|SQL Database 管理门户|
|Azure Active Directory|
|Access Control 服务|
|Visual Studio Online|
|Visual Studio Application Insights|
|Microsoft Azure 预览版门户|
Approximately from 23:00 Sept 2 2015 UTC, Engineers received reports that a very limited subset of customer might have experienced errors when trying to login to Azure Management Portal. (https://manage.windowsazure.com). The recommended workaround is to use new Azure Preview portal (https://portal.azure.com). Our support team continue to work with the impacted customers directly.
From 3 Sept, 2015 21:06 to 22:05 UTC, customers in Australia East, Australia Southeast regions, Southeast Asia and East Asia may have experienced intermittent connectivity issues to Azure hosted services. SQL Database and Azure Active Directory may be impacted by this connectivity issue. Engineers have validated all network connectivity are back to normal and the incident is now fully mitigated.
From 01 Sep, 2015 16:00 to 18:43 UTC a subset of customers using SQL Database in West US, East US, and North Central US sub-regions may have experienced errors while performing service management or “Get” operations. Existing SQL environments would not have been impacted. This incident has now been mitigated.
From 27 Aug, 2015 02:50 to 06:30 UTC customers using Azure Active Directory in Multiple Regions may have experienced timeout issues when authenticating. Retries will have succeeded in some cases. Engineers identified a problematic Active Directory Queue, and rerouted traffic to resolve this. This incident has now been mitigated.
The main Active Directory incident has been resolved, and authentication issues for additionally impacted services have now also been mitigated, as of 06:30 UTC.
From 27 Aug, 2015 00:33 to 03:30 UTC customers using SQL Database version 12.0 in West US may have experienced issues performing service management operations. Retrieving information about servers and databases through the Azure Management Portal may have resulted in an error or timeout. Server and database create, drop, rename, and change edition or performance tier operations may also not have completed successfully. Availability (connecting to and using existing databases) was not impacted. Engineers have implemented and verified a fix for this issue, and the incident has now been mitigated.
As early as 22:30 UTC on 25th Aug, 2015 to 01:45 UTC on 27th Aug, 2015, Event Hub customers deployed in West US experienced creation failures (control-plane) on new Service Bus Namespaces. This issue did not impact run time operations for active Event Hubs. Engineers have validated that full control-plane functionality has been restored to Event Hub customers in West US. This incident is now closed.
Visual Studio Online have received reports of some customers continuing to experience residual impact from the licensing issue on August 24th. Impacted customers will receive a downgraded "stakeholder" license when retrieving licenses. Engineers have deployed a mitigation for this issue as of 16:40 UTC on 26th Aug, 2015, and customers are encouraged to view additional information on http://aka.ms/vsoblog
From 14:34 UTC on 24 Aug, 2015 to ~00:30 UTC on 26 Aug, 2015 customers using Visual Studio Online may have experienced issues where valid licenses are being downgraded to stakeholder licences. Engineers have confirmed that all systems are back to normal and identified a potential cache-miss related to our licensing calls and refreshed the cache to provide updated license information for each of our users. Engineers are continuing to monitor the service actively for any licensing issues, should they reoccur. We understand that customers rely on VS Online as a critical service and apologize for any impact this incident caused. Additional detail can be found on the VSO Blog: http://aka.ms/vsoblog
Starting 23 Aug, 2015 from 21:13 to 22:10 UTC, Stream Analytics customers may have experienced significant latency for starting Jobs in North Europe. Engineers have validated that Job start times in North Europe are within normal ranges. This incident is now closed.