I am curious if anyone else has been experiencing problems in the last few days.  I have an Azure application with SQL in South Central US, and a user reported receiving some errors on Sunday, but I could find no problems in the SQL log which usually tells me when there was a problem.

Then today I uploaded a new instance of the application to the staging area and it could not connect to the SQL Azure at all - it was like the connectionstring was corrupt.  So I deleted it uploaded it again and it worked..or so I thought.

Another user had a similar error today, and it appears that one of the 2 instances works and the other does not - how can this be?

I am not quite sure how to debug this, and as it is only a recent problem which makes me think that I am probably not alone.

Has something changed recently? Anyone else have the same i

We've been experiencing the same issues off and on since Friday, 6/28. We have two cloud services in the same affinity group in US East that cannot connect to their SQL Azure db in US East.

We suspected it could be a firewall issue as non-cloud service clients can hit without issue. For example, our local development sites can access the same SQL Azure db without fail. Based on that assumption, we added our cloud service's public IPs to the firewalls rules for the db. It "appeared' to have helped, albeit, temporarily in Friday. Even with those rules, it is failing for us.

Our two cloud services are currently 'down' because of this issue.

Definitely experiencing similar issues here. It seems to be per-instance on our end though. Publishing a new version of the app resulted in failed connections to SQL. Randomly, re-imaging worked, but any future attempts to update re-introduced the problem

Can report that I can run the app locally with the Azure connection string without issue, so it's definitely an  issue with either the cloud service VMs themselves, or the network they operate on.

All East US, btw.

Can either of you confirm that this issue has gone away? I have published several updates to different apps this morning and have not had the issue yet.

We are continuing to experience the issue. We had a new build and deployment at about 30 minutes ago and it had the issue. The issue resolved on its own while I was replying to MS Azure technical support. It hasn't been as long lasting as the other days.

Your re-imaging trick has helped, though, can take many attempts for it to finally have any impact. As you said, it's somewhat random.

I am still getting more than usual emails about site issues but because I use my db for error tracking I cannot tell if it is related.  I have also re-imaged without any immediate problems, but I continue to monitor the situation.

I have not contacted tech support about this because it is impossible to demonstrate.  

I noticed yesterday at the height of the problem that my ELMAH emails weren't going out either, and there was a thread somewhere in the azure forums (can't find it now) about someone not having any access to outbound connections from his VMs. I'm wondering if the issue is a delayed firewall configuration or something - the problem definitely seems to resolve itself shortly after deployment (within 1-2 hours) when it does manifest. That said, I made at least 4 deployments today and did not experience the problem.