This is what we do in the implementation phase:
Configure Remote Access. Set up remote access from WSO2 to your Amazon EC2 instances.
Set up the environments (e.g., Development, Test, Pre-Production, and Production).
Commit the artifacts such as scripts, diagrams, and documents to the repository for versioning and history.
Hand over the Production environment to you with WSO2 Carbon user accounts that have admin privileges to access the Management Consoles.
Note that WSO2 can facilitate the following upon your request:
- Arrange a third-party consultant to carry out penetration tests.
- Provide reports and dashboards on the Production environment.
- Arrange trainings and workshops for you. <Link to a training page.>
- Provide read-only access to the monitoring dashboards and the log viewer.
Configure remote access
WSO2 does all Managed Cloud deployments in an Amazon Virtual Private Cloud (Amazon VPC). A VPC enables you to launch Amazon Web Services (AWS) into a virtual network that you define. A VPC improves the security of your data by providing network-level control and isolation for the AWS. You can keep your data and configurations in a private space and expose them through the DMZ. T his virtual network closely resembles a traditional network, but with improved security and scalability.
To set up your Cloud environments, WSO2 requires access to your Amazon EC2 instances. We access these instances over SSH only, with a Bastion host working as the SSH gateway. The Bastian host can either reside in the VPC or in your own datacenter. The diagrams below depict both scenarios.
Bastian host in the VPC
The Bastion host is in the public subnet and allows SSH traffic only to the WSO2 network via a non-standard port. All other hosts are configured to accept SSH requests from the Bastion host only.
Bastian host in your datacenter
The Bastion host is in your datacenter, and the other hosts are configured to accept SSH requests from the Bastion host only. When WSO2 DevOps want to connect to the Bastion host via SSH, they do it remotely via a client console.
In addition to the AWS instances, WSO2 requires access to the following resources:
|Need access to||Purpose||Prerequisites|
|AWS management console||To access and manage your AWS.|
WSO2 needs separate user accounts with the following form you :
|AWS API service||To execute automated tools to bring up the infrastructure services such as the VPC, network setup, databases etc.|
WSO2 needs the following from you:
Set up the environments
- Things to be done from both the customer-end and the MC-team-end
- What we can do from an MC PoV Vs things that require Dev Services and Delivery sort of support
- Firewall access that we need and what are the specific services that we require in network architecture terminology. For example, DNS, SMTP and NTP services should be explained just like we have explained the access via Bastion nodes.>
Implement monitoring and alerting
WSO2 hosts all monitoring services in a separate subnet in the same VPC where your Cloud services are hosted. We configure Nagios Remote Plugin Executor (NRPE) in all Linux hosts to monitor the resource utilization and set thresholds. If any resource gets utilized beyond a certain threshold, or if an application isn’t responding properly, NRPE triggers alerts and notifications.
We collect statistics about resource utilization (i.e., disk, CPU and memory utilization, JVM heap usage etc.) and a pplication health. All statistics collected via the NRPE agents are presented using ICinga , the monitoring and dashboard tool. We also configure all Linux hosts with Simple Network Management Protocol (SNMP) and host the st atistics that are collected via SNMP using Cacti , the the network graphing solution. All statistical d ashboards are exposed only to the WSO2 network over HTTP/S. To communicate with the third-party services required to extend alerts, all monitoring hosts need to have Internet connectivity. However, this doesn't mean that the monitoring hosts are placed in the public subnet.
W SO2 requires an email server with SMTP Authentication enabled to send direct email alerts and notifications to other servers. If the customer cannot provide an email server, WSO2 uses Amazon Simple Email Service (SES). We need support from the customer to verify the domain and set up DomainKeys Identified Mail ( DKIM ) , which is an email validation system designed to detect email spoofing.
We maintain application logs using the Logstash, ElasticSearch and Kibana solutions. WSO2 configures a Logstash agent in each host to collect application data and send over to ElasticSearch that is running on the monitoring host in a different subnet in the same VPC. The Kibana dashboard is exposed only to the WSO2 network over HTTP/S.
The monitoring and alerting implementation is depicted in the diagram below:
If the customer wants to synchronize their monitoring with that of WSO2, DevOps teams from both sides need to agree on certain technical requirements such as additional agents that must be installed on hosts, how to expose dashboards to other networks, send alerts to additional email addresses, phones, etc.
Implement backup and disaster recovery
<coming up soon>
Commit the artifacts
<Info about where the repository location is, what we mean by artifacts etc.>
Hand over the environments
WSO2 is to hand over the Production environment to you with WSO2 Carbon user accounts that have admin privileges to access the Management Consoles.
<A checklist of items? >
Next, go to Support and Maintenance.