Sudden Database Connection Failures from .NET Workloads in Kubernetes
Asked by
MellowPine42
Our .NET applications running in Kubernetes clusters suddenly started failing to connect to the database. The database server is online, the databases are available, and we haven't deployed any application changes since yesterday. Has anyone else noticed a cloud service outage or a networking issue that could explain these connection failures?
Answers
There are several possible causes besides the database itself. Check whether Entra authentication is involved, whether the database uses an IP allowlist, and whether the outbound public IP changed in the affected region. Also review recent network-policy or automated tenant-policy changes. Connecting to one of the affected containers and running DNS, routing, and connectivity tests should help narrow it down.
Reply to this answer
The provider’s public status page may not show an incident immediately. It can take hours for an outage to be officially confirmed, so checking independent outage reports can sometimes provide an earlier signal. Health-status email alerts may arrive sooner, but they’re often delayed too.
MellowPine42: We may have identified the issue on our side and are working on a fix, so it might not be a broader service outage after all.
Reply to this answer
Add your answer
Answers are reviewed before publication.
MellowPine42: We found that our load balancers were degraded somehow, and we’re working on replacing or switching them.