Default region change to us-east-1

[jsmolar]

AMIs are created by default in us-east-1. The default region we use in Jenkins jobs is eu-west-1.
This PR changes the default region to match AMIs default location.

Testing

• No testing is needed

PR pre-checks (self review)

• I added the relevant backport labels
• I didn't leave commented-out/debugging code

Reminders

• Add New configuration option and document them (in sdcm/sct_config.py)
• Add unit tests to cover my changes (under unit-test/ folder)
• Update the Readme/doc folder relevant to this change (if needed)

[fruch]

@jsmolar why should we change it ? what's the problem with current situation ?

i.e. it can backfire, and leave us with not enough resources on us-east-1, if everyone by default would work on it.

[vponomaryov]

I agree with @fruch

At first, the problem is unclear. It doesn't matter in which region images get built at a first place, they, anyway, get copied to other supported regions.

Second, I faced problems with allocations exactly in us-east-1 much more than in other regions.

Third, where are the majority of people who work with test VMs? - Biggest part is in Europe and Israel which are much closer to eu VMs and, hence, have better ping connecting to such VMs.

[jsmolar]

@fruch @vponomaryov
During the summit, I heard some developers express frustration that when using BYO AMI, the image is built by default in the us-east-1 region (as indicated by the COPY_AMI_TO_REGIONS parameter in this Jenkins job). With the current default region set to eu-west-1, one developer ran into an issue where their AMI was not available in the default region, forcing them to rebuild the job.

This PR was intended as a proposal to address that minor inconvenience by aligning the default region with where the AMI is initially built.

That said, we could also consider removing the default region entirely or simply accept the inconvenience and leave things as they are.

[fruch]

@fruch @vponomaryov During the summit, I heard some developers express frustration that when using BYO AMI, the image is built by default in the us-east-1 region (as indicated by the COPY_AMI_TO_REGIONS parameter in this Jenkins job). With the current default region set to eu-west-1, one developer ran into an issue where their AMI was not available in the default region, forcing them to rebuild the job.

This PR was intended as a proposal to address that minor inconvenience by aligning the default region with where the AMI is initially built.

That said, we could also consider removing the default region entirely or simply accept the inconvenience and leave things as they are.

if that's the reason, we can think of better options that can help developers

• copy AMIs also in BYO jobs
• teach them to use the BYO option in SCT, that can make sure the AMI is build in correct place
• copy AMI as needed from us-east-1 in SCT code

I would suggest opening an issue to discuss it, we need stronger reasons for changing the default region.

[vponomaryov]

@fruch @vponomaryov During the summit, I heard some developers express frustration that when using BYO AMI, the image is built by default in the us-east-1 region (as indicated by the COPY_AMI_TO_REGIONS parameter in this Jenkins job). With the current default region set to eu-west-1, one developer ran into an issue where their AMI was not available in the default region, forcing them to rebuild the job.

This PR was intended as a proposal to address that minor inconvenience by aligning the default region with where the AMI is initially built.

That said, we could also consider removing the default region entirely or simply accept the inconvenience and leave things as they are.

It doesn't sounds as conceptual problem.

It is known limitation that AMIs are region-based.
BYO CI jobs allow to configure copying of images to other regions as you already referred to.
Moreover, SCT has BYO feature which utilizes mentioned CI job types and resolves this problem automatically allowing you to run SCT BYO CI jobs in any region.

The switch will lead to a single region resources overcommit failing more CI jobs than it is now and I consider it a bigger problem.