feat: migrate to production veda stac catalog and titiler AND update GHG URLs

[anayeaye]

Linked PR: NASA-IMPACT/veda-ui#1019

Changed

• Base STAC API and Raster API urls updated for production openveda.cloud
• GHG center STAC and Raster API urls updated to align with recent GHG path changes unrelated to openveda.cloud

Why are you creating this Pull Request?

To migrate the Dashboard usage of VEDA backend APIs to new production openveda.cloud VEDA instance.

How tested

Steps taken to validate the new production VEDA instance are documented here.

[netlify]

✅ Deploy Preview for earth-information-center ready!

Name	Link
🔨 Latest commit	2a16472
🔍 Latest deploy log	https://app.netlify.com/sites/earth-information-center/deploys/669544c0bcc8ca0008a2d193
😎 Deploy Preview	https://deploy-preview-120--earth-information-center.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[agurvich]

I don't have access to veda-architecture so I can't read the documentation to test.

[sandrahoang686]

@agurvich The checklist linked is more of prod readiness checklist the data services team uses to ensure the new endpoints are ready to go before the dashboard makes this update. The things we'd like to validate on our end though is to make sure anything referencing these endpoints are tested to make sure they still work as is. So looks like we should be testing the E&A feature like adding layers and running analysis to see if everything is still working as is 👍🏼

[anayeaye]

@agurvich sorry about that link to an inaccessible veda-architecture issue. The list was the fine grained details of how the data services team audited the production system after moving the COGs, database, and services to the new higher security AWS account.

We used a search of this repo, veda-config-eic, and veda-config to identify which data should be promoted to the new production account. After auditing for parity we are opening this PR to cut EIC over to the new production system (away from the staging system which will be persisted but is intended for more experimental data). Please reach out with any issues and let us know how we can support transitioning EIC to production!

EDIT: sorry missed @sandrahoang686's comment, she already covered it!

[agurvich]

Gotcha, thanks for explaining! So what am I responsible for testing here? I don't fully understand what changed for the end-user or what was touched in the backend.

[sandrahoang686]

@agurvich @aboydnw

We should make sure any datasets that may have recently been merged are in the prod DB
For EIC instance, lets make sure /analysis page works the same as before this change.

• Check Datasets/layers
• Run analyses and make sure everything runs as like before

[aboydnw]

I went through and manually explored the datasets and layers, and created some time series charts in the analysis page. The only issue I found was with the U.S. Gridded Anthropogenic Methane Emissions Inventory dataset. None of the layers loaded in the exploration page. I also see this on production, though, so maybe it just my machine, or a separate issue?

[anayeaye]

@aboydnw thanks for doing that manual exploration! It looks like the that methane emissions problem is related to a GHG dataset (not in openveda.cloud). I think the problem is actually caused by GHG layer URLs here https://github.com/NASA-IMPACT/veda-config-eic/blob/develop/datasets/epa-ch4emission-v2express.data.mdx?plain=1#L22 I think those were recently updated?

[sandrahoang686]

• Looks like analysis for U.S. Gridded Anthropogenic Methane Emissions Inventory dataset looks good to now in the deploy preview ✅
• Running Analysis on GEOGLAM Crop Conditions gave a 500 Internal Service Error. This does not seem to happen in production ❌ (Update: Sometimes, It happened 3 times for me and then I ran it again and seemed to work)

[anayeaye]

• Running Analysis on GEOGLAM Crop Conditions gave a 500 Internal Service Error. This does not seem to happen in production ❌ (Update: Sometimes, It happened 3 times for me and then I ran it again and seemed to work)

@sandrahoang686 and I have now both re-run the geoglam analysis in the preview successfully. It seems to be an ephemeral problem related to the preview, not the API