add draft config for opera-rtc-s1-browse service #1
Conversation
|
I've confirmed that we can successfully generate browse imagery with worldview metadata. |
| @@ -225,7 +225,7 @@ LOG_STDOUT=false | |||
| # a variable name of GIOVANNI_ADAPTER_IMAGE you would specify giovanni-adapter (converting to lowercase | |||
| # and dash case and dropping _IMAGE. Make sure if the image for the service is not publicly available | |||
| # that you have built the docker image locally, otherwise the service will fail to start. | |||
| LOCALLY_DEPLOYED_SERVICES=harmony-service-example | |||
| LOCALLY_DEPLOYED_SERVICES=harmony-service-example,opera-rtc-s1-browse,hybig | |||
There was a problem hiding this comment.
We'll revert the change to this line before we merge into nasa/harmony
There was a problem hiding this comment.
The Harmony team points out that we can override any of these settings in our local .env file to facilitate local development, so we can avoid committing one-off changes like this in the future.
| env: | ||
| <<: *default-turbo-env | ||
| STAGING_PATH: public/asf/opera-rtc-s1-browse | ||
| umm_s: 'TODO' |
There was a problem hiding this comment.
We'll need to create a UMM-S service record in CMR UAT and CMR Prod and update this field before merging into nasa/harmony.
| @@ -23,6 +23,35 @@ x-turbo-config: &default-turbo-config | |||
|
|
|||
| https://cmr.earthdata.nasa.gov: | |||
|
|
|||
| - name: asf/opera-rtc-s1-browse | |||
There was a problem hiding this comment.
We'll want to create another copy of this service under https://cmr.uat.earthdata.nasa.gov before merging into nasa/harmony.
| @@ -511,3 +511,9 @@ NET2COG_LIMITS_CPU=128m | |||
| NET2COG_LIMITS_MEMORY=512Mi | |||
| NET2COG_INVOCATION_ARGS='./docker-entrypoint.sh' | |||
| NET2COG_QUEUE_URLS='["ghcr.io/podaac/net2cog:SIT,http://sqs.us-west-2.localhost.localstack.cloud:4566/000000000000/net2cog.fifo"]' | |||
|
|
|||
| OPERA_RTC_S1_BROWSE_IMAGE=ghcr.io/asfhyp3/opera-rtc-s1-browse:test | |||
There was a problem hiding this comment.
We'll want to update the image label here (and below in SERVICE_QUEUE_URLS) before merging to nasa/harmony. I imagine we'll used a fixed version, e.g. 1.0.0, and use that for both the UAT and Production configs. Then the deployment workflow is ASF releases a new container, PR goes into nasa/harmony to update the version number, change to nasa/harmony is deployed to harmony UAT and validated, then change is deployed to harmony Prod.
There was a problem hiding this comment.
Per the Harmony team, the image tags here are not the source of truth for what's deployed to each Harmony environment. They're only the image that is deployed the very first time a service is added to an environment. After that teams can manage the specific tag deployed to a specific environment via the Harmony service-image-tag api
Leaving this open as a draft PR into ASFHyP3/harmony:main while we work. We'll eventually merge it into nasa/harmony:main when we're ready to integrate with Harmony UAT and Harmony Prod.