[flagd] add reconnect e2e tests

[toddbaert]

Implement the "reconnect e2e test suite" here by adding it to the existing e2e test suite.

How to:

• start ghcr.io/open-feature/flagd-testbed-unstable and ghcr.io/open-feature/sync-testbed-unstable containers (these restart every 5s)
• run the gherkin suite about twice (once for in-process and once for the RPC provider)

See the Java implementation PR here.

[NeerajGartia21]

@toddbaert I think you have missed the link to the Java implementation PR. Can you please update it?

[toddbaert]

@toddbaert I think you have missed the link to the Java implementation PR. Can you please update it?

Updated! Thanks!

[NeerajGartia21]

@toddbaert please update the link to the flagd-testbed to add flagd-reconnect.feature in the repo.

[toddbaert]

@toddbaert please update the link to the flagd-testbed to add flagd-reconnect.feature in the repo.

I'm sorry, I don't understand your request.

Do you mean update the git submodule? If so, you should be able to do that in a branch.

[NeerajGartia21]

Yeah, I mean the git submodule.
I'll update that in the PR.

[toddbaert]

Sure, just cd into the submodule and do a checkout (disregard if it gives you "detatched head" errors). If you have trouble, let me know and I can do it and then you can rebase on it.

[NeerajGartia21]

@toddbaert I tried adding reconnect e2e tests but I'm having a trouble debugging the issue with the PROVIDER_ERROR callback. The ProviderErrorCallback is not being called when connection breaks. Here is my implementation if you have time please have a look into it.

[Kavindu-Dodan]

@NeerajGartia21 sorry for the delay on checking back on this.

I checked your code and I think root cause is not defining providerErrorCallback function as a global function. Can you attempt that and see if this get resolved ?