Forge tunnel uses ngrok - which is blocked by corporate network

Forge
,
1

Good day,

While following the Atlassian Developer guide (hello world confuence) - it’s required that Docker is installed & the command forge tunnel is run.

This forge tunnel command uses ngrok which is unfortunately where the problem lies since any tunnel process is blocked by our security team:

image
image1259×412 67.3 KB

Can anyone provide insight into getting this process to work while satisfying the security constraints?

Since I’m a novice in this area (my expertise is predominantly front-end development) & the Atlassian Development ecosystem, please provide clear steps on how to go about resolving the issue.

Thank you for your time & responses.

2

Hi @DylanWhittakerdigiou. Welcome to the developer community. If tunneling is restricted by your corporate network security policies, it is unlikely that you’ll be able to benefit from the use of forge tunnel (that is without trying to come up with clever ways to circumvent your security policies, which isn’t recommended).

That being said, while forge’s tunnel feature is useful for quick debugging and testing without having to redeploy manually, it isn’t necessary for you to develop, debug, or test your Forge app.

3

Thanks for quick response @nmansilla ,

Can you expand more on this:
“it isn’t necessary for you to develop, debug, or test”

I.e. for the layman developer, what’s the recommended development flow in order to refrain from redeploying the app every 10min to test code, cloud api requests, etc?

4

what’s the recommended development flow in order to refrain from redeploying the app every 10min to test code, cloud api requests, etc?

You’ll have to run forge deploy each time you make a code change in order to test on an instance. There’s no way around that, unless you’re using forge tunnel. I would recommend reaching out to IT/security to see if they can make a specific use-case exception.

5

Thanks @nmansilla ,

Our security team is investigating further to look at what can be done regarding the use of ngrok.

It would be great to hear from other members of the community that may have experienced the same issue and perhaps their (or their companies) method of navigating around it.

Maybe I’m being dramatic - but I must say the process of redeploying (which takes 2-5min) for every 5-15min of writing code means that I don’t really want to do Atlassian development anymore :smiley: - hopefully Atlassian can address this scenario in the future.