Indradhanush Gupta
3 min read

Categories

Tags

Last week at work I was working on a bash script, part of which needed to get the status of a Kubernetes node. Specifically I was running this command:

$ kubectl get nodes node-0

And if everything looked well, the output would be something similar to:

NAME      STATUS     ROLES     AGE       VERSION
node-0    Ready      <none>    22h       v1.11.1

As part of the script I was using cut to extract the value of the second column by doing:

1
2
3
4
5
6
#!/bin/bash

result=$(kubectl get nodes node-0 | tail -n 1)
state=$(echo $result | cut -d ' ' -f2)

echo $state

Let’s break that down:

On line three, I pipe the output of kubectl get nodes node-0 to tail and store only the second line of the output since the first line contains the column headings and is not useful in this case.

On line four, I use the cut command to split the words by the character space, indicated by the -d ' ' flag and extract the value of the second column, indicated by the -f2 flag. When I ran the script I saw the output as expected:

Ready

However, shellcheck complained about the line number four of the script (it complained about the last line as well, but that’s not relevant to this post):

In test.sh line 4:
state=$(echo $result | cut -d ' ' -f2)
             ^-- SC2086: Double quote to prevent globbing and word splitting.

Behold the mighty bug 🐛

So I fixed the warning by wrapping it around double quotes:

state=$(echo "$result" | cut -d ' ' -f2)

But to my surprise, all I had now was an empty string as output. To debug this I printed the value of $result by:

echo $result

Which gave the output:

node-0 Ready <none> 3d v1.11.1

Debugging 🔬

I copied this and ran piped it to cut -d ' ' -f2 directly on the bash shell. And I got the expected output – Ready.

I copied this and ran the code directly on the shell this time:

$ result="node-0 Ready <none> 3d v1.11.1"
$ state=$(echo "${result}" | cut -d ' ' -f2)
$ echo $state
Ready

I was able to get the desired output this time. I asked in my company’s Slack and eventually found out that if I ran the following in the shell directly:

$ result=$(kubectl get nodes node-0 | tail -n 1)
$ echo "$result"
node-0    Ready     <none>    3d        v1.11.1

Versus, if I ran:

$ echo $result
node-0 Ready <none> 3d v1.11.1

Notice the difference yet? The difference is the presence of double quotes around $result. When we wrap a variable around double quotes we are forcing bash to print the variable as is and to quote the linux documentation project:

Using double quotes the literal value of all characters enclosed is preserved, except for the dollar sign, the backticks (backward single quotes, ``) and the backslash.

Lessons learned 📖

I learned two things out of this:

  1. The output of kubectl get nodes is not separated by a single space character.
  2. Bash cleans up extra spaces when writing to stdout unless you use variable quoting.

Now that the problem had been identified, the solution was easy – I needed to squeeze out the extra spacing between the columns before using cut on it:

state=$(echo "${result}" | tr -s "[:space:]" | cut -d ' ' -f2)

Another important lesson that I learned from this exercise is that when manipulating strings plain text format isn’t the best option. kubectl can output in both yaml and json. I prefer json as it’s easier to parse by the naked eye and I can use jq to manipulate json output programmatically. As a result I was able to minimize my script to a single line:

state=$(kubectl get nodes node-0 -o json | jq ".status.conditions[-1])"

The -o json flag instructs the Kubernetes API to return the reponse in json and we pipe the result to jq for further processing, where the filters are specific to Kubernetes.

Addendum 📢

  1. It’s generally considered good practice to wrap bash variables around {}. So instead of echo "$result", using echo ${result} is safer and can help avoid bugs involving string expansion in bash.

  2. If you’re into Kubernetes, kubectl also supports the -o jsonpath that lets you directly specify the json filter and returns the minimized output instead of the entire json string. I tried using jsonpath but in my case I noticed that negative indexing raised a panic. I haven’t yet filed an issue about this on the Kubernetes project’s issue tracker and intend to do so once I’ve verified this with the latest Kubernetes components.