EC2 maintenance updates-Summary and effective experience sharing

October 11, 2014

After the 11th holiday, we would like to report some summary information about EC2 maintenance updates and some effective measures to improve system availability. In September 30, we restarted less than 10% of EC2 instances to prevent any security risks related to xen security notification (XSA-108.

According to the relevant security notification disclosure mechanism of xen, we are obligated to comply with the confidentiality requirements of relevant issues until October 1 until it is published to the public. The xen community (we are also an active participant) designed a two-phase information disclosure process, as shown below:

• Prior to the widespread disclosure of relevant information, the information should be pre-disclosed to specific members (This list is maintained and regularly evaluated by the xen security team based on a series of public standards established by the xen project community ), so that specific members can take measures and implement updates within a certain period of time.
• The disclosure date is completely disclosed to all

Because we ensure the security of our customers is our top priority, and this problem may cause potential harm to our customers, we need to take immediate action to protect them. For the above reason, we must keep confidential the specific reason for such behavior before the featured disclosure date.

All the retries in different regions have been successfully completed as planned. We have maintained close communication and collaboration with our customers to ensure smooth restart arrangements.

We will continue to pay close attention to security issues and will do our best to protect all AWS customers from being affected by similar issues. As an AWS user, you can also take this opportunity to review your AWS architecture and find possible ways to enhance the availability of your system.

Here are some suggestions for your reference:

· Deploy instances in more than two zones

· Pay attention to your inbox and pay attention to alerts in the AWS Management Console ). Make sure that you fill in the relevant information in the "alternate contact" on the AWS billing console.

· Use trusted advisor to check your system architecture and submit relevant cases through AWS Premium Support when implementing system architecture best practices.

· Use the chaos monkey (chaos monkey) to detect various possible faults in a controlled environment.

· Check and consider expanding your use of Amazon route 53 and elastic load balancing to ensure that network traffic can be processed by healthy running instances.

· Use auto scaling to ensure the normal operation of a certain number of healthy instances.

In addition, you can refer to our security practices White Paper to learn more about AWS and security.

EC2 maintenance updates-Summary and effective experience sharing