timckt
commented
3 days ago - For
MariaDB,MySQL,PostgreSQLwith larger than 399allocated_storageandgp2storage type
printf "%-40s %-18s %-12s %-12s %-15s %-45s %-25s %-25s\n" \
"DBInstanceIdentifier" "AllocatedStorage" "Engine" "StorageType" "Environment" "Namespace" "SlackChannel" "InfraSupport"
aws rds describe-db-instances \
--query 'DBInstances[?AllocatedStorage > `399` && (Engine == `postgres` || Engine == `mysql` || Engine == `mariadb`) && StorageType == `gp2`]' \
--output json | \
jq -r '.[] | [.DBInstanceIdentifier, .AllocatedStorage, .Engine, .StorageType, .DBInstanceArn] | @tsv' | \
while IFS=$'\t' read -r db_id storage engine storage_type arn; do
tags=$(aws rds list-tags-for-resource --resource-name "$arn" --output json)
namespace=$(echo "$tags" | jq -r '.TagList[] | select(.Key=="namespace") | .Value')
slack_channel=$(echo "$tags" | jq -r '.TagList[] | select(.Key=="slack-channel") | .Value')
infra_support=$(echo "$tags" | jq -r '.TagList[] | select(.Key=="infrastructure-support") | .Value')
environment=$(echo "$tags" | jq -r '.TagList[] | select(.Key=="environment-name") | .Value')
namespace=${namespace:-"N/A"}
slack_channel=${slack_channel:-"N/A"}
infra_support=${infra_support:-"N/A"}
storage_type=${storage_type:-"N/A"}
environment=${environment:-"N/A"}
printf "%-40s %-18s %-12s %-12s %-15s %-45s %-25s %-25s\n" \
"$db_id" "$storage" "$engine" "$storage_type" "$environment" "$namespace" "$slack_channel" "$infra_support"
doneThe below output is as of Oct 24 20:00, removed the email address in InfraSupport here for data privacy.
We have 14 DB as of Oct 24 20:00 in this case.
DBInstanceIdentifier AllocatedStorage Engine StorageType Environment Namespace SlackChannel InfraSupport
cloud-platform-22ccc42afd500eb2 830 postgres gp2 production hmpps-book-secure-move-api-production N/A *
cloud-platform-2d4c157062b20f5d 10000 postgres gp2 dev hmpps-activities-management-dev N/A *
cloud-platform-35c188a93c8a060d 500 mariadb gp2 development intranet-dev N/A *
cloud-platform-43f547bdfa94ce12 550 postgres gp2 stage hmpps-delius-alfresco-stage ask-probation-webops *
cloud-platform-59b5cf9e5de6b794 600 postgres gp2 development hmpps-dpr-fake-dps-service ask_dpr *
cloud-platform-5a6f44384e7e06e6 1491 postgres gp2 preprod hmpps-activities-management-preprod N/A *
cloud-platform-5b6e8eb8ba215b7b 1500 postgres gp2 prod hmpps-manage-adjudications-api-prod N/A *
cloud-platform-5c6377808488916a 512 postgres gp2 preprod offender-case-notes-preprod public_move-and-improve *
cloud-platform-6c035586d92ac925 2000 postgres gp2 preprod hmpps-manage-adjudications-api-preprod N/A *
cloud-platform-7520cc9334a6f330 16000 postgres gp2 prod visit-someone-in-prison-backend-svc-prod N/A *
cloud-platform-78407cd5fbd86ed5 750 postgres gp2 production hmpps-workload-prod N/A *
cloud-platform-a066adb9989d7f7c 1500 postgres gp2 preprod hmpps-delius-alfresco-preprod ask-probation-hosting *
cloud-platform-b6244d8c893e1d12 512 postgres gp2 prod offender-case-notes-prod public_move-and-improve *
cloud-platform-c6e5432ed45afd19 16000 postgres gp2 prod visit-someone-in-prison-backend-svc-prod N/A *```
TL;DR:
When bumping module, users will need to take action and set the
db_iopsvalue to be at least12000for below 14 + 8 = 22 DBsFor 14 postgres/mariadb
For 8 oracle
Background
We have cut the major
8.0.0release for RDS module and we are going to bump the modules across our platform.For all RDS DB engines except RDS for SQL Server, the baseline storage performance for gp3 volumes increases when the storage size reaches certain thresholds. This means that when user upgrade the module and switch to
gp3, they may need to adjust thedb_iopsvalue accordingly to ensure optimal performance, especially for RDS instances with large volumes.gp3Storage TypeWhen bumping module, users will need to take action and set the
db_iopsvalue to be at least12000for :Db2,MariaDB,MySQL,PostgreSQLwith larger than 399allocated_storageandgp2storage_type(We have 14 DB as of Oct 24 20:00 in this case.)Oraclewith larger than 199allocated_storageandgp2storage_type(We have 8 DB as of Oct 24 20:00 in this case.)The complete output are in below comment.
Proposed user journey
Approach
Which part of the user docs does this impact
Communicate changes
Questions / Assumptions
Definition of done
Reference
How to write good user stories