The topic contains the answers to the most frequent questions related to EPAM Cloud services.
This topic contains the following sections:
- How can I configure Cloud Services?
- How can I monitor my Zabbix server performance?
- Are there any restrictions on Zabbix Monitoring Services usage?
- What is Default Monitoring?
- What is the Log Aggregation Service?
- How can I start collecting logs from a VM?
- How do I access my Graylog Server provided by the Log Service?
- Can I choose a Chef Server for my project?
- Can you show an example of CI/CD? Is Jenkins the major component in it or is it Maestro?
- If I deploy Jenkins as a service and I have an issue with Jenkins, who supports that? Who is responsible for installing plugins into Jenkins?
- Is there a solution for the Git issue which occurs during an empty repository cloning via SSH?
- What additional configuration is required for using ATG as a Service?
- If there a way to add multiple instances to a schedule simultaneously?
- I didn't find the answer to my question here. Where should I address?
How can I configure Cloud Services?
The Auto Configuration Service, Cloud Monitoring Service, Load Balancer Service and Log Aggregation Service can be managed via the Manage Services Wizard on the Cloud Management Console as well as via the Maestro CLI.
Other services are managed via the Maestro CLI. Please see the Services Guide for details.
How can I monitor my Zabbix server performance?
When the Monitoring Service is set up, you can see the details about the server performance on the Monitoring page of Orchestration Management console.
Are there any restrictions on Zabbix Monitoring Services usage?
Zabbix Monitoring service is currently not available for EPAM-KZ1 and AWS-* regions.
There is also a set of recommendations that apply to Zabbix Monitoring service usage:
- For the correct Zabbix Monitoring service performance, it is recommended to add the custom image-based instances to the monitoring list (or2-start-monitoring command) only after they change to the running state.
- It is recommended not to stop or reboot the machine with the project Zabbix server. This may cause unexpected behaviour.
What is Default Monitoring?
The Default Monitoring service is a service that does not need activation and provides you with the Chef Server statistics. The service is available on the UI Monitoring page.
Default Monitoring service is currently not available for EPAM-KZ1 and AWS-* regions.
What is the Log Aggregation Service?
Log Aggregation Service is a GrayLog-based service providing you with the ability to aggregate log information from the images in your infrastructure in one place. The GrayLog server is created on an automatically created and configured VM and can be easily accessed via HTTP connection.
How can I start collecting logs from a VM?
To make the GrayLog server collect the data on an instance, you should add this instance to the logging list. This can be done with the or2-start-logging (or2log) command:
or2log -p project -r region -i instance_id
By default, the GrayLog server gets the logs info from the event log files on Windows instances and SysLog files on Linux. To specify a custom log file, use the --log parameter with the or2log command. In this case, the data from both default and custom files will be collected.
How do I access my Graylog Server provided by the Log Service?
You can access the server via HTTP connection, using the DNS name of the Graylog Server VM. The server DNS name can be retrieved by or2audit command launched for the Maestro Stack used to launch the Log Service, or, when the monitoring list has at least one instance, you can get the server DNS with the or2-describe-logging (or2dlog) command.
The default user/password combination for logging in is user/graylog.
Can I choose a Chef Server for my project?
By default, the common Chef Server is used for all production environment machines. However, you can run Chef Server as a service for your project and select one of the available Chef Server Modes with the following command:
or2cm -m mode [default, epc, user] -p project -r region
- Default mode - the default mode for all projects in the EPAM Cloud. - EPC mode - use project-specific Chef server created by EPAM Orchestrator for the specified project. - User mode - use project-specific Chef server created and properly configured by the user. When switching to this mode, the user should provide Chef server instance ID (or instance IP) and manually upload validation.pem file to the Orchestrator file storage. The user should also provide the path to validation.pem file during the command invocation.
Can you show an example of CI/CD? Is Jenkins the major component in it or is it Maestro?
Yes, Jenkins is the main component of CI/CD in Maestro. Maestro provides all the necessary services and is a secondary component that manages all audit events related to Jenkins. On the other hand, when new services are activated for a project, EPAM Orchestrator can add new Jenkins jobs and install specific Jenkins plugins where necessary.
Currently Jenkins service is implemented in the basic configuration, with minimum number of additional functionality and maximum flexibility. At the moment, we are working on the service update according to the requests and suggestions we received from project teams.
If I deploy Jenkins as a service and I have an issue with Jenkins, who supports that? Who is responsible for installing plugins into Jenkins?
With most services, EPAM Orchestrator creates a VM for a project and configures necessary applications on it. All the default configurations are basic, and each project can customize the provided service for its needs.
The configurations provided by EPAM Orchestrator are checked and verified. As part of our services, EPAM Cloud Support team can create and support any specific customization to EPAM Orchestrator.
Is there a solution for the Git issue which occurs during an empty repository cloning via SSH?
Windows users setting up integration with Jenkins and Gerrit via EPAM Orchestrator sometimes face a Git issue while cloning empty repositories via SSH. When an empty repository is cloned, Git returns the "git clone could not fetch refs from %your_repo%" error message.
Currently, this issue can be resolved by using the following workaround: clone the repository via http (not SSH), then switch the remote origin from http to SSH URLs.
What additional configuration is required for using ATG as a Service?
For the ATG environment to work, create the Jenkins jobs which will compile, build and deploy the ATG application on the virtual machine which is to function as the ATG server. Additionally, create Jenkins jobs to start and stop JBoss on the same virtual machine.
Currently, ATG as a Service contains the components prepared for installing any ATG application. The related services are not automatically installed during the deployment of ATG as a Service. For example, the following services are not available automatically:
- Publishing (BCC)
- Agent (Service center)
- Front/Pages (Production Server)
- Staging server
- Loader (Reporting)
Install and configure the application components according to your requirements. You can install an application on the ATG server and perform the initial setup. Alternatively, you can install a demo CRS/CSA application on the VM and complete the configuration using CIM.
If the Loader server is configured, OBI (Oracle Business Intelligence) will not be automatically installed. OBI support is planned in subsequent releases of EPAM Orchestrator. Please follow the release announcements.
Activation of ATG as a Service includes start-up scripts for Oracle DB instances and Endeca service. However, for JBoss you need to create your own start-up script or use the appropriate Jenkins job.
If there a way to add multiple instances to a schedule simultaneously?
EPAM Orchestrator allows setting the same schedule for all instances in the same project and region. For that purpose, use the following command:
or2addsch -p project -r region -n stop -c "0 0 17 ? * 1-5" --all
To assign a schedule to a group of instances, use tags. Add the same tag to certain machines which you would like to work by the same schedule and then specify the tag in the schedule settings:
or2addsch -p project -r region -n stop -c "0 0 17 ? * 1-5" -t user:tag=QA
This command stops all instances with QA tag at 17:00 UTC every day from Monday till Friday
You can see whether a schedule affects individual instances, instances with tags or all instances in a region in the response to the or2dsch command. The "type" column shows schedule types as "INSTANCE", "REGION" or "TAG" depending on their application:
These schedule operations are available only in Maestro CLI.
I didn't find the answer to my question here. Where should I address?
If this section does not contain the answer to your question, please, address our Consulting Team.