"Key is not unique for value ''" thrown by read_csv - ansible

Question

My playbook used to be able to run successfully without any issues.
But recently, I ran the playbook again (without changing anything) and got this error:
TASK [read_csv] *****************************************************
fatal: [localhost]: FAILED! => {"changed": false, "msg": "Key 'username' is not unique for value ''"}
This is the playbook:
---
- name: Read User
hosts: localhost
tasks:
- read_csv:
path: opt/ansible.csv
key: username
fieldnames: username,usergroups,ipaddress,lastlogindate
delimiter: ','
register: usersdata
This is an example of the CSV file that I am using:
Username
User_groups
IP Address
Last Login Date
Apple
A
192.34.50.183
12/4/2021
Blizz
B
192.34.50.145
3/8/2021
What is value ''? And why is using 'username' as the key now wrong?
===============================================================
Update:
I tried the playbook suggested by Vladimir Botka below, and it was able to run successfully.
PLAY [Read User] ***************************************************************
TASK [Gathering Facts] *********************************************************
ok: [localhost]
TASK [read_csv] ****************************************************************
ok: [localhost]
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.dict": {}
}
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.list": [
{
"IP Address": "192.169.60.220 - XXX.XXX.60.254",
"Last Login Date": "30-Jul-21",
"User_groups": "B",
"Username": "Vijay"
},
{
"IP Address": "XXX.XXX.60.146 - XXX.XXX.60.147",
"Last Login Date": "1-Jan-21",
"User_groups": "ecquaria",
"Username": "wangjcecquaria"
},
…
TASK [read_csv] ****************************************************************
ok: [localhost]
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.dict": {
"Kyawryobi": {
"IP Address": "XXX.XXX.60.184 - XXX.XXX.60.185",
"Last Login Date": "28-Jul-21",
"User_groups": "Ryobi",
"Username": "Kyawryobi"
},
"Nikhil": {
"IP Address": "XXX.XXX.60.1 - XXX.XXX.60.80",
"Last Login Date": "30-Jul-21",
"User_groups": "A",
"Username": "Joshua"
},
"Vijay": {
…
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.list": []
}
TASK [read_csv] ****************************************************************
ok: [localhost]
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.dict": {}
}
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.list": [
{
"ipaddress": "IP Address",
"lastlogindate": "Last Login Date",
"usergroups": "User_groups",
"username": "Username"
},
{
"ipaddress": "192.169.60.220 - XXX.XXX.60.254",
"lastlogindate": "30-Jul-21",
"usergroups": "A",
"username": "Joshua"
},
{
…
TASK [read_csv] ****************************************************************
ok: [localhost]
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.dict": {
"Kyawryobi": {
"ipaddress": "XXX.XXX.60.184 - XXX.XXX.60.185",
"lastlogindate": "28-Jul-21",
"usergroups": "Ryobi",
"username": "Kyawryobi"
},
"Nikhil": {
"ipaddress": "XXX.XXX.60.1 - XXX.XXX.60.80",
"lastlogindate": "30-Jul-21",
"usergroups": "A",
"username": "Joshua"
},
"Username": {
…
TASK [debug] *******************************************************************
ok: [localhost] => {
"usersdata.list": []
}
PLAY RECAP *********************************************************************
localhost : ok=13 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
Next, I tried running my playbook again without changing anything else in the settings or playbook and there is no more error.
PLAY [Read User] ***************************************************************
TASK [Gathering Facts] *********************************************************
ok: [localhost]
TASK [read_csv] ****************************************************************
ok: [localhost]
PLAY RECAP *********************************************************************
localhost : ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
I think the error, "Key is not unique for value ''", might be the result of a bug, and running a playbook that triggers ansible to output a dictionary and list using the usersdata variable might have resolved it. I have checked and there is nothing wrong with the data in my CSV. There are no missing values in the 'Username' column, neither do I have the word 'username' duplicated anywhere else.

Answer 1

By default, the module community.general.read_csv takes the first line of a file as a header. You can declare fieldnames if the header is missing. Quoting:
A list of field names for every column. This is needed if the CSV does not have a header.
Given the file
shell> cat /tmp/ansible.csv
Username,User_groups,IP Address,Last Login Date
Apple,A,192.34.50.183,12/4/2021
Blizz,B,192.34.50.145,3/8/2021
You don't have to declare fieldnames because the first line is the header
- read_csv:
path: /tmp/ansible.csv
register: usersdata
gives (see the notes below on the output dictionary vs. list)
usersdata.list:
- IP Address: 192.34.50.183
Last Login Date: 12/4/2021
User_groups: A
Username: Apple
- IP Address: 192.34.50.145
Last Login Date: 3/8/2021
User_groups: B
Username: Blizz
If you do declare fieldnames all lines are treated as data
- read_csv:
path: /tmp/ansible.csv
fieldnames: username,usergroups,ipaddress,lastlogindate
register: usersdata
gives the list with renamed fields including an additional item of the original header
usersdata.list:
- ipaddress: IP Address
lastlogindate: Last Login Date
usergroups: User_groups
username: Username
- ipaddress: 192.34.50.183
lastlogindate: 12/4/2021
usergroups: A
username: Apple
- ipaddress: 192.34.50.145
lastlogindate: 3/8/2021
usergroups: B
username: Blizz
If you declare a key in addition to the fieldnames
- read_csv:
path: /tmp/ansible.csv
key: username
fieldnames: username,usergroups,ipaddress,lastlogindate
register: usersdata
you get the dictionary of this key's values
usersdata.dict:
Apple:
ipaddress: 192.34.50.183
lastlogindate: 12/4/2021
usergroups: A
username: Apple
Blizz:
ipaddress: 192.34.50.145
lastlogindate: 3/8/2021
usergroups: B
username: Blizz
Username:
ipaddress: IP Address
lastlogindate: Last Login Date
usergroups: User_groups
username: Username
Notes
The output depends on the parameter key.
a) Without a key
- read_csv:
path: /tmp/ansible.csv
register: usersdata
- debug:
var: usersdata.dict
- debug:
var: usersdata.list
you get an empty dictionary and a list of the lines
usersdata.dict: {}
usersdata.list:
- IP Address: 192.34.50.183
Last Login Date: 12/4/2021
User_groups: A
Username: Apple
- IP Address: 192.34.50.145
Last Login Date: 3/8/2021
User_groups: B
Username: Blizz
b) If you specify a key (it's case-sensitive)
- read_csv:
path: /tmp/ansible.csv
key: Username
register: usersdata
you get a dictionary of this key's values and an empty list
usersdata.dict:
Apple:
IP Address: 192.34.50.183
Last Login Date: 12/4/2021
User_groups: A
Username: Apple
Blizz:
IP Address: 192.34.50.145
Last Login Date: 3/8/2021
User_groups: B
Username: Blizz
usersdata.list: []
Example of a complete playbook for testing
- hosts: localhost
tasks:
- read_csv:
path: /tmp/ansible.csv
register: usersdata
- debug:
var: usersdata.dict
- debug:
var: usersdata.list
- read_csv:
path: /tmp/ansible.csv
key: Username
register: usersdata
- debug:
var: usersdata.dict
- debug:
var: usersdata.list
- read_csv:
path: /tmp/ansible.csv
fieldnames: username,usergroups,ipaddress,lastlogindate
register: usersdata
- debug:
var: usersdata.dict
- debug:
var: usersdata.list
- read_csv:
path: /tmp/ansible.csv
key: username
fieldnames: username,usergroups,ipaddress,lastlogindate
register: usersdata
- debug:
var: usersdata.dict
- debug:
var: usersdata.list