A question about the GoodData.CN Container Edition - we'd like to use this to push declarative layouts at via the Python SDK however we've found that doing this causes the GoodData local docker instance to fall over and become unresponsive. Has anyone had success at using the docker image as a local development space befor migrating layouts over to another instance (say on cloud)?

Hi Cam, I’d have to check internally to see if what you want to do here is possible. In the meantime, would you mind sharing what kind of API calls or commands you are making that lead to the docker instance becoming unresponsive? Do you receive any error messages?

It throughs a 503 error through nginx. I am starting to suspect that the Python SDK 1.7.0 I'm using has compatibility issues with the 2.5.0 version of the docker image.

Hi Cam, You can find the compatibility versions in a nice table at the bottom of the documentation here: https://www.gooddata.com/docs/python-sdk/latest/installation/ Could you confirm the versions you are using? Also, could you describe in more details the steps you are taking before seeing the error and more precisely what the error says?

Cheers @Joseph Heun I've seen that compatibility chart before. We're using Python SDK 1.7.0 to be compatible with the latest features on our Cloud instance (we upgraded versions after the comments here) but I suspect the docker image isn't keeping up with what's in the cloud instance (interested in if this is something that might be looked at going forward as I think having a local development space outside of Cloud would be very valuable). I can get small calls from the SDK to localhost to complete but once we run a

load_and_put_declarative_workspaces

call to push a number of workspaces each with a number of dashboards, metrics, visualisations etc things start to timeout and the docker image fall over with a 503 error.

Hi Cam, Are you currently working with both products, please? GD Cloud and GD.CN? If this is the case, please b aware that CN version doesn’t get updated automatically with the Cloud one. it needs to be done manually, if needed. Please check our Upgrade Guides.

Hi @Ivana Gasparekova we are using Cloud only but would like a local development environment to test config as code changes before pushing them to our production Cloud instance which is what the repo I linked earlier seems to suggest we can do (https://github.com/gooddata/gooddata-git-workflow-example) - locally the docker instance falls over with a 503 error returning from Nginx, the instance restarts and 5-10mins later it's ready to accept API calls again. Very strange.

Hi Cam, apologies for the delay here! After discussing this one internally, I believe that the issue is most likely caused by the differences between GD.CN Container Edition 2.5.0 and the most recent GD Cloud; 2.5.0 uses the Python SDK 1.5, whereas you are on 1.7, which is compatible with the newest Cloud. The latest GD.CN Edition is 3.0.0, and it should be compatible with the newest SDK and GD Cloud. Have you tried updating your CN docker image to this version? 🙂

Do you mean CN or CN Container Edition 3.0.0? Also the 3.0.0 docker images was only made available 4 days ago 😉

Yes, you are right. There were lots of changes introduced recently. 😉 I assume that it would be best to check what’s currently best for your use case directly with your Account owner Greta Jakubovskaja.

Cheers @Ivana Gasparekova is there any release notes around this I can eyeball 🙂

Hey @Cam Findlay, As mentioned before, the best option is to contact @Greta directly, but sure thing - you can also check our Release Notes as well. Hope this helps a bit. 😉