Startup Procedure
Startup NFS
Log into the NFS server. If you are running Ubuntu, use this command:
sudo systemctl start nfs-server
If you are using RHEL, use this command:
sudo systemctl start nfs
Startup Elasticsearch
Elasticsearch is configured to startup automatically when the Linux host is started. To confirm this, log into each Elasticsearch node and run:
sudo systemctl status node1_elasticsearch
If the output from the previous command showed that the service is inactive, run the following command on each host on which the service was inactive:
sudo systemctl start node1_elasticsearch
Once all Elasticsearch instances are active, move on to the next step.
Startup SQL Server
Since SQL Server is not configured or managed via the Connect & Guest install process, it is outside the scope of this document. However, it should be started after starting Elasticsearch and before starting Redis.
Startup Redis
If this is performed on a 1-server system, skip to step 3. The Redis Sentinel service is configured to startup automatically when the Linux host is started. To confirm this, log into each Redis node and run:
sudo systemctl status sentinel_26379
If the output from the previous command showed that the service is inactive, run the following command on each host on which the service was inactive:
sudo systemctl start sentinel_26379
The Redis service is configured to startup automatically when the Linux host is started. To confirm this, log into each Redis node and run:
sudo systemctl status redis_6379
If the output from the previous command showed that the service is inactive, run the following command on each host on which the service was inactive:
sudo systemctl start redis_6379
Once all Elasticsearch instances are active, move on to the next step.
Startup RabbitMQ
Log into the primary RabbitMQ node and run:
sudo rabbitmqctl start_app
Log into each of the remaining RabbitMQ nodes and run:
sudo rabbitmqctl start_app
Startup Services
Portainer
Access Portainer.
Click on the “primary” endpoint.
Go to the Services page.
Increase “Items per page” in the lower right-hand corner to “100” to ensure you see all services.
For each service with “replicated” Scheduling Mode, click “Scale”, change the number to the appropriate number, and click the checkmark. For a redundant production environment, the default number of instances is given in the table below. For a single-server environment, each service should be scaled to “1”. If a non-default number of instances is being used then the user is responsible for noting the appropriate number at the time of shutdown (Shutdown procedure section A above).
Services | Replicas |
|---|---|
infrastructure_portainer_agent | 3 |
infrastructure_cadvisor | 3 |
infrastructure_prune | 3 |
infrastructure_alertmanager | 3 |
infrastructure_prometheus | 1 |
auth-web | 3 |
infrastructure_busybox | 3 |
cac-api | 3 |
cac-endpoint | 3 |
cac-web | 3 |
infrastructure_cerebro | 1 |
core-api | 3 |
core-endpoint | 3 |
exporter_elastic | 1 |
infrastructure_grafana | 1 |
idm-api | 3 |
idm-endpoint | 3 |
idm-web | 3 |
picfit | 3 |
portal-web | 3 |
public-api | 3 |
exporter_rabbitmq | 1 |
Exporter_redis | 1 |
registration-api | 3 |
registration-endpoint | 3 |
registration-web | 3 |
registry_api (offline only) | 1 |
root-web | 3 |
scheduler-endpoint | 3 |
exporter_sql | 1 |
sso-web | 3 |
symmetry-api | 3 |
symmetry-endpoint | 3 |
vms-api | 3 |
vms-endpoint | 3 |
vms-web | 3 |
Command Line
If you shut down the services using the command line, go to the same directory where the previous commands were run. There should be a “amag_services” file there. Run the following command to start up all the services:
cat amag_services | xargs -l docker service scale
Smoke Test
It is advised at this point that you log into Connect/Guest and run through your organization’s primary workflows (e.g. create an identity for the new employee, login as the new employee, request/approve/process credential, request/approve access, generate reports, etc.).