Return SourceAlias in FullStatus RPC, use to optimize VTOrc

[timvaillancourt]

Description

This PR implements the change described in #17297

TL;DR: FullStatus RPCs now include the tablet-alias of the replication source (not just the hostname/port). This alias is probably useful in general, but for now VTOrc uses it to avoid an sqlite table-scan in inst.ReadInstanceClusterAttributes, which is a hot-path function

In internal testing we saw a large performance win by adding an index to that hostname, port read - this PR should be even better because it uses primary key reads and no additional index, because alias is the PK

Related Issue(s)

Closes #17297

Checklist

• "Backport to:" labels have been added if this change should be back-ported to release branches
• If this change is to be back-ported to previous releases, a justification is included in the PR description
• Tests were added or are not required
• Did the new or modified tests pass consistently locally and on CI?
• Documentation was added or is not required

Deployment Notes

[vitess-bot]

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

• Ensure that the Pull Request has a descriptive title.
• Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

• Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

• Apply the release notes (needs details) label if users need to know about this change.
• New features should be documented.
• There should be some code comments as to why things are implemented the way they are.
• There should be a comment at the top of each new or modified test to explain what the test does.

New flags

• Is this flag really necessary?
• Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

• Each item in Jobs should be named in order to mark it as required.
• If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

• Protobuf changes should be wire-compatible.
• Changes to _vt tables and RPCs need to be backward compatible.
• RPC changes should be compatible with vitess-operator
• If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
• vtctl command output order should be stable and awk-able.

[codecov]

Codecov Report

Attention: Patch coverage is 6.00000% with 47 lines in your changes missing coverage. Please review.

Project coverage is 67.44%. Comparing base (22988df) to head (1b3970f).
Report is 6 commits behind head on main.

Files with missing lines	Patch %	Lines
go/vt/vttablet/tabletmanager/rpc_replication.go	0.00%	28 Missing ⚠️
go/vt/vtorc/inst/instance_dao.go	13.63%	19 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main   #17843   +/-   ##
=======================================
  Coverage   67.44%   67.44%           
=======================================
  Files        1592     1592           
  Lines      258173   258252   +79     
=======================================
+ Hits       174126   174189   +63     
- Misses      84047    84063   +16     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[GuptaManan100]

I have one concern in this PR. Its regarding the atomicity of the change of sourceAlias with the mysql fields. Let's say we call SetReplicationSource and it fails at some point. How do we guarantee there is no drift between the source set in the mysql instances and the source alias we store.

If the mysql is incorrect, but sourceAlias has been set to the correct value, will this change make VTOrc not notice that the replication needs to be fixed?

[GuptaManan100]

We need to set these field in a few more places like initializeReplication and reset it in disableReplication

[GuptaManan100]

maybe clear this in defer if the function fails 🤔

[timvaillancourt]

I have one concern in this PR. Its regarding the atomicity of the change of sourceAlias with the mysql fields. Let's say we call SetReplicationSource and it fails at some point. How do we guarantee there is no drift between the source set in the mysql instances and the source alias we store.

If the mysql is incorrect, but sourceAlias has been set to the correct value, will this change make VTOrc not notice that the replication needs to be fixed?

@GuptaManan100 I agree, it sounded like only a few places needed this change but the areas keep growing. I think I'd feel more confident if there were unit tests for this code, and perhaps I should start there - that way we could prove the transitions worked

It doesn't change the problem much, but would the sourceAlias field make more sense in tm.tmState? It seems to be used inconsistently so I wasn't sure

[timvaillancourt]

Let's say we call SetReplicationSource and it fails at some point. How do we guarantee there is no drift between the source set in the mysql instances and the source alias we store.

Sorry, I didn't respond to this specific point earlier. I wonder this too. For now I've placed the setting of the new value at the very end of funcs when we are about to return, with the assumption that a failure would happen earlier