-
Notifications
You must be signed in to change notification settings - Fork 1
Development Infrastructure Starting and Stopping
Our AWS development best practices expect that the development infrastructure is not left to run when not actively being used. Specifically weekends, but also overnight (especially if utilizing any expensive resources). This is largely for cost concerns as we can minimal the cost of our infrastructure by not only shutting down resources, but removing them when not in use.
The purpose of this document is to walk through building the infrastructure using terraform, shutting down the infrastructure in a manner that allows for recreation with minimal to no loss of data.
git clone https://github.com/sul-dlss/terraform-aws.git
cd terraform-aws
git checkout rialto-dev
git pull
cd organizations/development/rialto
terraform init
terraform plan
Note: If this is an initial startup or startup after a shutdown, you should see well over 100 resources in the "To add" category.
terraform apply
Note: It is common to see errors in terraform that are related to dependency between resources and modules. Attempt to run terraform apply
more than once to try to clear errors.
Terraform will error out if the final-snapshot-identifier
already exists.
- Log into AWS Console
- Select RDS
- Click Snapshots
- Select Snapshot (
rialto-development-rds-hold
) - Actions > Delete snapshot
- Select Neptune
- Click Snapshots
- Select Snapshot (
rialto-development-neptune-hold
) - Actions > Delete snapshot
skip_final_snapshot = false
...
final_snapshot_identifier = "rialto-development-rds-hold"
TBD
terraform destroy
- Log into AWS Console
- Select RDS
- Select Instance
rialto-development-rds
- Create Snapshot
- Instance actions
- Take snapshot
- Enter a snapshot name (i.e.
rialto-development-rds-hold
)
- Select Neptune
- Select Instance
rialto-dev
- Create Snapshot
- Instance actions
- Take snapshot
- Enter a snapshot name (i.e.
rialto-development-neptune-hold
)
TBD
- RIALTO Wiki Homepage
- RIALTO Use Cases
- RIALTO Architecture
- RIALTO Data Models
- RIALTO Acceptance Criteria
- RIALTO Data Sources
- Demo Videos
- Neptune/λ Integration
- Core/Combine Integration
- SPARQL Proxy λ
- Derivatives λ
- Entity Resolver Service
- Rebuild Trigger Task
- Solr Setup
- Ingest Service
- Combine Data Sources
- Data Mappings
- Load Procedure
- Starting & Monitoring ETL
- Counting # of Publications
- Jena/TDB vs Blazegraph
- Vitro Ingest Options
- VIVO/Vitro Assessment
- VIVO Community Convo Notes
- Vitro vs Stand-Alone Datastore
- Provisioning a VM
- Deployment Process
- Toggle inferencing
- Check Inferencing is On
- Recompute inferences
- Toggle indexing
- Working with Vitro Solr
- Vitro Solr Samples
- Ingest via Fuseki SPARQL-over-HTTP
- Ingest via Jena ARQ
- Ingest via Jena tdbloader
- Ingest via Vitro SPARQL-over-HTTP
- Ingest via TDB Java API
- Vitro Logging
- Detecting TDB Changes