Let's say that example.txt contains the following two lines.

~]# cat example.txt
foo
bar

The sed command with the d (delete) option can be used to delete lines from example.txt. In this example, lines matching "foo" will be removed.

sed '/foo/d' example.txt

Which should return the following.

bar

However, example.txt will still contain both foo and bar.

~]# cat example.txt
foo
bar

The -i or --in-place flag is needed to actually perform the removal.

sed -i '/foo/d' example.txt

Now example.txt will only contain bar.

~]# cat example.txt
bar

nobody owner

Let's say that example.txt is owned by John Doe.

~]# ls -l example.txt
-rw-r--r--. 1 john.doe john.doe 64 Nov 26 01:33 example.txt

When using the -i or --in-place flag to actually perform the removal, the file will be updated to be owned by the "nobody" user.

~]# ls -l example.txt
-rw-r--r--. 1 nobody john.doe 64 Nov 26 01:33 example.txt

The -c or --copy flag can be used to so that the file remains owned by John Doe.

sed --copy -i '/foo/d' example.txt