tabletmanager/grpctmclient: use vtrpcpb/gRPC error codes

[timvaillancourt]

Description

This PR causes tabletmanager RPCs to return real gRPC/vterrors error codes, to address the Unknown error code tabletmanager RPCs currently return for problems executing an RPC. Now tablets with downed-mysqld will return the gRPC Unavailable code with the error message unchanged, for example

To achieve this, the existing switch statement in tabletserver's convertErrorCode(err error) error func was moved to be a method of *sqlerror.SQLError: VtRpcErrorCode() vtrpcpb.Code. I did not name this .ErrorCode() because this would be confusing, as sqlerrors.ErrorCode is a type that is used for something else entirely

This has been tested in non-prod environment at Slack and things look good 👍

Before w/mysqld killed + StopReplicationAndGetStatus RPC (code = Unknown 👎)

rpc error: code = Unknown desc = TabletManager.StopReplicationAndGetStatus on REDACTED-TABLETALIAS: before status failed: net.Dial(/mnt/vitess/mysql/datadir/mysql.sock) to local server failed: dial unix /mnt/vitess/mysql/datadir/mysql.sock: connect: connection refused (errno 2002) (sqlstate HY000)

After w/mysqld killed + StopReplicationAndGetStatus RPC (code = Unavailable 👍)

rpc error: code = Unavailable desc = TabletManager.StopReplicationAndGetStatus on REDACTED-TABLETALIAS: before status failed: net.Dial(/mnt/vitess/mysql/datadir/mysql.sock) to local server failed: dial unix /mnt/vitess/mysql/datadir/mysql.sock: connect: connection refused (errno 2002) (sqlstate HY000)

and VTOrc (FullStatus RPC based) sees: (code = Unavailable 👍)

DiscoverInstance(REDACTED-TABLETALIAS) instance is nil in 0.007s (Backend: 0.002s, Instance: 0.005s), error=rpc error: code = Unavailable desc = TabletManager.FullStatus on REDACTED-TABLETALIAS: net.Dial(/mnt/vitess/mysql/datadir/mysql.sock) to local server failed: dial unix /mnt/vitess/mysql/datadir/mysql.sock: connect: connection refused (errno 2002) (sqlstate HY000)

🎉

Related Issue(s)

Partially addresses: #18528

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

❌ Patch coverage is 53.33333% with 63 lines in your changes missing coverage. Please review.
✅ Project coverage is 67.51%. Comparing base (2fb4fa7) to head (e2a92d2).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
go/vt/vttablet/grpctmclient/client.go	64.55%	28 Missing ⚠️
go/mysql/sqlerror/sql_error.go	22.85%	27 Missing ⚠️
go/vt/vttablet/tabletmanager/rpc_server.go	37.50%	5 Missing ⚠️
go/vt/vttablet/tabletserver/throttle/throttler.go	57.14%	3 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main   #18565   +/-   ##
=======================================
  Coverage   67.51%   67.51%           
=======================================
  Files        1607     1607           
  Lines      263548   263567   +19     
=======================================
+ Hits       177941   177958   +17     
- Misses      85607    85609    +2     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[timvaillancourt]

It turns out vterrors.ToGRPC is required for mapping Vitess errors -> gPRC codes. The other RPC servers are calling this, but not tabletmanager before this PR

[timvaillancourt]

Flattening (de-wrapping) this error should be fine, because nothing else is interested in the error chain before it becomes a gRPC-style error in the following lines

[timvaillancourt]

cc @arthurschreiber as this is something we discussed at an offsite (onsite?), also in #18528

[shlomi-noach]

@timvaillancourt please notice a couple commits are unsigned, we will have to force-push

[timvaillancourt]

@timvaillancourt please notice a couple commits are unsigned, we will have to force-push

@shlomi-noach thanks again!

I squashed my commits with a proper signing now, so DCO is resolved ✅

Also, I had to modify some throttler client code slightly to make tests pass, after baking vterrors.FromGRPC into grpctmclient. CI should be passing now, but I'd like to re-request your review on those changes - I believe I've made it backwards compatible with pre-v23 tablets 🙇

[timvaillancourt]

@shlomi-noach oh wow, the GitHub auto-merge feature did not consider that I re-requested your review

Looks like it merged this with 1/2 reviews (Harshit = 👍, you = re-requested), technically 👎

Please let me know if you see any problems with what was merged and I'll follow up with a PR quickly 🙇

[shlomi-noach]

@timvaillancourt that's fine. GitHub-wise, the way to do it correctly is to "dismiss" my review before re-requesting it.
I think the fix is correct.

Re: TODO for v24, would you please create an Issue and assign yourself, so that we have this tracked?

[timvaillancourt]

@timvaillancourt that's fine. GitHub-wise, the way to do it correctly is to "dismiss" my review before re-requesting it. I think the fix is correct.

Thanks!

Re: TODO for v24, would you please create an Issue and assign yourself, so that we have this tracked?

Sounds good, created that here: #18663 👍