[vtcombo] Expose --tablet_types_to_wait flag

[arthurschreiber]

Description

At GitHub, we use vtcombo in our CI system and our local development environments.

While investigating slow CI setup times, I noticed that vtcombo by default is starting 3 tablets (PRIMARY, RDONLY and REPLICA) per keyspace/cell. For a small topology, that's not really an issue, but we run multiple hundreds of keyspaces in CI, and just starting this setup takes almost a minute.

This is quite wasteful, especially because we don't use RDONLY and REPLICA tablets at all in CI or local development. To improve this, we can set replicaCount to 1 to only start a single REPLICA tablet (which will be promoted to primary), and rdonlyCount to -1 to start no RDONLY tablets.

Unfortunately, vtcombo will start vtgate and instruct it to wait for PRIMARY,REPLICA,RDONLY tablets to be available, which, based on the above configuration, won't ever happen. Eventually, the vtgate will run into a timeout while waiting and then start serving requests anyway.

Exposing the tablet_types_to_wait CLI flag allows setting the flag to e.g. PRIMARY only, which brings down the startup time for vtcombo to ~20 seconds in our setup, which is quite an improvement.

Related Issue(s)

N/A

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: 11 lines in your changes are missing coverage. Please review.

Comparison is base (696fe0e) 67.41% compared to head (de43b28) 67.48%.
Report is 28 commits behind head on main.

Files	Patch %	Lines
go/cmd/vtcombo/cli/main.go	0.00%	11 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #14951      +/-   ##
==========================================
+ Coverage   67.41%   67.48%   +0.06%     
==========================================
  Files        1560     1561       +1     
  Lines      192752   193343     +591     
==========================================
+ Hits       129952   130485     +533     
- Misses      62800    62858      +58     

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

[GuptaManan100]

Rest LGTM, but I don't think we should change the definition of replica count wherein it includes the primary tablet as well.

[arthurschreiber]

I changed this to not change how replicaCount and rdonlyCount are treated - I believe the change to expose --tablet_types_to_wait is all I need. 👍

[frouioui]

I removed the NeedsIssue label as I think the PR description is pretty self-explanatory