Developer

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy

Hi Guy,

Do you have found your solution ? the code look... 

do you have look pycentral (v1 ?) already make this stuff

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy

Hello,

I understand better now that the calls for the initial authentication to create a session, and then the call to obtain an auth code, are rate limited. So once I manage to get my access token I just need to keep refreshing that (at least once every 15 days I think?).

I did look at PyCentral, but I think that is for New Central only? Or have I got that wrong? I am trying to use the Classic Central API (we are actually on New Central but I believe there are certain API calls that can only be made to Classic that we will probably need to use when we start migrating to AOS 10).

Guy

Hi Guy,

Do you have found your solution ? the code look... 

do you have look pycentral (v1 ?) already make this stuff

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy

There is pycentral for Classic Central; and pycentral v2 for New Central.

While I have not played with it, pycentral v2 is supposed to be API compatible, so if you create your scripts now in pycentral, you should be able to migrate quite fast to the new Central.

Hello,

I understand better now that the calls for the initial authentication to create a session, and then the call to obtain an auth code, are rate limited. So once I manage to get my access token I just need to keep refreshing that (at least once every 15 days I think?).

I did look at PyCentral, but I think that is for New Central only? Or have I got that wrong? I am trying to use the Classic Central API (we are actually on New Central but I believe there are certain API calls that can only be made to Classic that we will probably need to use when we start migrating to AOS 10).

Guy

Hi Guy,

Do you have found your solution ? the code look... 

do you have look pycentral (v1 ?) already make this stuff

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy

Thanks Herman,

Do you know if Classic PyCentral offers any automation of token generation/refresh (ie management of tokens)? I've written a script now that will probably do the job of checking if a token is still valid and refreshing it if needed but it's not very pretty!

In general the jobs we are concerned about all relate to our migration from AOS 8, eg recreating the 300+ AP groups we have which I think currently can only be done via Classic Central. And things like pre-provisioning devices for when we start switching APs over to Central. I think we also need to configuire the AOS gateway firewall using API though I'm not sure if that is Classic Central or New Central.

I'm sure the list of things we need to do in bulk as part of the migration that would benefit from the API (Classic & New) will only grow.

Guy

There is pycentral for Classic Central; and pycentral v2 for New Central.

While I have not played with it, pycentral v2 is supposed to be API compatible, so if you create your scripts now in pycentral, you should be able to migrate quite fast to the new Central.

Hello,

I understand better now that the calls for the initial authentication to create a session, and then the call to obtain an auth code, are rate limited. So once I manage to get my access token I just need to keep refreshing that (at least once every 15 days I think?).

I did look at PyCentral, but I think that is for New Central only? Or have I got that wrong? I am trying to use the Classic Central API (we are actually on New Central but I believe there are certain API calls that can only be made to Classic that we will probably need to use when we start migrating to AOS 10).

Guy

Hi Guy,

Do you have found your solution ? the code look... 

do you have look pycentral (v1 ?) already make this stuff

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy

I believe that exactly is the purpose of using an SDK, to take care of token request,refreshing, etc... all the standard/low level things that you don't want to worry about, so you can focus on functionality.

For migration of AOS8 to AOS10 (and New Central configuration) there is quite good guides available as Validated Solution Guide.

If you work for a partner, you can click on the 'VSG Content for HPE Partners' on the VSG page and get even more detail. And you can get access to the VALID8 tool that can analyze your current config, and help migrating it to AOS10 in New Central configuration. And I know that VALID8 is expanded every now and then to support new use cases.

Thanks Herman,

Do you know if Classic PyCentral offers any automation of token generation/refresh (ie management of tokens)? I've written a script now that will probably do the job of checking if a token is still valid and refreshing it if needed but it's not very pretty!

In general the jobs we are concerned about all relate to our migration from AOS 8, eg recreating the 300+ AP groups we have which I think currently can only be done via Classic Central. And things like pre-provisioning devices for when we start switching APs over to Central. I think we also need to configuire the AOS gateway firewall using API though I'm not sure if that is Classic Central or New Central.

I'm sure the list of things we need to do in bulk as part of the migration that would benefit from the API (Classic & New) will only grow.

Guy

There is pycentral for Classic Central; and pycentral v2 for New Central.

While I have not played with it, pycentral v2 is supposed to be API compatible, so if you create your scripts now in pycentral, you should be able to migrate quite fast to the new Central.

Hello,

I understand better now that the calls for the initial authentication to create a session, and then the call to obtain an auth code, are rate limited. So once I manage to get my access token I just need to keep refreshing that (at least once every 15 days I think?).

I did look at PyCentral, but I think that is for New Central only? Or have I got that wrong? I am trying to use the Classic Central API (we are actually on New Central but I believe there are certain API calls that can only be made to Classic that we will probably need to use when we start migrating to AOS 10).

Guy

Hi Guy,

Do you have found your solution ? the code look... 

do you have look pycentral (v1 ?) already make this stuff

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy

central_info:   username: "<aruba-central-account-username>"   password: "<aruba-central-account-password>"   client_id: "<api-gateway-client-id>"   client_secret: "<api-gateway-client-secret>"   customer_id: "<aruba-central-customer-id>"   base_url: "<api-gateway-domain-url>" token_store:   type: "local"   path: "temp" ssl_verify: true 

central_info:   base_url: "<api-gateway-domain-url>"   token:     access_token: "<api-gateway-access-token>" ssl_verify: true

I believe that exactly is the purpose of using an SDK, to take care of token request,refreshing, etc... all the standard/low level things that you don't want to worry about, so you can focus on functionality.

For migration of AOS8 to AOS10 (and New Central configuration) there is quite good guides available as Validated Solution Guide.

If you work for a partner, you can click on the 'VSG Content for HPE Partners' on the VSG page and get even more detail. And you can get access to the VALID8 tool that can analyze your current config, and help migrating it to AOS10 in New Central configuration. And I know that VALID8 is expanded every now and then to support new use cases.

Thanks Herman,

Do you know if Classic PyCentral offers any automation of token generation/refresh (ie management of tokens)? I've written a script now that will probably do the job of checking if a token is still valid and refreshing it if needed but it's not very pretty!

In general the jobs we are concerned about all relate to our migration from AOS 8, eg recreating the 300+ AP groups we have which I think currently can only be done via Classic Central. And things like pre-provisioning devices for when we start switching APs over to Central. I think we also need to configuire the AOS gateway firewall using API though I'm not sure if that is Classic Central or New Central.

I'm sure the list of things we need to do in bulk as part of the migration that would benefit from the API (Classic & New) will only grow.

Guy

There is pycentral for Classic Central; and pycentral v2 for New Central.

While I have not played with it, pycentral v2 is supposed to be API compatible, so if you create your scripts now in pycentral, you should be able to migrate quite fast to the new Central.

Hello,

I understand better now that the calls for the initial authentication to create a session, and then the call to obtain an auth code, are rate limited. So once I manage to get my access token I just need to keep refreshing that (at least once every 15 days I think?).

I did look at PyCentral, but I think that is for New Central only? Or have I got that wrong? I am trying to use the Classic Central API (we are actually on New Central but I believe there are certain API calls that can only be made to Classic that we will probably need to use when we start migrating to AOS 10).

Guy

Hi Guy,

Do you have found your solution ? the code look... 

do you have look pycentral (v1 ?) already make this stuff

Hello,

I'm sure this must be something I'm doing wrong, but I can't figure out what. I am trying to work out the authentication/authorization process to use the Classic Central API. If I use cURL it works fine as per the online instructions, so I am now trying to turn that into a Python script. But after very few attempts (like 2 or 3 it seems) at running the script I get a rate limit exceeded message. This is my script so far (with contributions from ChatGPT! Needs must!):

def get_auth_classic(cust_id,client_id,api_url,username,pwd):    headers = {            "Content-Type": "application/json"            }    payload = {            "username": username,            "password": pwd            }    params = {            "client_id": client_id            }    url = "https://apigw-ukwest2.central.arubanetworks.com/oauth2/authorize/central/api/login"    # Session object can deal with cookies (equivalent to cURL --cookie-jar)    session = requests.Session()    response = session.post(            url,            params=params,            headers=headers,            json=payload,            allow_redirects=True # ?            )    print(f"response = {response.text}")    # Extract cookies    cookies = session.cookies.get_dict()    print("Cookies:", cookies)    # Extract CSRF token    csrf_token = cookies.get("csrftoken")    # Extract session ID    session_id = cookies.get("session")    if not csrf_token:        raise Exception("CSRF token not found in cookies")    print("CSRF Token:", csrf_token)    token_url = f"{api_url}/oauth2/authorize/central/api"    params = {        "client_id": client_id,        "response_type": "code",        "scope": "all"        }    data = {        #"grant_type": "authorization_code",        "customer_id": cust_id    }    headers = {        "Content-Type": "application/json",        "X-CSRF-Token": csrf_token    }    token_response = session.post(        token_url,        params=params,        json=data,        headers=headers,        allow_redirects=False   # ?    )    print("\nToken status:", token_response.status_code)    print("Token response:", token_response.text)    return token_response

The first login part seems to work, I get back the csrf token and session ID, but as I say, when I tweak the script to try to get the second part to work (obtain the actual re-useable token) I'm constantly blocked by having hit the rate limit and having to wait for that to clear, so testing is taking a long time! I _think_ it's the login action which is hitting that limit.

Is there something I'm doing wrong here that is causing this? I couldn't find Python examples for the auth process online.

Guy