Have  you got it fixed. If you have sorted the issue, can you update the answer here. In case, any of the comments were helpful, can you mark them as answer

If you have done inplace upgrade from 6.2, did you do crx2oak migration of the repository ? If not, please follow https://helpx.adobe.com/uk/experience-manager/6-4/sites/deploying/using/in-place-upgrade.html#content-repository-migrationPerforming an In-Place Upgrade . As mentioned in the document, y...

You can use oak run to check the connectivity faster. Below is the command for AWS S3, may need tweaking for azure (Jackrabbit Oak – Oak Run NodeStore Connection )Can you screen print your config (sanitised) and the install folder content (connector bundle etc)

If you are solution partner, there are brilliant materials via captivate courses. - solutionpartners.adobe.comhttps://experienceleague.adobe.com/#dashboardIf not, use experience league, recently launched - Experience League Intelligent Guidance | Adobe

1. Have you started AEM with crx3,crx3tar-nofds runmode ? crx3tar-nofds one is mandatory as default is fds. 2. If you check repository folder and if you see datastore, then you are still in fds and hence Azure cloud is not connecting.

As Jaideep mentioned, it's due to local storage of chrome getting full. Mostly, some javascript in the custom code might be running an infinite loop

Strange, I am getting a proper replication request on the default agent in author for each request in postmanhttp://localhost:4502/etc/replication/agents.author/publish.log.html#end Are you trying to invoke a custom replication agent ?

I was able to do without doing both the changes, I am using vanilla 6.4

Have you setup basic authentication itself in the authentication tab. After setting that one, it worked for me

A shared S3 datastore can make all the binaries to sit in the S3 rather than the instances. By setting the S3 configurations to make the nodestore to keep minimal data, you can make almost all coming from S3.