microsoft / CloudAdoptionFramework

Code samples and extended documentation to support the guidance provided in the Microsoft Cloud Adoption Framework
https://aka.ms/CAF
Creative Commons Attribution 4.0 International
854 stars 614 forks source link

naming & tagging convention template - Naming convention inconsistencies #187

Closed paraenggu closed 3 months ago

paraenggu commented 1 year ago

First of all thanks a lot for providing the Naming & Tagging standard template!

While studying the template, I noticed the following naming inconsistencies:

  1. In the "intro and components" sheet, virtual machines are prefixed with vm-, while in the "Naming convention" sheet the format and example is missing the dash (vm<policy name or appname><###> / vmsharepoint001)
    • Also the question arises whether the VM number (<###>) wasn't supposed to be separated by a dash as with most other assets.
  2. In the "Naming convention" sheet under Virtual Machines the format and the example for the entries of the Load Balancer and NIC do not match (dash)
    • lb-<app name or role><Environment><###> / lb-sharepoint-dev-001
    • nic-<##>-<vmname>-<subscription><###> / nic-02-vmhadoop1-prod-001

General question: In the "intro and components" sheet most resource name prefixes include a dash (-) as the last character, while some are lacking one (e.g. stor, ssimp, dls, dla and pbiemb). Do they use a different schema intentionally and if yes, is there a technical limitation or similar to it, which could be noted as a comment in the sheet? The same question goes for the <Environment> separation of those assets in the "Naming convention" sheet.

Thanks!

shurik-io commented 3 months ago

In the "intro and components" sheet most resource name prefixes include a dash (-) as the last character, while some are lacking one (e.g. stor, ssimp, dls, dla and pbiemb). Do they use a different schema intentionally and if yes, is there a technical limitation or similar to it, which could be noted as a comment in the sheet?

I assume this is because of the character limit (24 IIRC) for storage based services, which have a global scope, thus saving some chars by not using the dashes in the naming scheme.

In general it would be nice to get answer on this topic, as I saw these stuff as well and had to fix it manually.

Zimmergren commented 3 months ago

Thank you for reporting this, @paraenggu. The linked file is over 4 years old, and drifted out of sync with the source of truth (CAF docs) - I would recommend checking out the official docs:

Are you relying on the Excel file, specifically, or would you rather be able to keep up to date with the official docs, which are updated and maintained regularly? If I can understand how the Excel file is being used, we might be able to justify looking at an update. The way we see it right now, we're more likely to discontinue the Excel file and only keep the docs.

cc @shurik-io

paraenggu commented 3 months ago

thanks for your feedback on this issue @Zimmergren

Are you relying on the Excel file, specifically, or would you rather be able to keep up to date with the official docs, which are updated and maintained regularly? If I can understand how the Excel file is being used, we might be able to justify looking at an update. The way we see it right now, we're more likely to discontinue the Excel file and only keep the docs.

I originally opened this issue after following Step 5: Establish naming and tagging standards across the portfolio of the foundational alignment, which links to this file:

The same page also links to the initial-decisions-checklist which also contains a link to this template.

image

I don't directly relay on it, but found it practical to have a consolidated document which helps one to define the naming and tagging conventions and gather inputs from the involved people.

It's certainly OK, to discontinue the file and refer to the online docs, but I guess others will also end there as long as it's linked on that page and within that document.

Zimmergren commented 3 months ago

Thank you for the added thoughts on this, @paraenggu . We're planning to retire the file, and we'll then also remove the references from the content - we'll get this done asap to avoid further confusion, and we'll rely on the docs as the source of truth.

Thanks! Tobias.