Problem when browserstack.exe gets updated

[TraGicCode]

It looks like everytime this library downloads a new version of browerstack.exe from amazon and then tried to run it it fails to connect to browserstack when the file is downloaded. Below is the message that the browserstack.exe command line spits out when it fails

Failed to connect to browserstack

Here is the command line arguments this library is formulating to start the executable

"C:\Users\<Myuser>\.browserstack\BrowserStackLocal.exe"  -d start <mykey> -onlyAutomate -forcelocal -force -logFile C:\Users\<Myuser>\AppData\Local\JetBrains\Installations\ReSharperPlatformVs14_000\local.log -daemonInstance```

[vedharish]

@TraGicCode Can you share the contents of the local.log file? C:\Users\AppData\Local...\local.log ?

[TraGicCode]

@vedharish Hee is what the log shows

BrowserStackLocal v6.5

---

Setting the --only-automate parameter will stop local testing via command line from working with BrowserStack Live and BrowserStack Screenshots.
To enable local testing via the command line for Live and Screenshots, rerun BrowserStackLocal without --only-automate.

---

---

Updating BrowserStackLocal...

Updated to 6.6

Updated Binary

Note: A new version of BrowserStackLocal v6.6, is available, but the auto-upgrade failed
due to a slow connection.

Please download the latest version of BrowserStackLocal from https://www.browserstack.com/local-testing#command-line

Continuing with existing version v6.5

You can now access your local server(s) in our remote browser.

Press Ctrl-C to exit

[TraGicCode]

@vedharish I have more information regarding our issue

If no browserstacklocal executable exists it will download the latest from s3. This works everytime perfectly!

if there is already an existing browserstacklocal executable that is outdated. It will download the latest from s3 and then try and run it and connect and throw an exception containing the folllowing message "failed to connect to browserstacklocal".

There has to be other reports on this or am i the only customer using this :/

[pulkitsharma07]

Hi @TraGicCode, There is a timeout of 100 seconds when the BrowserStackLocal binary updates itself. The update failed because of that. I do not think this is an issue with the C# bindings.
When you remove the BrowserStackLocal binary from that location, the C# BrowserStackLocal bindings will download the latest version and place it there. Since there is no timeout to download the fresh binary, it is downloaded without any errors.
Look like a network issue for me. Closing this for now.

[pulkitsharma07]

Reopening the issue for now, @TraGicCode are you still facing this ?

[TraGicCode]

I will get back to you in the morning on this!

[TraGicCode]

Good Morning,

Yes we are still facing this issue. We have ended up resorting to a nasty hack, which is not ideal. Before each run we are deleting the BrowserStackLocal.exe in order to force a fresh download, which works everytime.

Our server is located in AWS if that helps at all on your end.

Do you have any recommendations?

[pulkitsharma07]

Can you share the contents of the local log file again ?
it will be helpful to us if you can provide the following information too:

• By your server, do you mean the server from which you are running the tests from ? If this is the case, then before you delete the BrowserStackLocal.exe (which from my understanding will be present there) can you check the binary version ? It can be fetched by executing the following command in the command prompt: BrowserStackLocal.exe --version. The latest version as of now is 6.7.
• Do you have a setup script for your server(like Ansible) which downloads files before tests are run on the machine ? If that is the case, does it download the BrowserStackLocal.exe file too ? is that the latest version ?
• For subsequent tests, do you turn off the test server and start again (and delete BrowserStackLocal.exe as you mentioned) or the server is kept running ?

[gawicks]

Having this problem with version 6.9 . In my experience this only happens when BrowserStackLocal is run as a daemon , and when it tries to update to 7 it prints out this error.
But on on flip side the instance seems to update and run properly. So the error is spurious.

The only problem that I'm having with this is that my CI script seems to think that an error has occurred and exits prematurely.