How to reuse existing Selenium browser session

Suppose you are running a long Selenium test. The test runs fine but fails towards the end as some element was not found or was not clickable. Ideally, you would like to fix the locator and check it immediately. Now you wonder if you can reuse your browser session from the same point where the failure occurred. This post shows you how to do just that! This would help you to debug the locators used quickly, instead of running your whole test again.

Extract session_id and _url from driver object

We are taking advantage of the fact that Selenium lets you attach a driver to a session id and an URL. You can extract your session id and _url from the driver object as mentioned below

session_url = driver.command_executor._url  
session_id = driver.session_id

Tie it up with your framework

In case you are using a framework, you can print out the session id and url in the method where you try to get the element. In an example shown below, I am using a get_element method which would return the DOM element. I am printing the session id and url whenever the find_element method fails

def get_element(self,xpath):
        "Return the DOM element of the xpath or 'None' if the element is not found "
        dom_element = None
        try:
            dom_element = self.driver.find_element(xpath)
        except Exception,e:
            self.write(str(e),'debug')
            self.write("Check your locator-'%s"%xpath)
            #Print the session_id and _url in case the element is not found
            self.write("In case you want to reuse session, the session_id and _url for current browser session are: %s,%s"%(driver.session_id ,driver.command_executor._url))
 
        return dom_element

Use the session id and url

Now you can use the session id and url, in order to debug the locators. Open python command line tool and use the below code

 
from selenium import webdriver
 
#Pass the session_url you extracted
driver = webdriver.Remote(command_executor=session_url,desired_capabilities={})
 
#Attach to the session id you extracted
driver.session_id = session_id
 
#Resume from that browser state
elm = driver.find_element_by_xpath("xpath")
 
elm.click() #Perform required action

Note1: Currently I am able to reuse the session extracted only from Chrome browser. For Firefox there are some Issue. Link 1 and Link 2

Note2: This code is present in our open-sourced Python-Selenium test automation framework based on the page object pattern. We strongly recommend you check it out!

If you are a startup finding it hard to hire technical QA engineers, learn more about Qxf2 Services.

Related posts:

Default ThumbnailObtaining BrowserStack screenshots and video links Default ThumbnailSelenium to get browser console log Default Thumbnailpytest: Cross browser, cross platform Selenium tests
Avinash Shetty

38 thoughts on “How to reuse existing Selenium browser session”

  1. I am trying to implement this however i am not able to find the solution can anyone help me with this

    Reply

  2. Hello, thank you very much for the post, I have followed the steps to be able to reuse the browser, I am using selenium 4.19.0
    options = Options()
    options.add_argument(“–disable-notifications”)
    options.add_argument(‘–no-sandbox’)
    options.add_argument(‘–disable-dev-shm-usage’)
    options.add_argument(‘–enable-logging’)
    options.add_argument(‘–disable-blink-features=AutomationControlled’)
    options.add_experimental_option(“detach”, True)
    driver = webdriver.Chrome(options=options)

    print(“ID: “, driver.session_id )
    print(“URL:”, driver.command_executor._url)

    When I get the id and the url I use it to reuse the browser with the following, (Example http://localhost:58822 is what driver.command_executor._url gives)

    driver = webdriver.Remote(command_executor=’http://localhost:58822′, options=options)
    driver.close()
    driver.quit()
    driver.session_id = session_id

    But it is giving me the following error (I have tried with localhost, 127.0.0.1, local ip)

    Ocurrió un error: HTTPConnectionPool(host=’127.0.0.1′, port=58822): Max retries exceeded with url: /session (Caused by NewConnectionError(‘: Failed to establish a new connection: [Errno 61] Connection refused’))

    Has it ever happened to you? I take back my thanks

    Reply

    1. Hi,
      driver.close() & driver.quit() statements before assigning the old session id to the new driver could be causing the exception. Can you remove those statements and try?

      Reply

    2. Thanks for your response, I tried to delete it

      driver.close()
      driver.quit()

      and I still get the same error 🙁

      Reply

      2. Of course the complete error is

        HTTPConnectionPool(host=’localhost’, port=50974): Maximum retries exceeded with URL: /session (caused by NewConnectionError(‘: Could not establish a new connection: [ Errno 61] Connection refused’))

        the steps as explained above

        first I generate the browser like this

        options = Options()
        options.add_argument(“–disable-notifications”)
        options.add_argument(‘–no-sandbox’)
        options.add_argument(‘–disable-dev-shm-usage’)
        options.add_argument(‘–enable-logging’)
        options.add_argument(‘–disable-blink-features=AutomationControlled’)
        options.add_experimental_option(“detach”, True)
        driver = webdriver.Chrome(options=options)

        print(“ID: “, driver.session_id )
        print(“URL:”, driver.command_executor._url)

        then I get that session id and url to connect to the existing browser with another function

        driver = webdriver.Remote(command_executor=session_url, options=options)
        driver.close()
        driver.quit()
        driver.session_id = session_id

      3. Hi Roberth,
        Can you please make sure your script has similar code to save a session details to use it further to reconnect to the existing browser:

        session_url = driver.command_executor._url
        session_id = driver.session_id

        And Could you give it a try with replacing webdriver.Remote to RemoteWebDriver .
        Also please reorder driver.session_id = session_id line above driver.close() line.

      4. Thank you for your response, yes I have reviewed it and only in the provided code I made the assignments and also adjusted the following code

        driver.session_id = session_id
        driver.close()

        Could it be because of the version of selenium that I am using?

        the version is
        selenium==4.19.0

  3. Hello, sorry for the recent response, I have tried the code that you have provided me and it is not the same, I still get the same error.

    urllib3.exceptions.MaxRetryError: HTTPConnectionPool(host=’localhost’, port=59311): Max retries exceeded with url: /session/892e8bacd4d8674ee98a72981e39987d/screenshot (Caused by NewConnectionError(‘: Failed to establish a new connection: [Errno 61] Connection refused’))

    It will be the version of python, I use Python 3.9.6.

    and again I reiterate my thanks for trying to find the error

    Reply

    1. Hi,
      I notice a disparity in the URI endpoint against which you have hit HTTPConnectionPool exception.
      In the first snippet the error is seen against /session endpoint and in the second snippet it is against /session/$sessionId/screenshot endpoint.

      These endpoints are hit at two different method calls. Calling driver.start_session method makes a POST request against /session endpoint and calling driver.get_screenshot_as_png makes a GET request against /session/$sessionId/screenshot.

      The error in the second snippet you have attached urllib3.exceptions.MaxRetryError: HTTPConnectionPool(host=’localhost’, port=59311): Max retries exceeded with url: /session/892e8bacd4d8674ee98a72981e39987d/screenshot (Caused by NewConnectionError(‘: Failed to establish a new connection: [Errno 61] Connection refused’)) proves that the session is being reused?

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *