Tag: ci/cd

Signing MacOSX apps with Linux

Do you, like me, develop desktop applications for MacOSX? Do you, like me, do it on Linux because it makes for a much cheaper and easier to manage gitlab CI/CD build farm? Do you still sign your apps using a MacOSX machine, or worse (yes, like me), not sign them at all, leaving ugly popups like the one below?

With the impending trustpocalypse next month a lot of third-party (non-app-store) apps for MacOSX are going to start having deeper trust issues than they’ve had previously, no doubt meaning more, uglier popups than that one, or worse, not being able to run at all.

I suspect this trust-tightening issue, whilst arguably a relatively good thing to do to in the war against malware, will adversely affect a huge number of open-source Mac applications where the developer/s wish to provide Mac support for their users but may not wish to pay the annual Apple Developer tax even though it’s still relatively light, or may not even own any Apple hardware (though who knows how they do their integration testing?). In-particular this is likely to affect very many applications built with Electron or NWJS, into which group this post falls.

Well, this week I’ve been looking into this issue for one of the apps I look after, and I’m pleased to say it’s at a stage where I’m comfortable writing something about it. The limitation is that you don’t sidestep paying the Apple Developer tax, as you do still need valid certs with the Apple trust root. But you can sidestep paying for more Apple hardware than you need, i.e. nothing needed in the build farm.

First I should say all of the directions I used came from a 2016 article, here. Thanks very much to Allin Cottrell.

Below is the (slightly-edited) script now forming part of the build pipeline for my app. Hopefully the comments make it fairly self-explanatory. Before you say so, yes I’ve been lazy and haven’t parameterised directory and package names yet.

#!/bin/bash

#########
# This is a nwjs (node) project so fish the version out of package.json
#
VERSION=$(jq -r .version package.json)

#########
# set up the private key for signing, if present
#
rm -f key.pem
if [ "$APPLE_PRIVATE_KEY" != "" ]; then
    echo "$APPLE_PRIVATE_KEY" > key.pem
fi

#########
# temporary build folder/s for package construction
#
rm -rf build
mkdir build && cd build
mkdir -p flat/base.pkg flat/Resources/en.lproj
mkdir -p root/Applications;

#########
# stage the unsigned applicatio into the build folder
#
cp -pR "../dist/EPI2MEAgent/osx64/EPI2MEAgent.app" root/Applications/

#########
# fix a permissions issue which only manifests after following cpio stage
# nw.app seems to be built with owner-read only. no good when packaging as root
#
chmod go+r "root/Applications/EPI2MEAgent.app/Contents/Resources/app.nw"

#########
# pack the application payload
#
( cd root && find . | cpio -o --format odc --owner 0:80 | gzip -c ) > flat/base.pkg/Payload

#########
# calculate a few attributes
#
files=$(find root | wc -l)
bytes=$(du -b -s root | awk '{print $1}')
kbytes=$(( $bytes / 1000 ))

#########
# template the Installer PackageInfo
#
cat <<EOT > flat/base.pkg/PackageInfo
<pkg-info format-version="2" identifier="com.metrichor.agent.base.pkg" version="$VERSION" install-location="/" auth="root">
  <payload installKBytes="$kbytes" numberOfFiles="$files"/>
  <scripts>
    <postinstall file="./postinstall"/>
  </scripts>
  <bundle-version>
    <bundle id="com.metrichor.agent" CFBundleIdentifier="com.nw-builder.epimeagent" path="./Applications/EPI2MEAgent.app" CFBundleVersion="$VERSION"/>
  </bundle-version>
</pkg-info>
EOT

#########
# configure the optional post-install script with a popup dialog
#
mkdir -p scripts
cat <<EOT > scripts/postinstall
#!/bin/bash

osascript -e 'tell app "Finder" to activate'
osascript -e 'tell app "Finder" to display dialog "To get the most of EPI2ME please also explore the Nanopore Community https://community.nanoporetech.com/ ."'
EOT

chmod +x scripts/postinstall

#########
# pack the postinstall payload
#
( cd scripts && find . | cpio -o --format odc --owner 0:80 | gzip -c ) > flat/base.pkg/Scripts
mkbom -u 0 -g 80 root flat/base.pkg/Bom

#########
# Template the flat-package Distribution file together with a MacOS version check
#
cat <<EOT > flat/Distribution
<?xml version="1.0" encoding="utf-8"?>
<installer-script minSpecVersion="1.000000" authoringTool="com.apple.PackageMaker" authoringToolVersion="3.0.3" authoringToolBuild="174">
    <title>EPI2MEAgent $VERSION</title>
    <options customize="never" allow-external-scripts="no"/>
    <domains enable_anywhere="true"/>
    <installation-check script="pm_install_check();"/>
    <script>
function pm_install_check() {
  if(!(system.compareVersions(system.version.ProductVersion,'10.12') >= 0)) {
    my.result.title = 'Failure';
    my.result.message = 'You need at least Mac OS X 10.12 to install EPI2MEAgent.';
    my.result.type = 'Fatal';
    return false;
  }
  return true;
}
    </script>
    <choices-outline>
        <line choice="choice1"/>
    </choices-outline>
    <choice id="choice1" title="base">
        <pkg-ref id="com.metrichor.agent.base.pkg"/>
    </choice>
    <pkg-ref id="com.metrichor.agent.base.pkg" installKBytes="$kbytes" version="$VERSION" auth="Root">#base.pkg</pkg-ref>
</installer-script>
EOT

#########
# pack the Installer
#
( cd flat && xar --compression none -cf "../EPI2MEAgent $VERSION Installer.pkg" * )

#########
# check if we have a key for signing
#
if [ ! -f ../key.pem ]; then
    echo "not signing"
    exit
fi

#########
# calculate attribute
: | openssl dgst -sign ../key.pem -binary | wc -c > siglen.txt

#########
# xar the Installer package
#
xar --sign -f "EPI2MEAgent $VERSION Installer.pkg" \
    --digestinfo-to-sign digestinfo.dat --sig-size $(cat siglen.txt) \
    --cert-loc ../dist/tools/mac/certs/cert00 --cert-loc ../dist/tools/mac/certs/cert01 --cert-loc ../dist/tools/mac/certs/cert02

#########
# construct the signature
#
openssl rsautl -sign -inkey ../key.pem -in digestinfo.dat \
        -out signature.dat

#########
# add the signature to the installer
#
xar --inject-sig signature.dat -f "EPI2MEAgent $VERSION Installer.pkg"

#########
# clean up
#
rm -f signature.dat digestinfo.dat siglen.txt key.pem

With all that you still need a few assets. I built and published (internally) corresponding debs for xar v1.6.1 and bomutils 0.2. You might want to compile & install those from source – they’re pretty straightforward builds.

Next, you need a signing identity. I used XCode (Preferences => Accounts => Apple ID => Manage Certificates) to add a new Mac Installer Distribution certificate. Then used that to sign my .app once on MacOS in order to fish out the Apple cert chain (there are probably better ways to do this)

productsign --sign LJXXXXXX58 \
        build/EPI2MEAgent\ 2020.1.14\ Installer.pkg \
        EPI2MEAgent\ 2020.1.14\ Installer.pkg

Then fish out the certs

xar -f EPI2MEAgent\ 2020.1.14\ Installer.pkg \
        --extract-certs certs
mac:~/agent rmp$ ls -l certs/
total 24
-rw-r--r--  1 rmp  Users  1494 15 Jan 12:06 cert00
-rw-r--r--  1 rmp  Users  1062 15 Jan 12:06 cert01
-rw-r--r--  1 rmp  Users  1215 15 Jan 12:06 cert02

Next use Keychain to export the .p12 private key for the “3rd Party Mac Developer Installer” key. Then openssl it a bit to convert to a pem.

openssl pkcs12 -in certs.p12 -nodes | openssl rsa -out key.pem

I set this up the contents of key.pem as a gitlab CI/CD Environment Variable APPLE_PRIVATE_KEY so it’s never committed to the project source tree.

Once all that’s in place it should be possible to run the script (paths-permitting, obviously yours will be different) and end up with an installer looking something like this. Look for the closed padlock in the top-right, and the fully validated chain of certificate trust.

In conclusion, the cross-platform application nwjs builds (Mac, Windows, Linux) all run using nw-builder on ubuntu:18.04, and the Mac (and Windows, using osslsigncode, maybe more on that later) also all run on ubuntu:18.04. Meaning one docker image for the Linux-based Gitlab CI/CD build farm. Nice!

Proxy testing with IP Namespaces and GitLab CI/CD

CC-BY-NC https://www.flickr.com/photos/thomashawk/106559730

At work, I have a CLI tool I’ve been working on. It talks to the web and is used by customers all over the planet, some of them on networks with tighter restrictions than my own. Often those customers have an HTTP proxy of some sort and that means the CLI application needs to negotiate with it differently than it would directly with a web server.

So I need to test it somehow with a proxy environment. Installing a proxy service like Squid doesn’t sound like too big a deal but it needs to run in several configurations, at a very minimum these three:

  • no-proxy
  • authenticating HTTP proxy
  • non-authenticating HTTP proxy

I’m going to ignore HTTPS proxy for now as it’s not actually a common configuration for customers but I reckon it’s possible to do with mkcert or LetsEncrypt without too much work.

There are two other useful pieces of information to cover, firstly I use GitLab-CI to run the CI/CD test stages for the three proxy configurations in parallel. Secondly, and this is important, I must make sure that, once the test Squid proxy service is running, the web requests in the test only pass through the proxy and do not leak out of the GitLab runner. I can do this by using a really neat Linux feature called IP namespaces.

IP namespaces allow me to set up different network environments on the same machine, similar to IP subnets or AWS security groups. Then I can launch specific processes in those namespaces and network access from those processes will be limited by the configuration of the network namespace. That is to say, the Squid proxy can have full access but the test process can only talk to the proxy. Cool, right?

The GitLab CI/CD YAML looks like this (edited to protect the innocent)

stages:
 - integration
 
.integration_common: &integration_common |
   apt-get update
   apt-get install -y iproute2

.network_ns: &network_ns |
   ip netns add $namespace
   ip link add v-eth1 type veth peer name v-peer1
   ip link set v-peer1 netns $namespace
   ip addr add 192.168.254.1/30 dev v-eth1
   ip link set v-eth1 up
   ip netns exec $namespace ip addr add 192.168.254.2/30 dev v-peer1
   ip netns exec $namespace ip link set v-peer1 up
   ip netns exec $namespace ip link set lo up
   ip netns exec $namespace ip route add default via 192.168.254.1

noproxynoauth-cli:
   image: ubuntu:18.04
   stage: integration
   script:
     - *integration_common
     - test/end2end/cli

proxyauth-cli:
   image: ubuntu:18.04
   stage: integration
   script:
     - *integration_common
     - apt-get install -y squid apache2-utils
     - mkdir -p /etc/squid3
     - htpasswd -cb /etc/squid3/passwords testuser testpass
     - *network_ns
     - squid3 -f test/end2end/conf/squid.conf.auth && sleep 1 || tail -20 /var/log/syslog | grep squid
     - http_proxy=http://testuser:testpass@192.168.254.1:3128/ https_proxy=http://testuser:testpass@192.168.254.1:3128/ ip netns exec $namespace test/end2end/cli
     - ip netns del $namespace || true
   variables:
     namespace: proxyauth

proxynoauth-cli:
   image: ubuntu:18.04
   stage: integration
   script:
     - *integration_common
     - apt-get install -y squid
     - *network_ns
     - squid3 -f test/end2end/conf/squid.conf.noauth && sleep 1 || tail -20 /var/log/syslog | grep squid
     - http_proxy=http://192.168.254.1:3128/ https_proxy=http://192.168.254.1:3128/ test/end2end/cli
     - ip netns del $namespace || true
   variables:
     namespace: proxynoauth

So there are five blocks here, with three stages and two common script blocks. The first common script block installs iproute2 which gives us the ip command.

The second script block is where the magic happens. It configures a virtual, routed subnet in the parameterised $namespace.

Following that we have the three test stages corresponding to the three proxy (or not) configurations I listed earlier. Two of them install Squid, one of those creates a test user for authenticating with the proxy. They all run the test script, which in this case is test/end2end/cli. When those three configs are modularised and out like this with the common net namespace script as well it provides a good deal of clarity to the test maintainer. I like it a lot.

So then the last remaining things are the respective squid configurations: proxyauth and proxynoauth. There’s a little bit more junk in these than there needs to be as they’re taken from the stock examples, but they look something like this:

 visible_hostname proxynoauth
 acl localnet src 10.0.0.0/8 # RFC1918 possible internal network
 acl localnet src 172.16.0.0/12 # RFC1918 possible internal network
 acl localnet src 192.168.0.0/16 # RFC1918 possible internal network
 acl SSL_ports port 443
 acl Safe_ports port 80 # http
 acl Safe_ports port 443 # https
 acl CONNECT method CONNECT
 http_access deny !Safe_ports
 http_access deny CONNECT !SSL_ports
 http_access allow localhost manager
 http_access deny manager
 http_access allow localnet
 http_access allow localhost
 http_access deny all
 http_port 3128

and for authentication:

 visible_hostname proxyauth
 acl localnet src 10.0.0.0/8 # RFC1918 possible internal network
 acl localnet src 172.16.0.0/12 # RFC1918 possible internal network
 acl localnet src 192.168.0.0/16 # RFC1918 possible internal network
 acl SSL_ports port 443
 acl Safe_ports port 80 # http
 acl Safe_ports port 443 # https
 acl CONNECT method CONNECT
 http_access deny !Safe_ports
 http_access deny CONNECT !SSL_ports
 http_access allow localhost manager
 http_access deny manager
 
 auth_param basic program /usr/lib/squid3/basic_ncsa_auth /etc/squid3/passwords
 auth_param basic realm proxy
 acl authenticated proxy_auth REQUIRED

 http_access allow authenticated
 http_access deny all
 http_port 3128

And there you have it – network-restricted proxy testing with different proxy configurations. It’s the first time I’ve used ip net ns without being wrapped up in Docker, LXC, containerd or some other libvirt thing, but the feeling of power from my new-found network-god skills is quite something :)

Be aware that you might need to choose different subnet ranges if your regular LAN conflicts. Please let me know in the comments if you find this useful or if you had to modify things to work in your environment.