Currently Azure File sync is not able to sync with a file storage which activates firewall function (and Service endpoint).

Enterprise data often contains secret data. Therefore I need that file sync can be used with file storage which activates firewall function.

Enable CSP to be used to fund the Visual Studio Cloud Subscriptions.

On Azure AD Application Proxy Connectors “Internal URL”, we cannot set URL which includes “?” or doesn't end with “/”.
We need to connect to on-premises apps from Azure AD Application Proxy connectors via Internet,
then tried to set the URL which doesn't end with "/" or includes "?", but couldn’t.
We hope this will be improved quickly.

extend trial period

It would be great if I could deploy docker images without having to worry about setting up Docker hosts.

Currently, when VNet Peering between virtual network in resource manager model and classic deployment model, the same supscription for the VNets are reqired.
I hope we can freely configure VNet Peering between ASM and ARM with different subscriptions in the near future.

When I want to create a SQL Server VM in Azure and use a license I already own (Bring Your Own License) I am unable to use an existing VM Image in Azure as these are pre-licensed. It would be good if I could use an existing VM image and supply just my license key.

Machine Learning workbench requires Internet connection, but please also support connection via VPN and via Express Route.

Also, I want you to be able to use some functions even in an environment without internet connection.

----In Japanese---
Machine Learning workbench はインターネット接続が必須ですが、VPN 経由での接続やExpress Route 経由での接続もサポートしてほしい。

また、インターネット接続がない環境でも、一部の機能だけでも使用できるようにしてほしい。

France has many laws regarding datas security, and in some sector like healthcare, it's impossible to host stuff outside of France.

I'd be a very good news for us that Microsoft could open a datacenter in France :)

There should be an option to rename Load Balancer name in Azure Portal. We follow naming convention of each resources in Azure and there is high chance of typo's sometimes when configuring resources manually.

Recently I created a Load balancer for HA/DR solution and later realised there is typo in Azure Load Balancer while creation. So I have to delete the load balancer and create it again which caused some delay to setup whole solution. So i think there should be a feature to rename resource name in Azure after creation.

We would like to have the possibility to allow users who are not members of the local "Administrators" group of workstations to connect to the VPN, without having the need to give them those privileges.

Please add R Services (RevoScale) to ADLA such that I can use ADL in a compute context in the same way I can currently use SQL Server with R Services in a compute context; bringing the data science to the data and not the other way around.

Deletion of link of integrated account is operated from Azure Resource Explorer, However, I think that all resource management should be done from the portal. Please make it possible to delete link of integrated account from Azure portal as well.

<In Japanese>
統合アカウントのリンクの削除は、Azure Resource Explorerから操作しますが、
すべてのリソース管理はAzure ポータルで一元管理したい。
統合アカウントのリンクの削除もAzureポータルからできるようにしてほしい。

Presently, it is only possible to call a webhook from an Azure Alert by providing a URL, and thus only allowing the passage of a security key via a query string parameter. This is bad security practice since the URL can be logged in plain text on web servers. Please provide an option to add additional data to the body of a POST/PUT webhook call so we can pass sensitive information safely

If a customer has a tenant/subscription with the same partner who has the other tenant, you should allow the partner to do the transfer of directory.

For each loop in Azure Logic apps.
Currently, you can not set Break processing, but please implement the function so that break processing can be set up

<In Japanese>
Azure Logicアプリケーションの各ループについて

現在、Break処理は設定できませんが、Break処理を設定できるように実装してください

I want you to be able to transfer the learned model in Chainer and Tensorflow directly to Machine Learning.