Using Recursion on Golang To Solve A Problem!

Summary

I was writing an SSL Certificate Decoder ( https://tools.woohoosvcs.com/certdecoder/ ) and came across an interesting problem. I wanted to validate a certificate chain but had some difficulty figuring it out. The problem was that I needed to find the most child certificate in a seemingly random list of certificates. What I am calling the most child certificate is the one at the bottom of the chain that is issued but does not issue one of its own. Users can provide the cert chain in any order so we cannot relay on the order they are submitted for this tool.

SSL Certificate Chains

To understand the problem, one needs to understand an SSL Certificate Chain. When traversing a chain, each certificate is an individual certificate. How they are linked is by their issuer. There is a subject section about the certificate itself and then an issuer section which indicates the certificate authority certificate that issued that. It goes all the way up until you hit a root certificate. The root certificates sign themselves.

Here you can see the cert chain for this site. The SAN or alternate names that allow tools.woohoosvcs.com be valid are further down below this screenshot.

SSL Certificate Chain
SSL Certificate Chain

Recursion

It had been a few decades since I sat in a Computer Science class but I do remember a few control structures. One of which was recursion. Recursion can be a difficult concept to grasp and an even more difficult concept to implement. Golang has an example of this here. At a high level a recursive function is one that calls itself. The call stack gets nested in each layer. In the case that you do not know how many certificates and iterations you need to go through to find the chain, recursive functions help with this. Alternatively you would have to use multiple layers of “for” loops to iterate through, matching the current cert against one that may be a child. In a past life I may have implemented a few layers of “for” loops to statically account for the most common scenarios.

Recursion can be tricky and if you do not use it right, it can involve stack overflows if the function indefinitely calls it self. The key to a good recursive function is a way out. It needs to be able to exit without continuing to call itself forever. There are also limits based on the platform as to how deep you can go in this recursion. Your platform will have a limit at which point an exception or error is thrown.

For a much better description of recursion you can read the Wikipedia article here – https://en.wikipedia.org/wiki/Recursion_(computer_science)

The Code

It is still a work in progress but after iterations of playing with linked lists, multiple for loops, this is what I landed on.

func findChildOfCert(certs []x509.Certificate, cert x509.Certificate) x509.Certificate {
	if len(certs) <= 1 {
		return cert
	}
	result := cert
	for _, item := range certs {
		if item.Issuer.CommonName == cert.Subject.CommonName {
			return findChildOfCert(certs, item)
		}
	}

	return result
}

It is kicked off in a parent function via

childCert := findChildOfCert(cs, cs[0])

Where cs is a slice (Golang speak for array) of certificates much like “certs” in the recursive function. We pass it the list of certificates and the very first one.

On the first call it checks to see if this certificate issued any others in the list. If so, it calls the function again with the issued certificate( more child than the current ) and does the process over again.

When it cannot find a certificate issued by the currently iterated certificate (most child record), the for loop exits and it simply passes the original cert that the function was called with. At that point, the stack completely unwinds, passing the “answer” certificate all the way down. That result is provided to the childCert variable.

Validating the Cert

Golang provides a few options for validating the cert. Once you have the most child certificate, you can do something like the below.

for i := range cs {
	if !cs[i].Equal(&childCert) {
		roots.AddCert(&cs[i])
	}
}

opts := x509.VerifyOptions{
	Intermediates: roots,
}

opts2 := x509.VerifyOptions{
	Roots: roots,
}

if _, err := childCert.Verify(opts); err != nil {
	status = append(status, "Not Trusted By Root - failed to verify certificate: "+err.Error())
} else {
	status = append(status, "Trusted Chain")
}

if _, err := childCert.Verify(opts2); err != nil {
	status = append(status, "Not Valid(contiguous) - failed to verify certificate: "+err.Error())
} else {
	status = append(status, "Valid Chain")
}

I load up a “roots” slice of the roots provided. I also exclude the child certificate from this. From there I perform two validations. One that the chain is trusted, meaning it rolls up to one that is trusted by the source used. The other validation is that the chain is valid. Is there continuity in the chain or is it broken. A chain can be valid but un trusted. Knowing the difference may help you in a rare case.

Stack Overflow

I actually found a stack overflow doing a regression test with a self signed certificate. The code above actually ended up comparing the certificate to itself over and over again and trying to keep going down the rabit hole. It ended up with the following

runtime: goroutine stack exceeds 1000000000-byte limit
fatal error: stack overflow

runtime stack:
runtime.throw(0x1332651, 0xe)
	/usr/local/go/src/runtime/panic.go:774 +0x72
runtime.newstack()
	/usr/local/go/src/runtime/stack.go:1046 +0x6e9
runtime.morestack()
	/usr/local/go/src/runtime/asm_amd64.s:449 +0x8f

goroutine 24 [running]:
tools.woohoosvcs.com/certdecoder.findChildOfCert(0xc0000d6b00, 0x1, 0x1, 0xc0000b8800, 0x3da, 0x3ea, 0xc0000b8804, 0x2c2, 0x3e6, 0xc0000b89a0, ...)

tools.woohoosvcs.com/certdecoder/certdecoder.go:180 +0x1d0 fp=0xc020114ef8 sp=0xc020114ef0 pc=0x128a210
tools.woohoosvcs.com/certdecoder.findChildOfCert(0xc0000d6b00, 0x1, 0x1, 0xc0000b8800, 0x3da, 0x3ea, 0xc0000b8804, 0x2c2, 0x3e6, 0xc0000b89a0, ...)

tools.woohoosvcs.com/certdecoder/certdecoder.go:184 +0x1a3 fp=0xc020116920 sp=0xc020114ef8 pc=0x128a1e3
tools.woohoosvcs.com/certdecoder.findChildOfCert(0xc0000d6b00, 0x1, 0x1, 0xc0000b8800, 0x3da, 0x3ea, 0xc0000b8804, 0x2c2, 0x3e6, 0xc0000b89a0, ...)

tools.woohoosvcs.com/certdecoder/certdecoder.go:184 +0x1a3 fp=0xc020118348 sp=0xc020116920 pc=0x128a1e3

Luckily my unit testing caught this and this would have never gone to production due to that unit testing. If you’re not sure what unit testing is check out my article Unit Testing Golang on App Engine. The fix was simple though in that to make sure my recursive function doesn’t compare a cert to itself using the .IsEquals() function

Final Words

In this we walked through a useful use case for recursion and the background of the technology that needed it. I even provided some of the errors that happen when you fail to use recursion properly as I accidentally did!

Unit Testing Golang on App Engine

Summary

If you have landed here, it is most likely because you attempted to implement the subject and ran into errors. In my use case, I ran into errors when splitting up my Golang app into packages and added unit testing to it. To compound it, I was automatically deploying via triggers with CloudBuild.

This builds upon the article IP Subnet Calculator on Google App Engine. Reading that article will help give you a good reference point for this article.

This article is not a “How To” on actual unit testing but getting it to work under this unique combination of tools. A great unit testing article I found is here – https://blog.alexellis.io/golang-writing-unit-tests/

How To Setup

Before we talk about what might have gone wrong, let’s talk about how to set this up properly. There has been a lot of confusion since Golang went to version 1.11 and started supporting modules. After supporting modules it appears using GOPATH is a less of a supported method. With that said some tools still look for it.

Directory Structure

To get a final state visual of my directory structure, it ended up as follows

tools.woohoosvcs.com
tools.woohoosvcs.com/app.yaml
tools.woohoosvcs.com/go.mod
tools.woohoosvcs.com/subnetcalculator
tools.woohoosvcs.com/subnetcalculator/subnetcalculator.go
tools.woohoosvcs.com/subnetcalculator/subnetcalculator_test.go
tools.woohoosvcs.com/cloudbuild.yaml
tools.woohoosvcs.com/html
tools.woohoosvcs.com/html/root.html
tools.woohoosvcs.com/html/view.html
tools.woohoosvcs.com/README.md
tools.woohoosvcs.com/static
tools.woohoosvcs.com/static/submitdata.js
tools.woohoosvcs.com/static/subnetcalculator.css
tools.woohoosvcs.com/root
tools.woohoosvcs.com/root/root.go
tools.woohoosvcs.com/main.go

Modules

Modules help with the dependency needs of Golang applications. Previously, private packages were difficult to manage as well as specific versioned public packages. Modules helps many of these issues. Here is a good article on modules – https://blog.golang.org/using-go-modules

The key to my issues seemed to be involving modules. Google App Engine’s migrating to 1.11 guide recommends the following.

The preferred method is to manually move all http.HandleFunc() calls from your packages to your main()function in your main package.

Alternatively, import your application’s packages into your main package, ensuring each init() function that contains calls to http.HandleFunc() gets run on startup.

https://cloud.google.com/appengine/docs/standard/go111/go-differences

My app’s directory is “tools.woohoosvcs.com” so I named the root, the same. Run “go mod init”

$ go mod init tools.woohoosvcs.com
go: creating new go.mod: module tools.woohoosvcs.com
dwcjr@Davids-MacBook-Pro tools.woohoosvcs.com % cat go.mod
module tools.woohoosvcs.com

go 1.13

Importing Private Packages

This then lets us to refer to packages and modules under it via something similar.

import ( "tools.woohoosvcs.com/subnetcalculator" )

CloudBuild

My cloudbuild.yaml ended up as follows

steps:
  - id: subnetcalculator_test
    name: "gcr.io/cloud-builders/go"
    args: ["test","tools.woohoosvcs.com/subnetcalculator"]
    #We use modules but this docker wants GOPATH set and they are not compatible.
    env: ["GOPATH=/fakepath"]
  - name: "gcr.io/cloud-builders/gcloud"
    args: ["app", "deploy"]

Interesting tidbit is the “name” is a docker image from Google Cloud Repository or gcr.io

The first id runs the “go” image and runs “go test tools.woohoosvcs.com/subnetcalculator”. It seems the go image wants GOPATH but go test fails with it set so I had to set it to something fake.

It then uses the gcloud deployer which consumes app.yaml

How I Got Here

Unit tests drove me to wanting to split out logic, particularly the subnetcalculator (check it out via https://tools.woohoosvcs.com/subnetcalculator )

Before implementing modules I could get it to run locally by importing ./subnetcalculator but then “gcloud app deploy” would fail.

2019/11/22 01:39:13 Failed to build app: Your app is not on your GOPATH, please move it there and try again.
building app with command '[go build -o /tmp/staging/usr/local/bin/start ./...]', env '[PATH=/go/bin:/usr/local/go/bin:/builder/google-cloud-sdk/bin/:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin HOSTNAME=ace17fcba136 HOME=/builder/home BUILDER_OUTPUT=/builder/outputs DEBIAN_FRONTEND=noninteractive GOROOT=/usr/local/go/ GOPATH=/go GOPATH=/tmp/staging/srv/gopath]': err=exit status 1, out=go build: cannot write multiple packages to non-directory /tmp/staging/usr/local/bin/start.

The error was vague but I noticed it was related to the import path. I tried moving the folder to $GOPATH/src and it seemed to deploy via “gcloud app deploy” but then failed via CloudBuild automated trigger.

------------------------------------ STDERR ------------------------------------
2019/11/21 21:31:39 staging for go1.13
2019/11/21 21:31:39 GO111MODULE=auto, but no go.mod found, so building with dependencies from GOPATH
2019/11/21 21:31:39 Staging second-gen Standard app (GOPATH mode): failed analyzing /workspace: cannot find package "tools.woohoosvcs.com/subnetcalculator" in any of:
	($GOROOT not set)
	/builder/home/go/src/tools.woohoosvcs.com/subnetcalculator (from $GOPATH)
GOPATH: /builder/home/go
--------------------------------------------------------------------------------

ERROR
ERROR: build step 0 "gcr.io/cloud-builders/gcloud" failed: exit status 1

It was like a balancing act with a 3 legged chair! Once I initialized the module though and adjusted the imports it worked great

Final Worlds

Here we worked through a best practice of using modules and an internal package to do automated build deploy on Google App Engine. Unit testing with automated deploys are important so that broken builds do not get pushed to production.

IP Subnet Calculator on Google App Engine

Summary

Over the past few days I put a few hours towards writing a web app to perform some basic CIDR subnet calculations. I wanted to both share the link to it ( https://tools.woohoosvcs.com/subnetcalculator/ ). I also wanted to walk through how I deployed it to Google App Engine.

What is a Subnet Calculator?

When provisioning subnets you typically need to determine your requirements. Determining the size of the subnet based on the number of hosts or what the mask will be can take a little bit of time. This is particularly so when you do not calculate this often. Subnet calculators help save some time on this.

Why did I do it?

I have written a few over the years. My early ones were written in C. Golang has a few libraries though in the new package to help with this. I had to use one the other day and figured why not quite my own. While doing this, why not share how to deploy your own app?

Deploying Golang on Google App Engine

In this, we assume you already have the Google SDK. We walked through this in Hello World From Google App Engine Via PHP.

As with any App Engine app, we need an app.yaml. Mine looks like this

runtime: go112

service: tools

handlers:
 - url: /view\.html
   static_files: view.html
   upload: view.html

I have a main.go file and a static view.html file. Since I already have a default service I am calling this one service. That requires I use a dispatch for my custom domain name.

Don’t forget to add the custom domain to the Settings section!

dispatch:
- url: "tools.woohoosvcs.com/*"
  service: tools

The easiest way to get a Golang app running is as follow. The Google App Engine will run the main function in the main package. Google App Engine sets the PORT environment variable.

package main

func main() {
        port := os.Getenv("PORT")
        if port == "" {
                port = "8080"
                log.Printf("Defaulting to port %s", port)
        }

        http.HandleFunc("/subnetcalculator/", viewSubnetCalculator)

        log.Printf("Listening on port %s", port)
        if err := http.ListenAndServe(":"+port, nil); err != nil {
                log.Fatal(err)
        }
}

// This is the main page to view the Subnet Calculator
func viewSubnetCalculator(w http.ResponseWriter, r *http.Request) {
        http.ServeFile(w, r, "./view.html")
}

The best practice for this would dictate a separate main.go importing your own package but if you were familiar enough with that, you may not need this article.

It is out of scope for this article but “http.HandleFunc(“/subnetcalculator/”, viewSubnetCalculator)” tells Golang that the “/subnetcalculator/” URI be directed to the “viewSubnetCalculator” function. That function simply displays the view.html file

From there we simply run the following. I have tools.woohoosvcs.com under the same parent as dispatch.yaml

cd tools.woohoosvcs.com 
gcloud app deploy

cd ..
gcloud app deploy dispatch/dispatch.yaml

Securing Google App Engine

If you use a Web Application Firewall like Cloudflare, don’t forget to ACL GAE to only allow connections from it. Otherwise, it can be completely bypassed

Google App Engine - Firewall
Google App Engine – Firewall

Final Words

This was just a quick and dirty deploy of a golang app to Google App Engine. As with all the apps though, I run them through Cloudflare or at least some sort of Web Application Firewall.

Samsung Galaxy S9 Moisture Detected!

Summary

I was watching cartoons with my 4 year old when I looked down at my phone and all of a sudden saw a water droplet icon. I dragged it town to read a message that indicated moisture was detected in my charging port. This couldn’t be, because I keep my phone nowhere near water. I had no accidents with it that could cause this, even though it is IP68 certified.

What is moisture detection?

S9 - Moisture Droplet
S9 – Moisture Droplet

IFIXIT has a great article on this. When reading up on this, that was the first link I clicked on. If you actually had your phone near water (which is perfectly acceptable on IP68), it has some great steps on rectifying it. In short, the phone has a sensor that can detect water/moisture and will disable the USB port while this detection is active. It helps to avoid a short or damage to the connectors. Apparently it detects this by monitoring resistance between pins.

Other issues like lint or other material could cause similar issues. The key to troubleshooting this is realizing it could be more than just water or moisture.

Other Things To Try

When I ran into this, I ran through the reboot and the hard reset without luck. I blew out with compressed air to no avail. When I did clean out my USB-C port with a toothpick I did find some debris but clearing this did not resolve the issue immediately.

What finally did it for me, which I was slightly surprised it worked was to power off the phone, plug it in and let it fully charge and then power on. I have a feeling the data for the port was cached and it just needed to clear after clearing debris out.

Here are a list of things to try, in this order.

  • Power off phone
  • Try to clean out with tissue and tooth picks
  • Reboot phone
  • Hard reset phone (power button and lower volume)
  • Blow out with compressed air but make sure none of the liquid comes out.
  • Power off and charge to 100% and then power back on
  • Go to Application Settings / USBSettings and clear cache and data and reboot
  • Factory Reset – Option of last resort because nobody likes doing this

Workarounds

If you are absolutely sure you do not have moisture or debris in there but need to charge, here are some options

  • Wireless charge ( this still works )
  • Power off the phone and try to charge powered off (if there is moisture in there this will potentially damage phone)

Sources

Here are a list of URLs I came across in my search. They may also help you further

Final Words

I hope you find this article before you went ahead and factory reset or set in for repair. With that said, even if you cannot get this error to go away, wireless charging still works. You will just be mostly unable to charge via wired. On some phones you can still charge when the phone is powered off.

Virtual Machine Orchestration on GCE

Summary

In this article we tackle VM orchestration. We I touched on in other articles, the desire is to dynamically spin up VMs as necessary. Some of the constructions in Google Cloud that are used are instance templates, instance groups, load balancers, health checks, salt (both state and reactor).

First Things First

In order to dynamically spin up VMs we need an instance group. For an instance group to work dynamically we need an instance template.

Instance Template

For this instance template, I will name it web-test. The name for this is important but we’ll touch on that later on.

GCE - Instance Template - Name
GCE – Instance Template – Name
GCE - Instance Template - CentOS
GCE – Instance Template – CentOS

For this demonstration we used CentOS 8. It can be any OS but our Salt state is tuned for CentOS.

GCE - Automation
GCE – Automation

As we touched on in the Cloud-init on Google Compute Engine article, we need to automate the provisioning and configuration on this. Since Google’s CentOS image does not come with this we use the startup script to load it. Once loaded and booted, cloud-init configures the local machine as a salt-minion and points it to the master.

Startup Script below

#!/bin/bash

if ! type cloud-init > /dev/null 2>&1 ; then
  # Log startup of script
  echo "Ran - `date`" >> /root/startup
  sleep 30
  yum install -y cloud-init

  if [ $? == 0 ]; then
    echo "Ran - yum success - `date`" >> /root/startup
    systemctl enable cloud-init
    # Sometimes GCE metadata URI is inaccessible after the boot so start this up and give it a minute
    systemctl start cloud-init
    sleep 10
  else
    echo "Ran - yum fail - `date`" >> /root/startup
  fi

  # Reboot either way
  reboot
fi

cloud-init.yaml below

#cloud-config

yum_repos:
    salt-py3-latest:
        baseurl: https://repo.saltstack.com/py3/redhat/$releasever/$basearch/latest
        name: SaltStack Latest Release Channel Python 3 for RHEL/Centos $releasever
        enabled: true
        gpgcheck: true
        gpgkey: https://repo.saltstack.com/py3/redhat/$releasever/$basearch/latest/SALTSTACK-GPG-KEY.pub

salt_minion:
    pkg_name: 'salt-minion'
    service_name: 'salt-minion'
    config_dir: '/etc/salt'
    conf:
        master: saltmaster263.us-central1-c.c.woohoo-blog-2414.internal
    grains:
        role:
            - web
GCE – Instance Template – Network tags – allow-health-checks

The network tag itself does not do anything at this point. Later on we will tie this into a firewall ACL to allow the Google health checks to pass.

Now we have an instance template. From our Intro to Salt Stack article we should have a salt server.

SaltStack Server

We have a state file here to provision the state but from our exposure we need salt to automagically do a few things.

Salt is a fairly complex setup so I have provided some of the files at the very bottom. I did borrow many ideas from this page of SaltStack’s documentation – https://docs.saltstack.com/en/latest/topics/tutorials/states_pt4.html

The first thing is to accept new minions as this is usually manual. We then need it to apply a state. Please keep in mind there are security implications of auto accepting. These scripts do not take that into consideration as they are just a baseline to get this working.

In order to have these automatically work, we need to use Salt reactor which listens to events and acts on them. Our reactor file looks like this. We could add some validation, particularly on the accept such as validating the minion name has web in it to push the wordpress state.

{# test server is sending new key -- accept this key #}
{% if 'act' in data and data['act'] == 'pend' %}
minion_add:
  wheel.key.accept:
  - match: {{ data['id'] }}
{% endif %}
{% if data['act'] == 'accept' %}
initial_load:
  local.state.sls:
    - tgt: {{ data['id'] }}
    - arg:
      - wordpress
{% endif %}

This is fairly simple. When a minion authenticates for the first time, acknowledge it and then apply the wordpress state we worked on in our articicle on Salt State. Since we may have multiple and rotating servers that spin up and down we will use Google’s Load Balancer to point Cloudflare to.

Cloudflare does offer load balancing but for the integration we want, its easier to use Google. The load balancer does require an instance group so we need to set that up first.

Instance Groups

Instance groups are one of the constructions you can point a load balancer towards. Google has two types of instance groups. Managed, which it will auto scale based on health checks. There is also managed which you have to manually add VMs to. We will choose managed

GCE - New Managed Instance
GCE – New Managed Instance

This name is not too important so it can be any one you like.

GCE - Instance Group
GCE – Instance Group

Here we set the port name and number, an instance template. For this lab we disabled autoscaling but in the real world this is why you want to set all of this up.

Instance Group - Health Check
Instance Group – Health Check

The HealthCheck expects to receive an HTTP 200 message for all clear. It is much better than a TCP check as it can validate the web server is actually responding with expected content. Since WordPress sends a 301 to redirect, we do have to set the Host HTTP Header here, otherwise the check will fail. Other load balancers only fail on 400-599 but Google does expect only a HTTP 200 per their document – https://cloud.google.com/load-balancing/docs/health-check-concepts

Instance Group Provisioning
Instance Group Provisioning

And here you can see it is provisioning! While it does that, let’s move over to the load balancer.

Firewall Rules

The health checks for the load balancer come from a set range of Google IPs that we need to allow. We can allow these subnets via network tags. Per Google’s Healthcheck document, the HTTP checks come from two ranges.

VPC - Allow Health Checks!
VPC – Allow Health Checks!

Here we only allow the health checks from the Google identified IP ranges to machines that are tagged with “allow-health-checks” to port 443.

Google Load Balancer

Initial

This is a crash course into load balancers if you have never set them up before. It is expected you have some understanding of front end, back end and health checks. In the VPC section we need to allow these

Google Load Balancer - Start configuration
Google Load Balancer – Start configuration
Google Load Balancer - Internet
Google Load Balancer – Internet

Back End Configuration

Google’s load balancers can be used for internal only or external to internal. We want to load balance external connections.

Google Load Balancer - Back End Create
Google Load Balancer – Back End Create

We will need to create a back end endpoint.

Luckily this is simple. We point it at a few objects we already created and set session affinity so that traffic is persistent to a single web server. We do not want it hopping between servers as it may confuse the web services.

Front End Configuration

Health Check Validation

Give the load balancer provisioning a few minutes to spin up. It should then show up healthy if all is well. This never comes up the first time. Not even in a lab!

Google Load Balancer - Healthy!
Google Load Balancer – Healthy!

Troubleshooting

The important part is to walk through the process from beginning to end when something does not work. Here’s a quick run through.

  • On provisioning, is the instance group provisioning the VM?
  • What is the status of cloud-init?
  • Is salt-minion installing on the VM and starting?
  • Does the salt-master see the minion?
  • Reapply the state and check for errors
  • Does the load balancer see health?

Final Words

If it does come up healthy, the last step is to point your DNS at the load balancer public IP and be on your way!

Since Salt is such a complex beast, I have provided most of the framework and configs here – Some of the more sensitive files are truncated but left so that you know they exist. The standard disclaimer applies in that I cannot guarantee the outcome of these files on your system or that they are best practices from a security standpoint.

Cloud-Init on Google Compute Engine

Summary

Yesterday I was playing a bit with Google Load Balancers and they tend to work best when you connect them to an automated instance group. I may touch on that in another article but in short that requires some level of automation. In an instance group, it will attempt to spin up images automatically. Based on health checks, it will introduce them to the load balanced cluster.

The Problem?

How do we automate provisioning? I have been touching on SaltStack in a few articles . Salt is great for configuration management but in an automated fashion, how do you get Salt on there? This was my goal. To get Salt Installed on a newly provisioned VM.

Method

Cloud-init is a very widely known method of provisioning a machine. From my brief understanding it started with Ubuntu and then took off. In Spinning Up Rancher With Kubernetes, I was briefly exposed to it. It makes sense and is widely supported. The concept it simple. Have a one time provisioning of the server.

Google Compute Engine

Google Compute Engine or GCE does support pushing cloud-init configuration (cloud-config) using metadata. You can set the “user-data” field and if cloud-init is installed it will be able to find this.

The problem is the only image that seems to support this out of the box is Ubuntu and my current preferred platform is CentOS although this is starting to change.

Startup Scripts

So if we don’t have cloud-init, what can we do? Google does have the functionality for startup and shutdown scripts via “startup-script” and “shutdown-script” meta fields. I do not want a script that runs every time. I also do not want to re-invent the wheel writing a failsafe script that will push salt-minion out and reconfigure it. For this reason I came up with a one time startup script.

The Solution

Startup Script

This is the startup script I came up with.

#!/bin/bash

if ! type cloud-init > /dev/null 2>&1 ; then
  echo "Ran - `date`" >> /root/startup
  sleep 30
  yum install -y cloud-init

  if [ $? == 0 ]; then
    echo "Ran - Success - `date`" >> /root/startup
    systemctl enable cloud-init
    #systemctl start cloud-init
  else
    echo "Ran - Fail - `date`" >> /root/startup
  fi

  # Reboot either way
  reboot
fi

This script checks to see if cloud-init exists. If it does, move along and don’t waste cpu. If it does not, we wait 30 seconds and install it. Upon success, we enable and either way we reboot.

Workaround

I played with this for a good part of a day, trying to get it working. Without the wait and other logging logic in the script, the following would happen.

2019-11-14T18:04:37Z DEBUG DNF version: 4.0.9
2019-11-14T18:04:37Z DDEBUG Command: dnf install -y cloud-init
2019-11-14T18:04:37Z DDEBUG Installroot: /
2019-11-14T18:04:37Z DDEBUG Releasever: 8
2019-11-14T18:04:37Z DEBUG cachedir: /var/cache/dnf
2019-11-14T18:04:37Z DDEBUG Base command: install
2019-11-14T18:04:37Z DDEBUG Extra commands: ['install', '-y', 'cloud-init']
2019-11-14T18:04:37Z DEBUG repo: downloading from remote: AppStream
2019-11-14T18:05:05Z DEBUG error: Curl error (7): Couldn't connect to server for http://mirrorlist.centos.org/?release=8&arch=x86_64&repo=AppStream&infra=stock [Failed to connect to mirrorlist.centos.org port 80: Connection timed out] (http://mirrorlist.centos.org/?release=8&arch=x86_64&repo=AppStream&infra=stock).
2019-11-14T18:05:05Z DEBUG Cannot download 'http://mirrorlist.centos.org/?release=8&arch=x86_64&repo=AppStream&infra=stock': Cannot prepare internal mirrorlist: Curl error (7): Couldn't connect to server for http://mirrorlist.centos.org/?release=8&arch=x86_64&repo=AppStream&infra=stock [Failed to connect to mirrorlist.centos.org port 80: Connection timed out].
2019-11-14T18:05:05Z DDEBUG Cleaning up.
2019-11-14T18:05:05Z SUBDEBUG
Traceback (most recent call last):
  File "/usr/lib/python3.6/site-packages/dnf/repo.py", line 566, in load
    ret = self._repo.load()
  File "/usr/lib64/python3.6/site-packages/libdnf/repo.py", line 503, in load
    return _repo.Repo_load(self)
RuntimeError: Failed to synchronize cache for repo 'AppStream'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.6/site-packages/dnf/cli/main.py", line 64, in main
    return _main(base, args, cli_class, option_parser_class)
  File "/usr/lib/python3.6/site-packages/dnf/cli/main.py", line 99, in _main
    return cli_run(cli, base)
  File "/usr/lib/python3.6/site-packages/dnf/cli/main.py", line 115, in cli_run
    cli.run()
  File "/usr/lib/python3.6/site-packages/dnf/cli/cli.py", line 1124, in run
    self._process_demands()
  File "/usr/lib/python3.6/site-packages/dnf/cli/cli.py", line 828, in _process_demands
    load_available_repos=self.demands.available_repos)
  File "/usr/lib/python3.6/site-packages/dnf/base.py", line 400, in fill_sack
    self._add_repo_to_sack(r)
  File "/usr/lib/python3.6/site-packages/dnf/base.py", line 135, in _add_repo_to_sack
    repo.load()
  File "/usr/lib/python3.6/site-packages/dnf/repo.py", line 568, in load
    raise dnf.exceptions.RepoError(str(e))
dnf.exceptions.RepoError: Failed to synchronize cache for repo 'AppStream'
2019-11-14T18:05:05Z CRITICAL Error: Failed to synchronize cache for repo 'AppStream'

Interestingly it would work on the second boot. I posted on ServerFault about this. – https://serverfault.com/questions/991899/startup-script-centos-8-yum-install-no-network-on-first-boot. I will try to update this article if it goes anywhere as the “sleep 30” is annoying. The first iteration had a sleep 10 and it did not work.

It was strange because I could login and manually run the debug on it and it would succeed.

sudo google_metadata_script_runner --script-type startup --debug

Cloud-Init

Our goal was to use this right? Cloud-Init has a nice module for installing and configuring Salt – https://cloudinit.readthedocs.io/en/latest/topics/modules.html#salt-minion

#cloud-config

yum_repos:
    salt-py3-latest:
        baseurl: https://repo.saltstack.com/py3/redhat/$releasever/$basearch/latest
        name: SaltStack Latest Release Channel Python 3 for RHEL/Centos $releasever
        enabled: true
        gpgcheck: true
        gpgkey: https://repo.saltstack.com/py3/redhat/$releasever/$basearch/latest/SALTSTACK-GPG-KEY.pub

salt_minion:
    pkg_name: 'salt-minion'
    service_name: 'salt-minion'
    config_dir: '/etc/salt'
    conf:
        master: salt.example.com
    grains:
        role:
            - web

This sets up the repo for Salt. I prefer their repo over Epel as Epel tends to be dated. It then sets some simple salt-minion configs to get it going!

How do you set this?

You can set this two ways. One is from the command line if you have the SDK.

% gcloud compute instances create test123-17 --machine-type f1-micro --image-project centos-cloud --image-family centos-8 --metadata-from-file user-data=cloud-init.yaml,startup-script=cloud-bootstrap.sh

Or you can use the console and paste it in plain text.

GCE - Automation - Startup and user-data
GCE – Automation – Startup and user-data

Don’t feel bad if you can’t find these settings. They are buried here.

Finding Automation Settings
Finding Automation Settings

Final Words

In this article we walked through automating the provisioning. You can use cloud-init for all sorts of things such as ensuring its completely up to date before handing off as well as adding users and keys. For our need, we just wanted to get Salt on there so it could plug into config management.

SaltStack – Assisting Windows Updates With win_wua

Summary

Today I got to play with the Salt win_wua module. Anyone that manages Windows servers, they know all about the second Tuesday of the month. the win_wua module can help greatly. I have recently been toying with Salt as mentioned in some of my other articles like Introduction to SaltStack.

Update Methodology

In the environments I manage, we typically implement Microsoft Windows Server Update Services (WSUS) but in a manual fashion so that we can control the installation of the patches. WSUS is more of a gatekeeper against bad patches. We approve updates immediate to only test servers. This lets us burn them in for a few weeks. Then when we’re comfortable, we push them to production. This greatly helped mitigate this conflict of Windows Updates – https://community.sophos.com/kb/en-us/133945

The process to actually install though is manual since we need to trigger the install. It involved manually logging into various servers to push the install button and then reboot. In my past complaints of this I was unable to find something to easily trigger the installation of windows updates.

Win_wua to the rescue

I originally thought I would need a salt state to perform this but the command line module is so easy, I did not bother.

salt TESTSERVER win_wua.list
TESTSERVER:
    ----------
    9bc4dbf1-3cdf-4708-a004-2d6e60de2e3a:
        ----------
        Categories:
            - Security Updates
            - Windows Server 2012 R2
        Description:
            Install this update to resolve issues in Windows. For a complete listing of the issues that are included in this update, see the associated Microsoft Knowledge Base article for more information. After you install this item, you may have to restart your computer.
        Downloaded:
.....

It then spews a ton of data related to the pending updates to be installed. Luckily it has an option for a summary. Surprisingly we use the same “list” to install by setting a flag. The install function expects a list of updates you wish to install but we just want to install all pending ones.

Before we install, check out the summary output

salt TESTSERVER win_wua.list summary=True
TESTSERVER:
    ----------
    Available:
        0
    Categories:
        ----------
        Security Updates:
            4
        Windows Server 2012 R2:
            4
    Downloaded:
        4
    Installed:
        0
    Severity:
        ----------
        Critical:
            3
        Moderate:
            1
    Total:
        4

Ok so let’s install and only see the summary

salt -t 60 LV-PSCADS01 win_wua.list summary=True install=True
LV-PSCADS01:
    Passed invalid arguments to win_wua.list: 'int' object is not callable
    
        .. versionadded:: 2017.7.0
    
        Returns a detailed list of available updates or a summary. If download or
        install is True the same list will be downloaded and/or installed.

Well that’s no fun! Not quite what we expected. It appears its a known bug on 2017.7.1 and fixed. Update your salt minion or perform the manual fix it listed and run again!

salt -t 60 TESTSERVER win_wua.list summary=True install=True
TESTSERVER:
    ----------
    Download:
        ----------
        Success:
            True
        Updates:
            Nothing to download
    Install:
        ----------
        Message:
            Installation Succeeded
        NeedsReboot:
            True
        Success:
            True
        Updates:
            ----------
            9bc4dbf1-3cdf-4708-a004-2d6e60de2e3a:
                ----------
                AlreadyInstalled:
                    False
                RebootBehavior:
                    Never Reboot
                Result:
                    Installation Succeeded
                Title:
                    2019-11 Servicing Stack Update for Windows Server 2012 R2 for x64-based Systems (KB4524445)
            9d665242-c74c-4905-a6f4-24f2b12c66e6:
                ----------
                AlreadyInstalled:
                    False
                RebootBehavior:
                    Poss Reboot
                Result:
                    Installation Succeeded
                Title:
                    2019-11 Cumulative Security Update for Internet Explorer 11 for Windows Server 2012 R2 for x64-based systems (KB4525106)
            a30c9519-8359-48e1-86d4-38791ad95200:
                ----------
                AlreadyInstalled:
                    False
                RebootBehavior:
                    Poss Reboot
                Result:
                    Installation Succeeded
                Title:
                    2019-11 Security Only Quality Update for Windows Server 2012 R2 for x64-based Systems (KB4525250)
            a57cd1d3-0038-466b-9341-99f6d488d84b:
                ----------
                AlreadyInstalled:
                    False
                RebootBehavior:
                    Poss Reboot
                Result:
                    Installation Succeeded
                Title:
                    2019-11 Security Monthly Quality Rollup for Windows Server 2012 R2 for x64-based Systems (KB4525243)

Of course, this is windows so we need a reboot. By default the win_system.reboot waits 5 minutes to reboot. With the flags below we can shorten that.

salt TESTSERVER system.reboot timeout=30 in_seconds=True

Salt State

If I wanted to automate the reboot after the update install, I could make this a state and check for the updates to trigger a reboot. In my scenario, I do not need it but if you want to try, check out this section for the win_wua states. The syntax is slightly different than the module we have been working with on this article.

Updating Multiple Server

If you want to update multiple servers at once you can do something like the following. The -L flag lets you set multiple targets as a comma separated

salt -t 60 -L TESTSERVER,TESTSERVER2,TESTSERVER3 win_wua.list summary=True install=True

salt -L TESTSERVER,TESTSERVER2,TESTSERVER3 system.reboot timeout=30 in_seconds=True

We could even set a salt grain to group these

salt -L TESTSERVER,TESTSERVER2,TESTSERVER3 grains.set wua_batch testservers
salt -G wua_batch:testservers win_wua.list summary=True install=True
salt -G wua_batch:testservers system.reboot timeout=30 in_seconds=True

Throttling

If you are running this on prem or just flat out want to avoid an update and boot storm, you can throttle it using “salt -b” as mentioned in Salt’s documentation.

# This would limit the install to 2 servers at a time
salt -b 2 -G wua_batch:testservers win_wua.list summary=True install=True

Final Words

This article is likely only good if you have salt in your environment somewhere but never thought about using it on Windows. It is a great tool at configuration management on Windows but most Windows admins think of other tools like GPO, SCCM, etc to manage Windows.

Salt State – Intro to SaltStack Configuration

Summary

This article picks up from Configuration Management – Introduction to SaltStack and dives into Salt State. It assumes you have an installed and working SaltStack. I call this intro to SaltStack configuration because this is the bulk of salt. Setting up the salt states and configuration. Understanding the configuration files, where they go and the format is the most important part of Salt.

The way I learn is a guided tour with a purpose and we will be doing just that. Our goal is first to create a salt state that installs apache.

Prepping Salt – Configuration

We need to modify “/etc/salt/master” and uncomment the following

#file_roots:
#  base:
#    - /srv/salt
#

This is where the salt states will be stored. Then we want to actually create that directory.

mkdir /srv/salt

vi /etc/srv/salt/webserver.sls

The contents of webserver.sls are as follows

httpd:
  pkg:
    - installed

This is fairly simple. We indicate a state “apache”, and define that package should be installed. We can apply it specifically as follows.

Applying Our First Salt State

# salt saltmaster1.woohoosvcs.com state.apply webserver
[WARNING ] /usr/lib/python3.6/site-packages/salt/transport/zeromq.py:42: VisibleDeprecationWarning: zmq.eventloop.minitornado is deprecated in pyzmq 14.0 and will be removed.
    Install tornado itself to use zmq with the tornado IOLoop.
    
  import zmq.eventloop.ioloop

saltmaster1.woohoosvcs.com:
----------
          ID: httpd
    Function: pkg.installed
      Result: True
     Comment: The following packages were installed/updated: httpd
     Started: 16:31:35.154411
    Duration: 21874.957 ms
     Changes:   
              ----------
              apr:
                  ----------
                  new:
                      1.6.3-9.el8
                  old:
              apr-util:
                  ----------
                  new:
                      1.6.1-6.el8
                  old:
              apr-util-bdb:
                  ----------
                  new:
                      1.6.1-6.el8
                  old:
              apr-util-openssl:
                  ----------
                  new:
                      1.6.1-6.el8
                  old:
              centos-logos-httpd:
                  ----------
                  new:
                      80.5-2.el8
                  old:
              httpd:
                  ----------
                  new:
                      2.4.37-12.module_el8.0.0+185+5908b0db
                  old:
              httpd-filesystem:
                  ----------
                  new:
                      2.4.37-12.module_el8.0.0+185+5908b0db
                  old:
              httpd-tools:
                  ----------
                  new:
                      2.4.37-12.module_el8.0.0+185+5908b0db
                  old:
              mailcap:
                  ----------
                  new:
                      2.1.48-3.el8
                  old:
              mod_http2:
                  ----------
                  new:
                      1.11.3-3.module_el8.0.0+185+5908b0db
                  old:

Summary for saltmaster1.woohoosvcs.com
------------
Succeeded: 1 (changed=1)
Failed:    0
------------
Total states run:     1
Total run time:  21.875 s

You’ll note 1) the annoying warning which I will be truncating from further messages but 2) that it installed httpd (apache). You can see it also installed quite a few other dependencies that apache required.

Let’s validate quickly

# systemctl status httpd
● httpd.service - The Apache HTTP Server
   Loaded: loaded (/usr/lib/systemd/system/httpd.service; disabled; vendor preset: disabled)
   Active: inactive (dead)
     Docs: man:httpd.service(8)

# ls -la /etc/httpd/
total 12
drwxr-xr-x.  5 root root  105 Nov 10 16:31 .
drwxr-xr-x. 80 root root 8192 Nov 10 16:31 ..
drwxr-xr-x.  2 root root   37 Nov 10 16:31 conf
drwxr-xr-x.  2 root root   82 Nov 10 16:31 conf.d
drwxr-xr-x.  2 root root  226 Nov 10 16:31 conf.modules.d
lrwxrwxrwx.  1 root root   19 Oct  7 16:42 logs -> ../../var/log/httpd
lrwxrwxrwx.  1 root root   29 Oct  7 16:42 modules -> ../../usr/lib64/httpd/modules
lrwxrwxrwx.  1 root root   10 Oct  7 16:42 run -> /run/httpd
lrwxrwxrwx.  1 root root   19 Oct  7 16:42 state -> ../../var/lib/httpd

Looks legit to me! We just installed our first salt state. You can “yum remove” httpd and apply again and it will install. The real power in configuration management is that it knows the desired state and will repeatedly get you there. It is not just a one and done. This is the main difference between provisioning platforms and configuration management.

Installing More Dependencies

WordPress also needs “php-gd” so let’s modify the salt state to add it and then reapply.

httpd:
  pkg:
    - installed

php-gd:
  pkg:
    - installed

Here you can see it did not try to reinstall apache but did install php-gd.

# salt saltmaster1.woohoosvcs.com state.apply webserver

saltmaster1.woohoosvcs.com:
----------
          ID: httpd
    Function: pkg.installed
      Result: True
     Comment: All specified packages are already installed
     Started: 16:41:36.742596
    Duration: 633.359 ms
     Changes:   
----------
          ID: php-gd
    Function: pkg.installed
      Result: True
     Comment: The following packages were installed/updated: php-gd
     Started: 16:41:37.376211
    Duration: 20622.985 ms
     Changes:   
              ----------
              dejavu-fonts-common:
                  ----------
                  new:
                      2.35-6.el8
                  old:
              dejavu-sans-fonts:
                  ----------
                  new:
                      2.35-6.el8
                  old:
              fontconfig:
                  ----------
                  new:
                      2.13.1-3.el8
                  old:
              fontpackages-filesystem:
                  ----------
                  new:
                      1.44-22.el8
                  old:
              gd:
                  ----------
                  new:
                      2.2.5-6.el8
                  old:
              jbigkit-libs:
                  ----------
                  new:
                      2.1-14.el8
                  old:
              libX11:
                  ----------
                  new:
                      1.6.7-1.el8
                  old:
              libX11-common:
                  ----------
                  new:
                      1.6.7-1.el8
                  old:
              libXau:
                  ----------
                  new:
                      1.0.8-13.el8
                  old:
              libXpm:
                  ----------
                  new:
                      3.5.12-7.el8
                  old:
              libjpeg-turbo:
                  ----------
                  new:
                      1.5.3-7.el8
                  old:
              libtiff:
                  ----------
                  new:
                      4.0.9-13.el8
                  old:
              libwebp:
                  ----------
                  new:
                      1.0.0-1.el8
                  old:
              libxcb:
                  ----------
                  new:
                      1.13-5.el8
                  old:
              php-common:
                  ----------
                  new:
                      7.2.11-1.module_el8.0.0+56+d1ca79aa
                  old:
              php-gd:
                  ----------
                  new:
                      7.2.11-1.module_el8.0.0+56+d1ca79aa
                  old:

Summary for saltmaster1.woohoosvcs.com
------------
Succeeded: 2 (changed=1)
Failed:    0
------------
Total states run:     2
Total run time:  21.256 s

The output of state.apply is rather long so we likely will not post too many more. With that said, I wanted to give you a few examples of the output and what it looks like.

Downloading Files

Next we need to download the WordPress files. The latest version is always available via https://wordpress.org/latest.tar.gz. Salt has a Salt State for managed file to download but it requires us to know the hash of the file to ensure it is correct. Since “latest” would change from time to time, we do not know what that is. We have two options. The first is to store a specific version on the salt server and provide that. The second is to use curl to download the file.

download_wordpress:
  cmd.run:
    - name: curl -L https://wordpress.org/latest.tar.gz -o /tmp/wp-latest.tar.gz
    - creates: /tmp/wp-latest.tar.gz

In order for salt to not download the file every time, we need to tell it what the command we are running will store. We tell it, it creates the “/tmp/wp-latest.tar.gz” so it should only download if that file does not exist.

Downloading is not all we need to do though, we also need to extract it.

Extracting

extract_wordpress:
  archive.extracted:
    - name: /tmp/www/
    - source: /tmp/wp-latest.tar.gz
    - user: apache
    - group: apache

The archive.extracted module allows us to specify a few important parameters that are helpful. Applying the state we can see its there!

[root@saltmaster1 salt]# ls -la /tmp/www/
total 4
drwxr-xr-x.  3 apache root     23 Nov 10 16:57 .
drwxrwxrwt. 11 root   root    243 Nov 10 16:59 ..
drwxr-xr-x.  5 apache apache 4096 Oct 14 15:37 wordpress

We actually want it in the “/var/www/html” root so the webserver.sls was modified to the following.

extract_wordpress:
  archive.extracted:
    - name: /var/www/html
    - source: /tmp/wp-latest.tar.gz
    - user: apache
    - group: apache
    - options: "--strip-components=1"
    - enforce_toplevel: False

The issue is the tar has the “wordpress” directory as the root and we want to strip that off. We need the options to pass to tar to strip it. We also need the enforce_toplevel to false as Salt expects a singular top level folder. I found this neat trick via https://github.com/saltstack/salt/issues/54012

# Before
# ls -la /var/www/html/wp-config*
-rw-r--r--. 1 apache apache 2898 Jan  7  2019 /var/www/html/wp-config-sample.php
# salt saltmaster1.woohoosvcs.com state.apply webserver

# After
# ls -la /var/www/html/wp-config*
-rw-r-----. 1 apache apache 2750 Nov 10 18:03 /var/www/html/wp-config.php
-rw-r--r--. 1 apache apache 2898 Jan  7  2019 /var/www/html/wp-config-sample.php

Sourcing the Config

We now have a stock WordPress install but we need to configure it to connect to the database.

For that I took a production wp-config.php and placed it in “/srv/salt/wordpress/wp-config.php” on the salt master. I then used the following salt state to push it out

/var/www/html/wp-config.php:
  file.managed:
  - source: salt://wordpress/wp-config.php
  - user: apache
  - group: apache
  - mode: 640

Set Running Salt State

What could would Apache do if it weren’t running. We do need a salt state to enable and run it!

start_webserver:
  service.running:
  - name: httpd
  - enable: True
# systemctl status httpd
● httpd.service - The Apache HTTP Server
   Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled; vendor preset: disabled)
   Active: active (running) since Sun 2019-11-10 18:27:24 CST; 1min 5s ago

Final Words

Through this we configured a webserver.sls salt state. We used it to install apache and a basic php module necessary as well as push out a configuration file. As you can likely tell from these instructions, it is an iterative approach to configuring the salt state for your need.

This first iteration of the webserver.sls is far from complete or best practice. It is meant as a beginner’s guide to walking through the thought process. Below is the full webserver.sls file for reference

httpd:
  pkg:
    - installed

php-gd:
  pkg:
    - installed

download_wordpress:
  cmd.run:
    - name: curl -L https://wordpress.org/latest.tar.gz -o /tmp/wp-latest.tar.gz
    - creates: /tmp/wp-latest.tar.gz

extract_wordpress:
  archive.extracted:
    - name: /var/www/html
    - source: /tmp/wp-latest.tar.gz
    - user: apache
    - group: apache
    - options: "--strip-components=1"
    - enforce_toplevel: False

/var/www/html/wp-config.php:
  file.managed:
  - source: salt://wordpress/wp-config.php
  - user: apache
  - group: apache
  - mode: 640

start_webserver:
  service.running:
  - name: httpd
  - enable: True

Configuration Management – Intro to SaltStack

Summary

SaltStack or Salt for short is an open source configuration management platform. It was first released in the early 2010’s as a potential replacement for Chef and Puppet. In this guide we will walk through some high level details of Salt and a basic install. If you already have Salt installed, please skip ahead to the next article when it is published.

Next – Salt State – Intro to SaltStack Configuration

What is Configuration Management?

A configuration management tool allows you to remotely configure and dictate configurations of machines. Through this multi-part series we will work through that with the use case of https://blog.woohoosvcs.com/. At some point this site may need multiple front ends. It has not been decided if that method will be Kubernetes, Google App Engine or VMs. If the VM route is chosen it will make sense to have an easy template to use.

What Configuration Management is not

Configuration management typically does not involve the original provisioning of the server. There are typically other tools for that such as Terraform.

Salt Architecture

Salt has three main components to achieve configuration management. Those are the salt master, minion and client. Salt can be configured highly available with multi master but it is not necessary to start out there. For the sake of this document and per Salt’s best practices we can add that later if necessary. – https://docs.saltstack.com/en/latest/topics/development/architecture.html

Salt Client

The salt client is a command line client that accepts commands to be issued to the salt master. It is typically on the salt master. You can use it to trigger expected states.

Salt Master

The Salt Master is the broker of all configuration management and the brains. Requests/commands received from the client make their way to the master which then get pushed to the minion.

Salt Minion

The minion is typically loaded onto each machine you wish to perform configuration management on. In our case, it will be the new front ends we spin up as we need them.

Firewall Ports

The Salt Master needs ports TCP/4505-4506 opened. The minions check in and connect to the master on those ports. No ports are needed for the minions as they do not listen on ports.

More on this can be found on their firewall page – https://docs.saltstack.com/en/latest/topics/tutorials/firewall.html

Where to install

Typically you want the master to be well connected since the minions will be connecting to it. Even if you are primarily on prem, it is not a bad idea to put a salt master in the cloud.

Installing Salt

Prerequisites

OS: CentOS 8
VM: 1 core, 1 GB RAM, 12 GB HDD
Install: Minimal

Installation

For the installation we will be closely following Salt’s documentation on installing for RHEL 8 – https://repo.saltstack.com/#rhel

For the sake of this lab we will have the client, master and minion all on the same server but it will allow us to build out the topology.

Now to the install!

# I always like to start out with the latest up to date OS
sudo yum update

# Install the salt repo for RHEL/CentOS
sudo yum install https://repo.saltstack.com/py3/redhat/salt-py3-repo-latest.el8.noarch.rpm

# Install minion and master
sudo yum install salt-master salt-minion

# Reboot for OS updates to take effect
reboot

Post Install Configuration

We need to setup a few things first. The Salt guide is great at pointing these out – https://docs.saltstack.com/en/latest/ref/configuration/index.html

On the minion we need to edit “/etc/salt/minion”. The following changes need to be made. If/when you roll this out into production you can use DNS hostnames.

#master: salt
master: 192.168.116.186

We will also open up the firewall ports on the master

firewall-cmd --permanent --zone=public --add-port=4505-4506/tcp
firewall-cmd --reload

Start up Configuration Management Tools

Now we are ready to start up and see how it runs!

sudo systemctl enable salt-master salt-minion
sudo systemctl start salt-master salt-minion

Wait about 5 minutes. It takes a little bit to initialize. Once it has you can run “sudo salt-key -L”. When a minion connects to the master, the master does not allow it to connect automatically. It has to be permitted/admitted. salt-key can be used to list minions and allow them.

$ sudo salt-key -L
Accepted Keys:
Denied Keys:
Unaccepted Keys:
saltmaster1.woohoosvcs.com
Rejected Keys:

$ sudo salt-key -A
The following keys are going to be accepted:
Unaccepted Keys:
saltmaster1.woohoosvcs.com
Proceed? [n/Y] Y
Key for minion saltmaster1.woohoosvcs.com accepted.
[dwcjr@saltmaster1 ~]$ sudo salt-key -L
Accepted Keys:
saltmaster1.woohoosvcs.com
Denied Keys:
Unaccepted Keys:
Rejected Keys:

We used salt-key -A to accept all unaccepted keys.

Testing

$ sudo salt saltmaster1 test.version
[WARNING ] /usr/lib/python3.6/site-packages/salt/transport/zeromq.py:42: VisibleDeprecationWarning: zmq.eventloop.minitornado is deprecated in pyzmq 14.0 and will be removed.
    Install tornado itself to use zmq with the tornado IOLoop.
    
  import zmq.eventloop.ioloop

No minions matched the target. No command was sent, no jid was assigned.
ERROR: No return received
[root@saltmaster1 ~]# salt '*' test.version
[WARNING ] /usr/lib/python3.6/site-packages/salt/transport/zeromq.py:42: VisibleDeprecationWarning: zmq.eventloop.minitornado is deprecated in pyzmq 14.0 and will be removed.
    Install tornado itself to use zmq with the tornado IOLoop.
    
  import zmq.eventloop.ioloop

saltmaster1.woohoosvcs.com:
    2019.2.2

Well that is an ugly error code. It seems to have been introduced in 2019.2.1 but not properly fixed in 2019.2.2. My guess is the next release will fix this but it seems harmless. – https://github.com/saltstack/salt/issues/54759. We do, however, get the response so this is a success.

Final Words

At this point we have a salt-master and salt-minion setup, albeit on the same host. We have accepted the minion on the master and they are communicating. The next article will start to tackle setting up Salt states and other parts of the salt configuration.

Next – Salt State – Intro to SaltStack Configuration

Mobile Fuel Payment – Securing Your Credit Card

Summary

In order to help cut down on Credit Card fraud and increase convenience, gas stations have been implementing mobile fuel payment options. This allows you to pay at the pump without inserting a credit card.

Prior to this, Credit Card companies have been implementing EMV. This is the chip “dip” or NFC contactless payment that we all have seen recently. These are much more secure because they cannot easily be copied/duplicated. Credit Card fraud in Europe was terrible until they started implementing this. It was more terrible than in the United States which is why the United States later adopted it.

The Problem Mobile Fuel Payment Helps

Nearly all retailers are required to provide this option today in the United States. That is, except for Gas Stations. They received an extension. Perpetrators of fraud realize this and therefore it is not uncommon for Credit Card skimmers to be installed. Skimmers help people intending to commit fraud by capturing card information as it is inserted. It is usually combined with a camera to help capture pin entries or other values on the card.

How Does Mobile Fuel Payment actually help this?

I have yet to see a fuel pump that accepts a chip card so any card inserted with a chip has its magnetic strip read. This is susceptible to cloning. Mobile Fuel Payment saves you from having to actually insert a card.

Many of the major gas companies have their own app.

Near me, Chevron and Sunoco both accept this but many more also accept it. Since those are the two major chains near me, I will provide my thoughts on it.

Chevron’s Mobile Fuel Payment

Chevron’s app was easy to download and register. Payment methods are unfortunately scarce. It accepts only PayPal and their own branded Advantage Card or Gift cards. I had PayPal already, so I linked my account to the app. I then went to try to use it and the payment failed. In PayPal, my checking account was the only linked account. The Chevron app gave me no indication of this but PayPal did email me with a failure. The Chevron app also hung in this scenario and I had to “force stop” it to recover. I had to do some digging with PayPal and then it suggested I add my debit/credit card.

After the first experience, I decided to update my bank information and add a credit card. Unfortunately the second time resulted in the same issue. To be fair though, this decline was from PayPal, not Chevron. It seems like fuel charge issues are common though as they have an article for this – https://www.paypal.com/us/smarthelp/article/why-was-my-fuel-purchase-declined-troubleshooting-tips-faq4022

Roughly 30 minutes on hold with PayPal, I gave up. I did not want to waste any more of my Sunday. Maybe I will try again later when they’re better staffed or wait until the Chevron App supports Google Pay.

Chevron's Mobile Fuel Payment App Features and Options
Chevron’s Mobile Fuel Payment App Features and Options

Sunoco’s Mobile Fuel Payment

Sunoco’s app has been very intuitive and it accepts Google Pay, major credit cards and their own rewards and gift cards. The process was very straight forward. It provided estimates on time to start as well as a timer that I had to complete the start fueling. The Google Pay method worked flawlessly.

Sunoco's Mobile Fuel Payment App Features and Options
Sunoco’s Mobile Fuel Payment App Features and Options

Common Features

Some of the common features between these apps are the ability to review receipts and find locations that support Mobile Fuel Payment. I like to keep a copy of my receipts, at least for a few days. In the event that I get a fill of bad gas, I want proof I went to the station. It has never happened but there will be that day.

One thing to note is just because Chevron supports Mobile Fuel Payment does not mean every station does. The stations are almost always a franchise with owners deciding when to perform upgrades. Even if they were not, it still takes time to upgrade all of the stations to support this.

Other Stations and Options

Many other gas stations offer this. I will not dive into the research for that. My hope is just to open you to the idea that your chain likely has this implemented. Feel free to test yours out and comment back with your experience.

Final Words

If you have not looked into using mobile fuel payment options and you keep getting your credit card compromised, give it a try. It is very likely to help with this situation.