Edward Thomson

Changing Titles in Google Authenticator

August 23, 2017  •  5:27 PM

Surely you know that the best practice for securing your accounts is to enable two-factor authentication:

When all that is between you and an attacker getting into your account is a single password, you’re running a risk that is far greater than what you need be taking. A password is one factor – “something you know”. Now if we add something you have such as your mobile phone and the email service verifies your identity when you first log on by sending an SMS to that thing you have, the security position of your email changes fundamentally.

Troy Hunt, 10 email security fundamentals for everyday people

And hopefully you're using an application as your second factor, instead of text messages. Text messages may not work when you travel to foreign countries, but you're also reliant upon your wireless carrier to keep your data secure:

Instead, use TOTP (Time-based One-Time Pad) to get a six digit number from a local application. There are many applications that support TOTP, but I keep it old school, and use the Google Authenticator application.

The problem with the Google Authenticator app, though, is that it doesn't let you edit the title of a website (the "issuer") once you've set it up. So you end up with a number that's missing a title, and there's no good way to identify it.

Here, the first entry is obviously for my Microsoft account, but the second entry…? I have no idea what it's for:

Google Authenticator Missing a Title

Thankfully, TOTP is a published standard, so you can actually create - and then scan - your own QR code based on the secret number that you're given when you turn on two-factor authentication:

Facebook 2FA Enablement

The QR code that you scan to set up a new account is generated by constructing a URL with the secret number and some metadata, and then encoding that with a QR generator. The format is:


The account_name - as the name suggests - reflects the name of your account on the website. This is your username or email address, generally. Google Authenticator shows this as the second line of the key.

The secret_key is the secret key that the web site gives you when you enable TOTP. (In the example above, it's XXXX ABCD XXXX ABCD).

Finally, the issuer is the name of the website itself. This is the larger header displayed above your key.

It's such a simple mechanism that you can just create a new URL with those values and then use your favorite QR generating tool to create a QR code for your custom URL. (Remember to URL-encode any of your values!)

If you don't have a QR generator (I didn't) then you can install the very simple qrencode package and generate a QR code into an image file.

Better still, you can specify ANSI as the output type:

% qrencode -t ANSI otpauth://totp/ethomson@edwardthomson.com?secret=XXXXABCDXXXXABCD&issuer=My%20Title

And it will dump a QR code straight to your console:

QR on the Console

Now you just point Google Authenticator at your terminal window, and you can see that it adds a secret with a custom title of "My Title":

Google Authenticator with a Custom Title


Upgrading git for CVE 2017-1000117

August 14, 2017  •  12:11 PM

A security vulnerability in Git has been announced: a bug in URL parsing can cause git clone to execute arbitrary commands. These URLs look quite suspicious, so it's unlikely that you'd be convinced through social engineering to clone them yourself. But they can be hidden in repository submodules.

Unless you're a Continuous Integration build agent, I hope that it's quite uncommon that you git clone --recursive a repository that you do not trust. So this vulnerability is rather uncommon, but as with any security vulnerability that has the possibility of remote code execution, you should upgrade your Git clients immediately.

Git version 2.14.1 is the latest and greatest version of Git, and has been patched. But most people don't actually build from source, so your version of Git is probably provided to you by a distribution. You may have different versions available to you - ones that have had the patches applied by your vendor - so you may not be able to determine if you're vulnerable simply by looking at the version number.

Here's some simple steps to determine whether you're vulnerable and some upgrade instructions if you are.

Are you vulnerable?

You can easily (and safely) check to see if your version of Git is vulnerable to this recent security vulnerable. Run this from a command prompt:

git clone -q ssh://-q/ /tmp/gittest

Note: this will not actually clone any repositories to your system, and it will not execute any dangerous commands.

If you see:

fatal: strange hostname '-q' blocked

Congratulations - you are already running a version of Git that is not vulnerable.

If, instead, you see:

fatal: Could not read from remote repository.

Please make sure you have the correct access rights
and the repository exists.

Then your version of Git is vulnerable and you should upgrade immediately.


Windows is quite easy to upgrade. Simply grab the newest version of Git for Windows (version 2.14.1) from https://git-for-windows.github.io/.


Apple ships Git with Xcode but unfortunately, they do not update it regularly, even for security vulnerabilities. As a result, you'll need to upgrade to the version that is included by a 3rd party. Homebrew is the preferred package manager for macOS.

  1. If you have not yet installed Homebrew, you can install it by running:

    /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

    at a command prompt.

  2. After that, you can use Homebrew to install git:

    brew install git
  3. Add the Homebrew install location (/usr/local) to your PATH.

    echo 'export PATH="/usr/local/bin:$PATH"' >> ~/.bashrc
  4. Close all open Terminal sessions, quit Terminal.app, and re-open it.

Linux (Debian, Ubuntu)

If you're using the current version of Ubuntu or Debian, then they'll have the latest version ready. If you're on a stable system, like a server, you should be running an LTS release - a "long term support" version - where they backport security patches like this one. So you should simply need to:

  1. Get the latest information about the available software versions from the remote repository:

    Debian, Ubuntu:

    sudo apt-get update

    Red Hat, CentOS:

    sudo yum update
  2. Install the latest version of git:

    Debian, Ubuntu:

    sudo apt-get install git

    Red Hat, CentOS:

    sudo yum update git

Ensuring that you're patched

Now if you run:

git clone -q ssh://-q/ /tmp/gittest

at a command prompt, then you should see:

fatal: strange hostname '-q' blocked

And now you're patched against the git security vulnerability, CVE 2017-1000117.

(Re)introducing git-dad

June 18, 2017  •  8:18 PM

After I dropped the git-recover script, @MordodeMaru asked me on Twitter if we could have a git dad command to help you out when you're in a jam:

But when I started thinking about my stepdad and his banter with his coworkers, I thought that if a git dad command was really going to help you out, it would help you when you mistyped the git add command. It could bring you a bit of levity… a dad joke!

And what better day than Father's Day to make that happen:

Now when you mistype git add as git dad, it will still add your file to the index, but it will also give you the prize of a dad joke.

All you have to do is grab git-dad and put it in your PATH.

On Dad Jokes and Calculus

I'd love to claim credit for this wonderful addition to the Git ecosystem, but just as I was getting ready to publish this, I did a quick search for "git dad" and I realized that Tim Petterson had already come up with the idea.

And, honestly, I would like to claim that I just happened to have the same idea. That this was totally independent discovery, like Calculus (and almost as important a contribution to humanity). But the truth is that I probably heard him talking about it. Perhaps it was in his awesome talk at Git Merge about aliases this year. Anyway, I'm sure that somewhere I got the idea from him and it stuck in my head, lying dormant until it was resurrected on Twitter.

But why would we need a second version of git dad? Surely one is enough.

You'll notice that this solution is a bit different than his solution, though. If you have an alias that starts with a bang (!) it will execute a non-Git command. (Normally, a Git alias just invokes another Git command; starting your alias with a ! allows you to invoke any command.)

But, if you have an alias that runs a non-Git command, then the alias will only be executed from the root of the repository's working directory. So an alias for !git add will work from the root of the repository's working directory, but not if you're inside some folder beneath that.

Using a script instead of an alias will solve this problem.

I did like his idea of using icanhazdadjoke.com instead of hardcoding some dad jokes. It's a bit slower than if they were hardcoded, but let's face it, that extra time spent is totally worth it to have a fresh, neverending supply.

Happy father's day!

Introducing git-recover

June 15, 2017  •  2:04 PM

I'm old enough to remember the old Norton UNERASE command: it was part of the old Norton Utilities for MS-DOS. It made clever use of the FAT filesystem to find files that were recently deleted, show them to you and let you undelete them.

git-recover brings that idea to your Git repositories.

Every time you add a version of a file to your git repository - that is to say, every time you run git add - Git will put a copy of that file in its object database. That means that if you accidentally delete a file that you were working on, if you ever ran git add on it, you can probably recover it.

Tell me more

I thought you'd never ask!

(Wait, you didn't? If you really aren't interested in the nitty gritty of how Git manages the index, then I guess you can skip this section. But who isn't interested in that!?!)

Git's index is a "staging area" that will become the next commit. If you recall from my discussion about how Git works, a commit in Git is a snapshot of the entire repository at a single point in time. And the index is also a snapshot: it contains a list of all the files in the repository that will make up the next commit.

You can see this if you look at the index, and Git provides a tool to do just that: git ls-files --stage. When I've just cloned a repository:

% git clone /tmp/foo_repo .
% git ls-files --stage
100644 6af0abcdfc7822d5f87315af1bb3367484ee3c0c 0   foo.txt

And when I add a new file to this repository, I can inspect the index again, and will see the new file:

% git add bar.txt
% git ls-files --stage
100644 ce013625030ba8dba906f756967f9e9ca394464a 0   bar.txt
100644 6af0abcdfc7822d5f87315af1bb3367484ee3c0c 0   foo.txt

Note that the entry for bar.txt contains the object ID of the file. When you run git add, Git actually adds the file to its object database, and takes the resulting object ID (the SHA-1 hash of the file) and places that in the index.

You can see the file on disk - Git has added it to the repository as a loose object:

% ls -Flas .git/objects/ce/013625030ba8dba906f756967f9e9ca394464a
4 -r--r--r--  1 ethomson  staff  21 14 Jun 23:58 .git/objects/ce/013625030ba8dba906f756967f9e9ca394464a

So Git has prepared this new file for our commit. But what if we don't commit this file? What if, instead, we git rm it? Or if we make some more changes to bar.txt and add those instead?

% echo "different changes" > bar.txt
% git add bar.txt

Now we've overwritten our original changes to bar.txt:

% git ls-files --stage
100644 4a95512212b2f24397fe2df5a2554935bd0a032a 0   bar.txt
100644 6af0abcdfc7822d5f87315af1bb3367484ee3c0c 0   foo.txt

You can see that the object ID for bar.txt had changed - reflecting our new file. But what's happened to the original file we added? Where is object ce01362?

It's still in our object database:

% ls -Flas .git/objects/ce/013625030ba8dba906f756967f9e9ca394464a
4 -r--r--r--  1 ethomson  staff  21 14 Jun 23:58 .git/objects/ce/013625030ba8dba906f756967f9e9ca394464a

But we never committed it, so this object is not pointed to by any commit in the graph. Nor is it in our index anymore. This unreference blob is "garbage" and - eventually - Git will garbage collect it.

But until it does, we can recover it!

Using git-recover

The simplest way to use git-recover is to use it in interactive mode: just run git recover -i. It will show you the first few lines of each "orphaned" file - those that were once git added to the repository but were never committed - and let you recover them (or not).

% git recover -i
Recoverable orphaned git blobs:

61c2562a7b851b69596f0bcad1d8f54c400be977  (Thu 15 Jun 2017 12:20:22 CEST)
> Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod
> tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim
> veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea
> commodo consequat. Duis aute irure dolor in reprehenderit in voluptate

Recover this file? [y,n,v,f,q,?]: 

You can also run git-recover without any arguments, and it will show you all the "orphaned" blobs that you can recover. You can then inspect an object to decide if it's something that you're interested in (using git show).

% git recover
Recoverable orphaned git blobs:

61c2562a7b851b69596f0bcad1d8f54c400be977  Thu 15 Jun 2017 12:20:22 CEST

When you find the object that you want to recover, you can run git-recover <objectid> to pull it out of the object database and write it to disk.

You can specify the filename to write with the (optional) -f flag:

% git recover 61c2562 -f greeking.txt
Writing 61c2562: greeking.txt.

Specifying the filename is helpful, because you may have rules set up in your .gitattributes file on a per-file or per-file extension basis. Using the -f flag will make sure that these rules are executed.

How to get it

git-recover is a shell script - you can just download it and go.

When you put git-recover in your PATH, then it becomes a proper git command, and you can run git recover (notice the space instead of the dash).

Please open an issue or a pull request if you have problems or improvements.

Git Conditional Includes

June 6, 2017  •  1:29 PM

One of the features that slipped quietly into Git 2.13 is the notion of "conditional includes" for configuration files, and it's one of my favorite new features. Although it's really simple, it's also extremely important for people who work both on Enterprise software and open source projects.

When I work on my open source projects, I end up pushing them to GitHub - and I have to authenticate when I do, so GitHub knows that I'm ethomson@edwardthomson.com. And when I'm working at my day job, I push my repositories to Visual Studio Team Services - and I have to authenticate there, too. So VSTS knows that I'm ethomson@microsoft.com.

But since Git is a distributed version control system, there's no authentication when I'm working in my local repository. When I commit, my changes go into the repository on my computer. That means that I need to tell Git who I am - my name and my email address - and Git will dutifully record that information the commits as I create them.

The problem here is that now I'm on my own to manage the settings for my name and email address - and I want to make sure that I keep my day job separate from the work I do in open source. I don't want to use my work email address on my open source projects because I don't want anybody to run git log and think that it's a project that's sponsored by my employer. And if I ever change jobs, the email address in the repository's history will stop working.

More importantly, I don't want to use my personal email address in our corporate Git history. People tend to get a little jumpy when they see that somebody has checked in code and their email address doesn't end in microsoft.com. And although my personal address is pretty boring, I imagine this is especially important for people who's email addresses are a bit more "creative". Otherwise, you may show up at work on monday morning with a clever new nickname. (I'm talking to you, reeferman42@example.com.)

So I try to make sure to keep my professional work separate from my personal work.

Before Git 2.13 introduced conditional includes, I would have had to set up my corporate email address in my global Git configuration, and then remember to change it in any new repository where I want to use my personal email address. Unfortunately, it's really easy to forget to do this when all you want to do is clone a repository and quickly create a quick pull request. That means that it's easy to accidentally end up pushing a change with the wrong identity.

With conditional configuration includes, you have a lot more control over how your Git configuration is applied. You can set up some configuration to be applied based on the directory that you're in, so it's much easier to set up.

I keep all my libgit2-related repositories in one directory: C:\LibGit2. This is where libgit2, LibGit2Sharp, Rugged and the like live; all my other open source projects live in C:\Projects. Every other repository on my machine is work-related. So my global Git configuration (C:\Users\ethomson\.gitconfig on my Windows machine, or /Users/ethomson/.gitconfig on my Mac) looks like:

    name = Edward Thomson
    email = ethomson@microsoft.com
[includeIf "gitdir:C:/LibGit2/"]
    path = .gitconfig.oss
[includeIf "gitdir:C:/Projects/"]
    path = .gitconfig.oss

And I have a second file in my home directory, .gitconfig.oss that specifies only the configuration that I want to use when I'm hacking on open source projects, like libgit2:

    email = ethomson@edwardthomson.com

Note that the trailing slash is required, and on Windows, you need to specify your path using forward slashes - not backslashes. I make sure to keep these updated in my dotfiles repository, so that I won't forget them on a new machine.

Now when I'm working on a project that lives beneath C:\Projects, like my little hexdump utility that lives in C:\Projects\HexDump - or when I'm working on LibGit2Sharp, which lives in C:\LibGit2\LibGit2Sharp, then Git will set my identity as ethomson@edwardthomson.com.

For everything else - like the Visual Studio Team Services repository - it will use my work email address, ethomson@microsoft.com, ensuring that nobody at work gives me an embarassing nickname.

At least, if they do, it won't be because of my email address.