Service Health

SUMMARY: Beginning Tuesday 19 August 2014, Cloud SQL users were unable to connect from hosts outside the Google network for a duration of 9 hours 53 minutes. We apologize for this service disruption. We are reviewing this incident in order to improve our service so we better uphold the high level of reliability we aim to provide.

DETAILED DESCRIPTION OF IMPACT: From Tuesday 19 August 2014 18:00 until Wednesday 20 August 2014 3:53 US/Pacific users who attempted to connect to their Cloud SQL instances from remote hosts outside of the Google network experienced failures to connect with the following message: "ERROR 2013 (HY000): Lost connection to MySQL server at 'reading initial communication packet', system error: 0". Connections from Google Compute Engine and Google App Engine were unaffected. This issue impacted 8% of Cloud SQL instances active during that time.

ROOT CAUSE: An update to Google infrastructure to widen usage of an internal authorization system resulted in some Cloud SQL traffic receiving a new classification. This traffic was then rejected as the new classification was not allowed by an access control check.

REMEDIATION AND PREVENTION: To resolve the issue Google engineers updated Cloud SQL to process the traffic with the new classification in the same way as the old classification. This change ended the outage. Google engineers are adding testing of connectivity to Cloud SQL from outside of Google, which will ensure issues of this nature are found and fixed in a much shorter time.

The problem with Cloud SQL connectivity from external networks was fully resolved as of Wednesday, 2014-08-19 05:32 AM US/Pacific. We apologize for the inconvenience and thank you for your patience and continued support. Please rest assured that system reliability is a top priority for the Google Cloud Platform, and we are making continuous improvements to make our systems better.

Cloud SQL connectivity from external networks has now been restored for most users. We are continuing to work on resolving this issue for the remaining users. We will provide next update by Wednesday, 2014-08-19 06:00 AM US/Pacific with current status.

We have identified the root cause behind the Google Cloud SQL connectivity issue and are are currently working on resolving it. We will provide next update by Wednesday, 2014-08-19 04:00 AM US/Pacific with current status.

We are currently experiencing an issue with Google Cloud SQL connectivity where some users can't connect to their Cloud SQL instances from external networks. Connectivity from Google App Engine or Google Compute Engine is not affected. For everyone who is affected, we apologize for any inconvenience you may be experiencing. We will provide next update by Wednesday, 2014-08-19 03:00 AM US/Pacific with current details.