Skip to content

Commit 82b13af

Browse files
authored
(docs) various updates (#1067)
* corrections - remove duplicate scp line - fix bad section reference in docs * add endpoint question to FAQ * Doc tweaks for 154 release
1 parent 2b80ec7 commit 82b13af

File tree

6 files changed

+69
-50
lines changed

6 files changed

+69
-50
lines changed

reference-artifacts/SCPs/ASEA-Guardrails-Sandbox.json

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,6 @@
77
"Action": [
88
"aws-marketplace:CreatePrivate*",
99
"aws-marketplace:AssociateProductsWithPrivate*",
10-
"aws-marketplace:CreatePrivate*",
1110
"aws-marketplace:DescribePrivate*",
1211
"aws-marketplace:DisassociateProducts*",
1312
"aws-marketplace:ListPrivate*",

reference-artifacts/SCPs/ASEA-Guardrails-Sensitive.json

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,6 @@
77
"Action": [
88
"aws-marketplace:CreatePrivate*",
99
"aws-marketplace:AssociateProductsWithPrivate*",
10-
"aws-marketplace:CreatePrivate*",
1110
"aws-marketplace:DescribePrivate*",
1211
"aws-marketplace:DisassociateProducts*",
1312
"aws-marketplace:ListPrivate*",

reference-artifacts/SCPs/ASEA-Guardrails-Unclass.json

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,6 @@
77
"Action": [
88
"aws-marketplace:CreatePrivate*",
99
"aws-marketplace:AssociateProductsWithPrivate*",
10-
"aws-marketplace:CreatePrivate*",
1110
"aws-marketplace:DescribePrivate*",
1211
"aws-marketplace:DisassociateProducts*",
1312
"aws-marketplace:ListPrivate*",

src/mkdocs/docs/faq/index.md

Lines changed: 60 additions & 41 deletions
Original file line numberDiff line numberDiff line change
@@ -346,6 +346,47 @@
346346

347347
Additionally, setting "populate-all-elbs-in-param-store": true for an account will populates all Accelerator wide ELB information into parameter store within that account. The sample PBMM configuration files set this value on the perimeter account, such that ELB information is available to configure centralized ingress capabilities.
348348

349+
??? faq "1.3.3. How do I deploy AWS Elastic Beanstalk instances?"
350+
351+
#### How do I deploy AWS Elastic Beanstalk instances?
352+
353+
If your deployed environment contains an SCP enforcing volume encryption of EC2 instances, your Elastic Beanstalk deployment will fail.
354+
355+
The SCP will contain an entry like this:
356+
357+
```json
358+
{
359+
"Sid": "EBS1",
360+
"Effect": "Deny",
361+
"Action": "ec2:RunInstances",
362+
"Resource": "arn:aws:ec2:*:*:volume/*",
363+
"Condition": {
364+
"Bool": {
365+
"ec2:Encrypted": "false"
366+
}
367+
}
368+
},
369+
370+
```
371+
A solution is to encrypt the root volume of the AMI that Elastic Beanstalk uses for your selected platform, and perform a custom AMI deployment of your Elastic Beanstalk application.
372+
373+
You can gather the AMI that Elastic Beanstalk uses via CLI with the following command:
374+
375+
```bash
376+
aws elasticbeanstalk describe-platform-version --region <YOUR_REGION> --platform-arn <ARN_EB_PLATFORM>
377+
```
378+
379+
Once you have gathered the AMI ID successfully, go to the EC2 console and:
380+
381+
- Click on the ‘AMIs’ option in the left navigation pane
382+
- Search for your AMI after selecting ‘Public Images’ from the dropdown list.
383+
- Select the AMI
384+
- Go to Actions and Copy AMI
385+
- Click on the checkbox to enable ‘Encryption’ and then select "Copy AMI".
386+
387+
Once the AMI is successfully copied, you can use this AMI to specify a custom AMI in your Elastic Beanstalk environments with root volume encrypted.
388+
389+
349390
## 1.4. Upgrades
350391

351392
??? faq "1.4.1. Can I upgrade directly to the latest release, or must I perform upgrades sequentially?"
@@ -873,47 +914,25 @@
873914

874915
![Logging](../installation/img/ASEA-Logging-Arch.png)
875916

876-
??? faq "1.6.17. How do I deploy AWS Elastic Beanstalk instances?"
877-
878-
#### How do I deploy AWS Elastic Beanstalk instances?
879-
880-
If your deployed environment contains an SCP enforcing volume encryption of EC2 instances, your Elastic Beanstalk deployment will fail.
881-
882-
The SCP will contain an entry like this:
883-
884-
```json
885-
{
886-
"Sid": "EBS1",
887-
"Effect": "Deny",
888-
"Action": "ec2:RunInstances",
889-
"Resource": "arn:aws:ec2:*:*:volume/*",
890-
"Condition": {
891-
"Bool": {
892-
"ec2:Encrypted": "false"
893-
}
894-
}
895-
},
896-
897-
```
898-
A solution is to encrypt the root volume of the AMI that Elastic Beanstalk uses for your selected platform, and perform a custom AMI deployment of your Elastic Beanstalk application.
899-
900-
You can gather the AMI that Elastic Beanstalk uses via CLI with the following command:
901-
902-
```bash
903-
aws elasticbeanstalk describe-platform-version --region <YOUR_REGION> --platform-arn <ARN_EB_PLATFORM>
904-
```
905-
906-
Once you have gathered the AMI ID successfully, go to the EC2 console and:
907-
908-
- Click on the ‘AMIs’ option in the left navigation pane
909-
- Search for your AMI after selecting ‘Public Images’ from the dropdown list.
910-
- Select the AMI
911-
- Go to Actions and Copy AMI
912-
- Click on the checkbox to enable ‘Encryption’ and then select "Copy AMI".
913-
914-
Once the AMI is successfully copied, you can use this AMI to specify a custom AMI in your Elastic Beanstalk environments with root volume encrypted.
915-
916-
917+
??? faq "1.6.17. Why are only select interface endpoints provisioned in the sample configuration files?"
918+
919+
#### Why are only select interface endpoints provisioned in the sample configuration files?
920+
921+
For economic reasons, most of the sample configuration files only include the following minimum set of required interface endpoints:
922+
923+
"ec2", "ec2messages", "ssm", "ssmmessages", "secretsmanager", "cloudformation", "kms", "logs", "monitoring"
924+
925+
The full sample configuration file included all interface endpoints that existed in the Canada (Central) region at the time the configuration file was originally developed:
926+
927+
"access-analyzer", "acm-pca", "application-autoscaling", "appmesh-envoy-management", "athena", "autoscaling", "autoscaling-plans", "awsconnector", "cassandra", "clouddirectory", "cloudformation", "cloudtrail", "codebuild", "codecommit", "codepipeline", "config", "datasync", "ebs", "ec2", "ec2messages", "ecr.api", "ecr.dkr", "ecs", "ecs-agent", "ecs-telemetry", "elasticbeanstalk", "elasticbeanstalk-health", "elasticfilesystem", "elasticloadbalancing", "elasticmapreduce", "email-smtp", "events", "execute-api", "git-codecommit", "glue", "kinesis-firehose", "kinesis-streams", "kms", "license-manager", "logs", "macie2", "monitoring", "notebook", "sagemaker.api", "sagemaker.runtime", "secretsmanager", "servicecatalog", "sms", "sns", "sqs", "ssm", "ssmmessages", "states", "storagegateway", "sts", "synthetics", "transfer", "transfer.server", "workspaces"
928+
929+
Since that time these additional endpoints have been launched in the ca-central-1 region and can be optionally added to customer configuration files to make them accessible from private address space:
930+
931+
"airflow.api", "airflow.env", "airflow.ops", "app-integrations", "appstream.api", "appstream.streaming", "auditmanager", "backup", "backup-gateway", "batch", "cloudhsmv2", "codedeploy", "codedeploy-commands-secure", "codestar-connections.api", "comprehend", "comprehendmedical", "databrew", "dms", "elasticache", "emr-containers", "finspace", "finspace-api", "fis", "fsx", "greengrass", "imagebuilder", "inspector2", "iot.data", "iot.fleethub.api", "iotsitewise.api", "iotsitewise.data", "kendra", "lakeformation", "lambda", "memory-db", "mgn", "models-v2-lex", "nimble", "panorama", "profile", "qldb.session", "rds", "rds-data", "redshift", "redshift-data", "rekognition", "runtime-v2-lex", "sagemaker.featurestore-runtime", "securityhub", "servicecatalog-appregistry", "ssm-contacts", "ssm-incidents", "sync-states", "textract", "transcribe", "transcribestreaming", "translate", "xray"
932+
933+
The aws.sagemaker.ca-central-1.studio interface endpoint was also launched, but cannot be auto-deployed by the Accelerator at this time as it does not utilize standardized naming and requires a code update to enable deployment.
934+
935+
Additional endpoints may exist in other AWS regions. Any endpoint can be added to any Accelerator configuration file, as long as it follows the standardized endpoint naming convention (e.g. com.amazonaws.{region}.{service}).
917936

918937
## 1.7. Network Architecture
919938

src/mkdocs/docs/installation/install.md

Lines changed: 8 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -215,7 +215,7 @@ Multiple options exist for downloading the GitHub Accelerator codebase and pushi
215215
- Do NOT download the code off the main GitHub branch, this will leave you in a completely unsupported state (and with beta code)
216216
3. Push the extracted codebase into the newly created CodeCommit repository, maintaining the file/folder hierarchy
217217
4. Set the default CodeCommit branch for the new repository to main
218-
5. Create a branch following the Accelerator naming format for your release (i.e. `release/v1.5.3`)
218+
5. Create a branch following the Accelerator naming format for your release (i.e. `release/v1.5.4`)
219219

220220
### 1.4.3. AWS Internal (Employee) Accounts Only
221221

@@ -318,7 +318,7 @@ If deploying to an internal AWS employee account and installing the solution wit
318318
- Add an `Email` address to be used for State Machine Status notification
319319
- The `GitHub Branch` should point to the release you selected
320320
- if upgrading, change it to point to the desired release
321-
- the latest stable branch is currently `release/v1.5.3`, case sensitive
321+
- the latest stable branch is currently `release/v1.5.4`, case sensitive
322322
- click `Next`
323323
7. Finish deploying the stack
324324
- Apply a tag on the stack, Key=`Accelerator`, Value=`ASEA` (case sensitive).
@@ -369,8 +369,9 @@ Current Issues:
369369

370370
- **NEW 2022-08-07** An issue with the version of cfn-init in the "latest" AWS standard Windows AMI will cause the state machine to fail during a new installation when deploying an RDGW host. RDGW hosts in existing deployments will fail to fully initialize if the state machine is or has been recently run and the auto-scaling group subsequently refreshes the host (default every 7 days).
371371

372-
- To temporarily workaround this issue, assume an administrative role in your `operations` account, open Systems Manager Parameter Store, and `Create parameter` with a Name of `/asea/windows-ami` and a value of `ami-0d336ea070bc06fb8` (which is the previous good AMI), accepting the other default values. Update your config file to point to this new parameter by changing `image-path` (under \deployments\mad) to `/asea/windows-ami` instead of `/aws/service/ami-windows-latest/Windows_Server-2016-English-Full-Base`. Rerun your state machine. If you have an existing RDGW instance it should be terminated to allow the auto-scaling group to redeploy it.
373-
- This config file entry should be reverted and state machine rerun once the next AWS Windows AMI is released (hopefully within the next week) to ensure you are always using the latest Windows AMI.
372+
- To temporarily workaround this issue, assume an administrative role in your `operations` account, open Systems Manager Parameter Store, and `Create parameter` with a Name of `/asea/windows-ami` and a value of `ami-0d336ea070bc06fb8` (which is the previous good AMI in ca-central-1), accepting the other default values. Update your config file to point to this new parameter by changing `image-path` (under \deployments\mad) to `/asea/windows-ami` instead of `/aws/service/ami-windows-latest/Windows_Server-2016-English-Full-Base`. Rerun your state machine. If you have an existing RDGW instance it should be terminated to allow the auto-scaling group to redeploy it.
373+
- In other regions you will need to lookup the previous working ami-id (you cannot use `ami-0d336ea070bc06fb8`)
374+
- This config file entry should be reverted and state machine rerun once the next (validated fixed) AWS Windows AMI is released to ensure you are always using the latest Windows AMI. **NOTE: Issue still exists 2022-10-05.**
374375

375376
- If dns-resolver-logging is enabled, VPC names containing spaces are not supported at this time as the VPC name is used as part of the log group name and spaces are not supported in log group names. By default in many of the sample config files, the VPC name is auto-generated from the OU name using a variable. In this situation, spaces are also not permitted in OU names (i.e. if any account in the OU has a VPC with resolver logging enabled and the VPC is using the OU as part of its name)
376377

@@ -382,7 +383,7 @@ Current Issues:
382383

383384
Issues in Older Releases:
384385

385-
- New installs to releases prior to v1.3.9 are no longer supported.
386+
- New installs to releases prior to v1.5.4 are no longer supported.
386387
- Upgrades to releases prior to v1.3.8 are no longer supported.
387388

388389
## 1.7. Post-Installation
@@ -391,6 +392,8 @@ The Accelerator installation is complete, but several manual steps remain:
391392

392393
1. Enable and configure AWS SSO in your `home` region (i.e. ca-central-1)
393394

395+
- **NOTE: AWS SSO has been renamed to AWS IAM Identity Center (IdC). The IdC GUI has also been reworked. The below steps are no longer click-by-click accurate. An update to the below documentation is planned, which will also include instructions to delegate AWS IdC administration to the Operations account enabling connecting IdC directly to MAD, rather than through an ADC.**
396+
394397
- Login to the AWS Console using your Organization Management account
395398
- Navigate to AWS Single Sign-On, click `Enable SSO`
396399
- Set the SSO directory to AD ("Settings" => "Identity Source" => "Identity Source" => click `Change`, Select Active Directory, and select your domain from the list)

src/mkdocs/docs/installation/upgrades.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -19,7 +19,7 @@
1919

2020
- Upgrades to `v1.5.1-a and above` from `v1.5.0` or `v1.5.1`:
2121
- Do not add the parameter: `"ssm-inventory-collection": true` to OUs or accounts which already have SSM Inventory configured or the state machine will fail
22-
- Follow the standard upgrade steps detailed in section 3.2 below
22+
- Follow the standard upgrade steps detailed in section 1.3 below
2323
- `v1.5.1` was replaced by v1.5.1-a and is no longer supported for new installs or upgrades
2424
- Upgrades to `v1.5.0` and `v1.5.1-a and above` from `v1.3.8 through v1.3.9`:
2525
- We recommend upgrading directly to v1.5.1-a

0 commit comments

Comments
 (0)