Azure Workstation Deployment Role Fails If I Do Not Delete Previous Deployment

I usually had a miserable time deploying my work roles in Azure. I will publish my working role once from Visual Studio, and everything will work fine. I will publish the working role again later, and the deployment will fail. The instance goes into a "recycle cycle". I spend hours trying to understand what I broke. I tried Intellitrace, but it always fails with the error message "Unable to load error message." Then, in the end, I will remove the deployment from the inside of the Azure Management Portal and try again, and the same code that does not work for several hours will magically work.

This does not always happen, and some projects seem to “fix” themselves and stop showing this behavior together. But it seems that publishing from Visual Studio will fail if I do not manually delete the removal of the existing deployment.

I know this may be a little vague, but I really have nothing to do next. Intellitrace never works, and I cannot work with Remote Desktop because it is being processed so quickly (which could also be due to the fact that Intellitrace is not working).

Does anyone know what could be here?

+4
source share
1 answer

I have done more research, and I think I know what is going on. Obviously, Visual Studio is trying to upgrade your in-place work roles during deployment. If this fails, for some reason, for example, when changing the service configuration between deployments, she simply complains that something is wrong with your role and that your instance is being processed.

The deployment options have the option "If the deployment cannot be updated, complete the full deployment" which will remove the existing deployment and will be deployed from scratch if the existing deployment cannot be updated. I am not sure why this is not checked by default, and not "crash mysteriously."

+3
source

Source: https://habr.com/ru/post/1487798/


All Articles