by-part basis, but the UI was sufficiently confusing that ordinary users are going to be easily confused.

Saturday, April 5, 2014

Announcing jsmime 0.2

Previously, I've been developing JSMime as a subdirectory within comm-central. However, after discussions with other developers, I have moved the official repository of record for JSMime to its own repository, now found on GitHub. The repository has been cleaned up and the feature set for version 0.2 has been selected, so that the current tip on JSMime (also the initial version) is version 0.2. This contains the feature set I imported into Thunderbird's source code last night, which is to say support for parsing MIME messages into the MIME tree, as well as support for parsing and encoding email address headers.

Thunderbird doesn't actually use the new code quite yet (as my current tree is stuck on a mozilla-central build error, so I haven't had time to run those patches through a last minute sanity check before requesting review), but the intent is to replace the current C++ implementations of nsIMsgHeaderParser and nsIMimeConverter with JSMime instead. Once those are done, I will be moving forward with my structured header plans which more or less ought to make those interfaces obsolete.

Within JSMime itself, the pieces which I will be working on next will be rounding out the implementation of header parsing and encoding support (I have prototypes for Date headers and the infernal RFC 2231 encoding that Content-Disposition needs), as well as support for building MIME messages from their constituent parts (a feature which would be greatly appreciated in the depths of compose and import in Thunderbird). I also want to implement full IDN and EAI support, but that's hampered by the lack of a JS implementation I can use for IDN (yes, there's punycode.js, but that doesn't do StringPrep). The important task of converting the MIME tree to a list of body parts and attachments is something I do want to work on as well, but I've vacillated on the implementation here several times and I'm not sure I've found one I like yet.

JSMime, as its name implies, tries to work in as pure JS as possible, augmented with several web APIs as necessary (such as TextDecoder for charset decoding). I'm using ES6 as the base here, because it gives me several features I consider invaluable for implementing JavaScript: Promises, Map, generators, let. This means it can run on an unprivileged web page—I test JSMime using Firefox nightlies and the Firefox debugger where necessary. Unfortunately, it only really works in Firefox at the moment because V8 doesn't support many ES6 features yet (such as destructuring, which is annoying but simple enough to work around, or Map iteration, which is completely necessary for the code). I'm not opposed to changing it to make it work on Node.js or Chrome, but I don't realistically have the time to spend doing it myself; if someone else has the time, please feel free to contact me or send patches.

Thursday, April 3, 2014

If you want fast code, don't use assembly

…unless you're an expert at assembly, that is. The title of this post was obviously meant to be an attention-grabber, but it is much truer than you might think: poorly-written assembly code will probably be slower than an optimizing compiler on well-written code (note that you may need to help the compiler along for things like vectorization). Now why is this?

Modern microarchitectures are incredibly complex. A modern x86 processor will be superscalar and use some form of compilation to microcode to do that. Desktop processors will undoubtedly have multiple instruction issues per cycle, forms of register renaming, branch predictors, etc. Minor changes—a misaligned instruction stream, a poor order of instructions, a bad instruction choice—could kill the ability to take advantages of these features. There are very few people who could accurately predict the performance of a given assembly stream (I myself wouldn't attempt it if the architecture can take advantage of ILP), and these people are disproportionately likely to be working on compiler optimizations. So unless you're knowledgeable enough about assembly to help work on a compiler, you probably shouldn't be hand-coding assembly to make code faster.

To give an example to elucidate this point (and the motivation for this blog post in the first place), I was given a link to an implementation of the N-queens problem in assembly. For various reasons, I decided to use this to start building a fine-grained performance measurement system. This system uses a high-resolution monotonic clock on Linux and runs the function 1000 times to warm up caches and counters and then runs the function 1000 more times, measuring each run independently and reporting the average runtime at the end. This is a single execution of the system; 20 executions of the system were used as the baseline for a t-test to determine statistical significance as well as visual estimation of normality of data. Since the runs observed about a constant 1-2 μs of noise, I ran all of my numbers on the 10-queens problem to better separate the data (total runtimes ended up being in the range of 200-300μs at this level). When I say that some versions are faster, the p-values for individual measurements are on the order of 10-20—meaning that there is a 1-in-100,000,000,000,000,000,000 chance that the observed speedups could be produced if the programs take the same amount of time to run.

The initial assembly version of the program took about 288μs to run. The first C++ version I coded, originating from the same genesis algorithm that the author of the assembly version used, ran in 275μs. A recursive program beat out a hand-written assembly block of code... and when I manually converted the recursive program into a single loop, the runtime improved to 249μs. It wasn't until I got rid of all of the assembly in the original code that I could get the program to beat the derecursified code (at 244μs)—so it's not the vectorization that's causing the code to be slow. Intrigued, I started to analyze why the original assembly was so slow.

It turns out that there are three main things that I think cause the slow speed of the original code. The first one is alignment of branches: the assembly code contains no instructions to align basic blocks on particular branches, whereas gcc happily emits these for some basic blocks. I mention this first as it is mere conjecture; I never made an attempt to measure the effects for myself. The other two causes are directly measured from observing runtime changes as I slowly replaced the assembly with code. When I replaced the use of push and pop instructions with a global static array, the runtime improved dramatically. This suggests that the alignment of the stack could be to blame (although the stack is still 8-byte aligned when I checked via gdb), which just goes to show you how much alignments really do matter in code.

The final, and by far most dramatic, effect I saw involves the use of three assembly instructions: bsf (find the index of the lowest bit that is set), btc (clear a specific bit index), and shl (left shift). When I replaced the use of these instructions with a more complicated expression int bit = x & -x and x = x - bit, the program's speed improved dramatically. And the rationale for why the speed improved won't be found in latency tables, although those will tell you that bsf is not a 1-cycle operation. Rather, it's in minutiae that's not immediately obvious.

The original program used the fact that bsf sets the zero flag if the input register is 0 as the condition to do the backtracking; the converted code just checked if the value was 0 (using a simple test instruction). The compare and the jump instructions are basically converted into a single instruction in the processor. In contrast, the bsf does not get to do this; combined with the lower latency of the instruction intrinsically, it means that empty loops take a lot longer to do nothing. The use of an 8-bit shift value is also interesting, as there is a rather severe penalty for using 8-bit registers in Intel processors as far as I can see.

Now, this isn't to say that the compiler will always produce the best code by itself. My final code wasn't above using x86 intrinsics for the vector instructions. Replacing the _mm_andnot_si128 intrinsic with an actual and-not on vectors caused gcc to use other, slower instructions instead of the vmovq to move the result out of the SSE registers for reasons I don't particularly want to track down. The use of the _mm_blend_epi16 and _mm_srli_si128 intrinsics can probably be replaced with __builtin_shuffle instead for more portability, but I was under the misapprehension that this was a clang-only intrinsic when I first played with the code so I never bothered to try that, and this code has passed out of my memory long enough that I don't want to try to mess with it now.

In short, compilers know things about optimizing for modern architectures that many general programmers don't. Compilers may have issues with autovectorization, but the existence of vector intrinsics allow you to force compilers to use vectorization while still giving them leeway to make decisions about instruction scheduling or code alignment which are easy to screw up in hand-written assembly. Also, compilers are liable to get better in the future, whereas hand-written assembly code is unlikely to get faster in the future. So only write assembly code if you really know what you're doing and you know you're better than the compiler.