I have 2 Web Apps (Azure web app) in the same service plan.
Those Web apps use virtual networks because they need to have a fixed and single outbound IP.

We have
WebApp1 using VirtualNetwork1.
WebApp2 using VirtualNetwork2.

The first Web app can access to myazuredatabaseserver.database.windows.net.
The second one cannot resolve.

The troubleshooting tool says that the DNS resolution is not OK.
They both use the default Azure DNS server.

Both virtual networks are configured the same way : they use routetables to route to one dedicated outbound IP, they use VNAT for the same purpose and they both have a specific subnet, with default usage, server farms and routetable in their config.

It has worked very well until this morning.
No change have occurred in their configuration before the problem appeared and I’m the only one admin for this tenant.

I don’t understand where this problem comes from.

I’ve rebuilt the virtual network from scratch : I’ve deleted all the components (vnet, vnat and routetable) and recreated it from scratch. The problem persists.

Everything is in the same region (france/central).

It may be interesting to note that the database server has the checkbox “allow azure resources…” checked.

What I’ve tried :

I’ve tried to rebuild the complete vnet.
I’ve tried to use the server IP instead of the endpoint.
I’ve tried to use other DNS severs
I’ve tried everything available in the vnet components configuration (vnet, vnat, routetable…)

I’ve read many many stackoverflow articles.

I need to have this problem solved because it occurs to my integration environment and all my devs are stuck.
I also need to understand what happened in case it occurs on another env (production, for exemple)and how to fix/prevent that.

Help would really be appreciated.

Does anyone have an idea ?

I can answer any question and provide logs and detail upon request.

Thanks in advance for your help !