If the management console hangs on login or pages stop responding, start by clearing the browser cache and cookies. Chrome and Firefox both accumulate stale session data that breaks the AWS UI. Try an incognito window — if that works, the issue is local cache. Also check that JavaScript is fully enabled; the console relies on it heavily.
Connection timeout to your instance usually means one of three things: the security group doesn't allow inbound traffic on port 22, the instance is in a stopped state, or the key pair used doesn't match. Go to EC2 → Security Groups and confirm port 22 is open for your IP. If you're using Elastic IP, make sure it's still associated after a restart.
Spikes in response time often come down to region selection. If your Lambda or EC2 is in us-east-1 but your users are in Europe, every call adds 80–120ms by default. Use CloudFront or pick a region closer to your load. Also check if you're hitting service quotas — throttled requests silently retry and inflate latency.
The most common cause is the security group attached to the RDS instance not allowing inbound connections from your application's subnet or IP. Check the VPC configuration and confirm the database is in a reachable subnet. If you recently changed instance type, the endpoint may have updated — grab the latest from the RDS console.
If cost alerts you configured in CloudWatch aren't firing, verify that billing alerts are enabled in Account Settings under the root account. This toggle is off by default and must be switched on before CloudWatch can track billing metrics at all.