How do I print the second column with awk?

awk '{print $2}' file. Fields are split on runs of whitespace by default, so it works on ls -l and ps output without setup. For a CSV, set the separator: awk -F',' '{print $2}' data.csv. $NF is the last column, $(NF-1) the one before it, and $0 the whole line. Note the single quotes around the program: without them the shell would expand $2 itself, usually to nothing.

How do I sum a column with awk?

awk '{sum += $1} END {print sum}' file adds up column 1 and prints the total once at the end. The average is the same idea with the line counter: awk '{sum += $3} END {print sum/NR}' file. For a CSV add -F','. Summing per key needs an array: awk '{s[$1] += $2} END {for (k in s) print k, s[k]}' totals column 2 for each distinct value in column 1.

What is the difference between sub and gsub in awk?

sub(/regex/, "new") replaces the first match, gsub(/regex/, "new") replaces all of them. Both edit their target in place ($0 unless you pass a third argument, like gsub(/,/, "", $3)) and return the number of replacements, not the new string. When you want the result as a value with the original untouched, gawk has gensub(), which returns the modified string instead.

Can awk use shell variables?

Yes, with -v: awk -v host="$HOSTNAME" '$1 == host {print $2}' servers.txt. Each -v defines one awk variable before the program runs. Splicing the variable into the program text with double quotes works until the value contains a quote or backslash, so -v is the habit worth keeping. Environment variables are also readable directly as ENVIRON["HOME"] without any flag.

When should I use awk instead of grep or sed?

grep only selects lines, and does that fastest. sed rewrites lines with regex and is best for pure find-and-replace. awk is the one that understands fields, so reach for it when the job mentions columns, arithmetic, or state across lines: print column 2, sum column 3, count occurrences per user. The three chain well, but a grep piped into awk can usually become just awk '/pattern/ {...}'.

awk Examples Cheat Sheet: print $1, -F & NR/NF (2026)

Print Columns

Fields split on whitespace by default. is the first field, the last, the whole line.

Command	What it does
	First column of every line
	Two columns, separated by a space
	Last column
	Second-to-last column
	Mix fields with literal text
	PID and command from ps
	Size and name from ls
	Number every line

Field Separators

sets the input separator; is what goes between fields on output.

Command	What it does
	Comma-separated input
	Tab-separated input
	Every username on the system
	The separator can be a regex: comma or semicolon
	Comma with optional spaces around it
	Same as -F, set inside the program
	Convert CSV to TSV ( forces the rebuild)
	Custom output separator

NR and NF

is the current line number, the number of fields on the line. Both double as filters.

Command	What it does
	Print line 5 only
	Skip the header line
	A line range
	Every second line
	Count lines, like wc -l
	How many fields each line has
	Drop blank lines (0 fields is false)
	Only lines with exactly 7 fields (catch bad CSV rows)
	FNR resets per file; print each file's name once

Patterns and Conditions

The shape is : the action runs on lines where the pattern holds. Either half can be omitted.

Command	What it does
	Lines matching a regex (like grep)
	Column 2 of matching lines
	Lines NOT matching
	Exact string match on a field
	Numeric comparison on a field
	Regex match on one field only
	Field does not match
	Combine with && and \|\|
	Range: from one match to the next
	Disk usage percent per mount

if/else lives inside the action:

awk '{if ($3 > 100) print $1 " high"; else print $1 " ok"}' data.txt
awk '{print $1, ($3 > 100 ? "high" : "ok")}' data.txt

BEGIN and END

runs before the first line, after the last: headers, totals, averages.

Command	What it does
	Sum a column (the classic)
	Average of a column
	Maximum of a column
	Count matches ( prints 0 instead of blank)
	Add a header line
	BEGIN is where FS/OFS setup goes

printf

formats output with C-style specifiers and adds no newline unless you write .

Command	What it does
	String, one per line
	Left-pad to 15, right-pad to 8: aligned columns
	Two decimal places
	Integer plus a literal percent sign
	Mix strings and numbers
	Works in END too

Arrays and Counting

awk arrays are string-keyed maps, which makes counting and grouping one-liners.

Command	What it does
	Count occurrences of each value in column 1
	Remove duplicate lines, order preserved (uniq without sort)
	First line per distinct column-1 value
	Sum column 2 per key in column 1
	Keys that appear more than once

The count-and-sort pipeline for "top talkers" in a log (see the grep cheat sheet for the pure-grep half):

awk '{print $1}' access.log | sort | uniq -c | sort -rn | head

String Functions

Command	What it does
	Line length in characters
	Lines longer than 80 characters
	First 3 characters of field 1
	Change case
	Position of a substring (0 if absent)
	Split a field on and use a piece
	Replace first match: trim leading spaces
	Replace all matches, here only in field 3
	Format into a string

and modify in place and return the replacement count. For pure stream find-and-replace, sed is usually the shorter tool.

Variables

Command	What it does
	Pass a value in with -v
	Pass a shell variable in safely
	Match against a variable (in a script)
	Read environment variables directly
	Output record separator: join lines with commas
	Paragraph mode: records split on blank lines

Scripts and Recipes

Past one line, put the program in a file and run it with , or give it a shebang.

awk -f report.awk access.log

#!/usr/bin/awk -f
BEGIN { FS = ","; print "user,total" }
NR > 1 { total[$1] += $3 }
END { for (u in total) print u "," total[u] }

Everyday pipelines:

Command	What it does
	Processes over 50% CPU
	Mounts over 80% full ( strips the %)
	Memory in use
	Listening addresses and ports
	Human (non-system) users

Gotchas

Single-quote the program. In double quotes the shell expands before awk runs, which silently breaks the script.
Not all awks are equal: Linux usually has gawk or mawk, macOS ships an older BSD awk. , , and -style classes are gawk extras; everything on this sheet sticks to the portable core.
Comparisons are type-sensitive: compares numbers, but if the field has stray characters awk may compare strings. Force a number with .
Assigning to any field (even ) rebuilds with OFS, which rewrites the separators on output. Sometimes that is exactly what you want; sometimes it is the bug.
Uninitialized variables are and 0 at the same time, so needs no setup, and a missing field prints as empty rather than erroring.
grep selects lines and sed rewrites them; see the grep cheat sheet and sed cheat sheet for the other two thirds of the toolkit.

awk Examples