Rozdíly

Zde můžete vidět rozdíly mezi vybranou verzí a aktuální verzí dané stránky.

--- skoda-ep:amazon-aws [16.02.2016 21:25]
Jan Šimůnek
+++ — (aktuální)
@@ Řádek 1: / Řádek 1: @@
-====== ŠKODA Engagement AWS Infrastructure ======
-This page contains information about Engagement platform infrastructure running
-on Amazon AWS.
-====== Infrastructure deployment diagram ======
-{{ :skoda-ep:skoda-ep-amazon_infrastructure.png?600 |}}
-Infrastructure is hosted on Amazon AWS. We use EC2 instances and several other services
-(RDS, S3, etc.).
-====== EC2 instances ======
-EC2 instances represents virtual machines for running core components. We use different instance configurations. We use autoscaling ability of Amazon AWS. Instances are located in Frankfurt datacenter.
-See [Amazon EC2 Instance types](http://aws.amazon.com/ec2/instance-types/) for more info.
-===== Operation systems =====
-All instances running Linux Ubuntu 14.04, AMI ID: ubuntu-trusty-14.04-amd64-server-20140927 (ami-b83c0aa5).
-===== Login to instances =====
-Login to servers is possible only by using SSH with private keys. Username and password login is disabled.
-====== Core components ======
-===== Web proxy =====
-Web proxy with NGINX handling all of incoming traffic to infrastructure. Traffic is forwarded to web servers based on proxy configuration.
-===== Web server =====
-Web server based on EC2 instance with autoscaling feature. NGINX handling incoming requests. Static resources requests are handled by NGINX directly. Requests to PHP sources are forwarded to PHP-FPM via linux socket.
-==== PHP engine ====
-We use PHP FPM engine to process PHP sources. See [http://php-fpm.org/](http://php-fpm.org/) for more info. Installed by Ansible.
-===== Amazon RDS =====
-We use Amazon RDS for all MySQL instances. RDS offers perfect out of the box solution for MySQL hosting including scalability, backup, security, etc.
-===== Amazon =====
-We use Amazon S3 service for storing uploaded content (images, movies, multimedia content, etc.). We plan to use CDN to serve this content to visitors.
-===== Mandrill (SMTP) =====
-Mandrill is used for sending e-mails as SMTP service. We use this service to send emails from EC2 instances and running applications including installed Wordpress CMS. There is postfix installed on every EC2 instance with relay configuration to Mandrill.
-Mandrill is used for newslettering also.
-====== Integrated systems ======
-There is no integration to other external systems now, especially ŠKODA AUTO internal IT.
-====== Ansible provisioning ======
-Infrastructure is provisioned (installed and maintained) by Ansible. Ansible is similar to Chef or Puppet. But it's
-more simple, with Python and SSH requirements only. See the official [documentation](http://docs.ansible.com/)
-for more info.
-====== Security ======
-We use a lot of different services to make a whole infrastructure secure. Details about security settings are private.
-====== Support ======
-There is information about infrastructure support parameters (SLA) in the following tables.
-===== SLA =====
-| Parameter                |                  |
-| Support Time:            | Mo-Su 8:00-18:00 |
-| Servicedesk :            | Mo-Su 8:00-18:00 |
-| Response time:           | 10 min           |
-| Uptime:                  | 95%              |
-| Fix time (by priority*): | 2h/8h/-h         |
-===== Priority and fix time =====
-| Priority  | Description | Fixtime |
-| Critical  | Website is down, does not accessible, not work at all. There is real seacurity problem, data lost, etc. | 2h |
-| High      | Some parts, functions of website does not work, cannot be used. | 8h |
-| Low | Some minor, not important functions of site does not work properly, there are some issues in backend, administration part, editors functions, etc. | will be specified during reporting the issue |
-===== Contact information =====
-**YOUR SYSTEM, s.r.o.** \\
-WWW: http://helpdesk.ys.cz \\
-E-mail: servicedesk@ys.cz (jan.simunek@ys.cz) \\
-Phone: +420 277 775 555 (in czech language only)
-===== Monitoring =====
-We use Amazon AWS monitoring tools (status check alarms, cloud watch, etc.) with notifications to ensure fix time as short as possible.