Hi Tae,

Hi,

I realise this case is old, but we developed a Python script that can keep making API calls until it retrieves all the people data.

Here is the script code. Hopefully it is useful.

====================================================================================

import requests

import json

from pathlib import Path

import csv

# Define Okta API base URL

base_okta_url = "https://YOUR-OKTA-NAME.okta.com"

# Define the API token required by Okta

okta_apikey = "PUT-YOUR-OKTA-API-TOKEN-HERE"

# Define record limit (200 is the maximum accepted by Okta)

limit = 200

# If get_all_data is True, then the script will call Get User until all data is retrieved.

get_all_data = True

#get_all_data = False

# Define the Okta Users endpoint

url = base_okta_url + "/api/v1/users" + "?limit=" + str(limit)

# CSV - JSON Response Mapping

# ADD OTHER OKTA ATTRIBUTES OR CHANGE THE ORDER OF COLUMNS FOR THE CSV BELOW

column_list = ["firstName","lastName","displayName","manager","title","login","email","department"]

# Construct the headers

headers = {

  'content-type': 'application/json',

'Authorization': 'SSWS ' + okta_apikey

}

# Setup path for current directory and file name for the Export file

data_folder = Path(".")

export_file = data_folder / "okta_user_export.csv"

exportfile = open(export_file,"w+")

# Join the values from the column list to build a CSV title row, with delimeters

col_headings = '"' + '", "'.join(column_list) + '"'

exportfile.write(col_headings)

csv_delim = '","'

# Set conditions for while loop. Used with pagination.

next = False

first = True

next_url = ""

# While loop use to control first and subsequent requests

while (next or first):

# If next_url is defined, this is not the first time through the loop,

# and we should have the next_url generated from Okta

response_json = []

# If this is the second (or greater) call to Get Users

if next:

# Make the API call

response = requests.request("GET", next_url, headers=headers)

response_json = response.json()

# Parse the LINK headers from the response.

response_links = requests.utils.parse_header_links(response.headers['Link'].rstrip('>').replace('>,<', ',<'))

next_url = ""

# Look for the 'next' url in the response links.

for linkobj in response_links:

if linkobj['rel'] == 'next':

next_url = linkobj['url']

next = True

else:

next = False

# If this is the first call to Get Users

if first:

# Make the request to Okta and record the response

response = requests.request("GET", url, headers=headers)

response_json = response.json()

# Parse the LINK headers from the response.

if (response.status_code == 200):

response_links = requests.utils.parse_header_links(response.headers['Link'].rstrip('>').replace('>,<', ',<'))

next_url = ""

for linkobj in response_links:

#print(linkobj['rel'])

if linkobj['rel'] == 'next':

next_url = linkobj['url']

next = True

# Turn off first flag so that the next option is selected on subsequent loops

first = False

else:

print(response.status_code)

print(response_json)

exit(1)

# write the response to file

for entry in response_json:

# Build the CSV row

csv_record = []

# Loop through in column in the column list

for col in column_list:

# If we can find column in profile object append it

if col in entry['profile']:

csv_record.append(entry['profile'][col])

# Else if we can find column in root object append it

elif col in entry:

csv_record.append(entry[col])

# Else if we can't find, write a dash to preserve column format

else:

csv_record.append("-")

# Write the CSV row to the file

csv_line = '"' + csv_delim.join( csv_record ) + '"'

exportfile.write("%s\n" % (csv_line))

# If we don't want to get all data (i,e we only want 'limit' records), exit the loop.

if get_all_data == False:

break

exportfile.close

print("Done.")

=====================================================================================

I'm really sorry to hear you're frustrated Deb. I reached out to our product team to share your feedback and they let me know this functionality is on our long term roadmap. I'll be sure to share more when we have more specific delivery timeline.

In the meantime, they recommended the following as a workaround:

The exportable Password Health report provides a list of all users (as well as their logins, current status, activation date, auth source, last login date and last password change date). Depending on what you want to do with the user data, this report may cover your needs for the time being

Hope this is helpful and again- we'll update everyone as we have more info on timing.

I'll second Deb's request. I've had to report on Okta users and suffer the same problem. I use an API script to do this, but Deb is right that for an enterprise product like Okta that is meant to scale to thousands of users, a simple user export function should not be missing. We need to routinely audit and report on Okta users. Please, improve the reporting in general in Okta. Too simplistic for a product of this sort.

Another dirty python example. If you need to pull more than 200 users, borrow code from the previous comments and integrate.

import requests

api_token = '<your token>'

headers = {

'Accept': 'application/json',

'Content-Type': 'application/json',

'Authorization': f"SSWS {api_token}"

}

url = 'https://<your okta domain>.okta.com/api/v1/users'

while len(url) > 0:

r = requests.get(url, headers=headers)

print(r.content)

links = r.links

url = links['next']['url']

And here's the PowerShell version if needed. This will automatically keep looping until the entire directory is downloaded. I then create a custom PS object to hold the properties i want and then spit it out to CSV.

Feel free to contribute and use the code.

https://github.com/ajhstn/okta-tools/blob/master/Export-AllOktaUsers.ps1

# Use TLS 1.2

[Net.ServicePointManager]::SecurityProtocol = [Net.SecurityProtocolType]::Tls12

# Setup Auth Header

$header = @{

Authorization = "SSWS <your token here>"

}

# API Resource Endpoint

$url = "https://<your okta domain here>.okta.com/api/v1/users"

$Users = @()

do {

$result = Invoke-WebRequest -Uri $url -Headers $header -UseBasicParsing

$json = $result.Content | convertfrom-json

$Users += $json

$next = (($result.Headers.Link -split "`n")[1] -replace "[\<\>]" -split ";")[0]

$url = $next

}

while ($result.Headers.Link[1] -match "next")

$Users | Foreach-object {

[pscustomobject]@{

FirstName = $_.profile.firstName

LastName = $_.profile.lastName

DisplayName = if ($_){$_.profile.displayName}else{$_}

login = $_.profile.login

email = $_.profile.email

Description = $_.profile.Description

Department = $_.profile.department

State = $_.status

StatusChanged = $_.statusChanged

Created = $_.created

LastLogin = $_.lastLogin

PasswordChanged = $_.passwordChanged

Error = if ($_ -eq $null){"Something wrong"}else{}

}

} | Export-Csv ./AllUsers-Feb2019.csv -Force

#Write-Output "($Users).count:OK"

Hi Andrew,

I tried your PowerShell code but still only get 200 users to download (I have 350 or so total). There must be a way to get all 350 users to download isn't there?

Thanks!