scriptoriumadmin/GitNexus
1
0
Fork 0
mirror of https://github.com/abhigyanpatwari/GitNexus.git synced 2026-10-07 02:58:02 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 8

test(ci): rebalance Windows shards from measured slow suites

Browse source
This commit is contained in:
Abhinav Pandey 2026-09-09 06:24:33 +05:30
parent c9cb49038f
commit d3eef2e66d
No known key found for this signature in database
2 changed files with 40 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

22
gitnexus/scripts/cross-platform-shard.ts
View file

@ -36,9 +36,9 @@
*
* Only files heavy enough to matter are listed; everything else is carried by
* {@link PER_FILE_OVERHEAD_SEC} alone. These are load-balancing hints, NOT
* assertions — no
* test asserts a runtime, and drift only makes the split slightly less even, so
* a stale entry is harmless and refreshing them is optional. Deliberately not
* assertions — no test measures elapsed time against this table. Missing or
* stale heavy entries can still overload a shard; refresh them from failed
* CI logs and replay that profile in the partition tests. Deliberately not
* auto-generated: a committed table is reviewable and works offline, and the
* alternative (timing files at CI runtime to decide the split) would make the
* partition depend on the very machine load it is trying to protect against.
@ -49,13 +49,15 @@ export const WINDOWS_WEIGHTS_SEC: Readonly<Record<string, number>> = {
'test/integration/cli-e2e.test.ts': 621,
'test/integration/worker-pool.test.ts': 222,
'test/unit/incremental-vector-extension-ordering.test.ts': 87,
// ESTIMATE, not a measurement (#2841): this suite drives more full
// `runFullAnalysis` cycles than the VECTOR sibling above, so the 8 s
// PER_FILE_OVERHEAD floor would badly under-charge it and skew the Windows
// split — the failure mode that produced the job timeouts this table exists
// to prevent. Scaled from the sibling's measured 87 s by analyze-run count.
// Replace with a real figure after the first green Windows matrix run.
'test/unit/incremental-index-extension-dml-gate.test.ts': 180,
// Measured on Windows in run 34014266125 (#3190, 2026-09-06). These DB
// suites landed together on shard 3: the old 180s estimate and missing
// entries made a ~27-minute recorded load look like an ~12-minute shard.
// Upstream speedups may reduce these figures; retaining conservative weights
// keeps the expensive suites distributed without changing the watchdog.
'test/unit/incremental-index-extension-dml-gate.test.ts': 414,
'test/integration/skills-e2e.test.ts': 444,
'test/integration/fts-extension-e2e.test.ts': 146,
'test/integration/analyze-wal-checkpoint-failure.test.ts': 86,
'test/integration/cli-limit-e2e.test.ts': 75,
'test/unit/hooks.test.ts': 26,
'test/integration/analyze-heap-oom-e2e.test.ts': 23,

29
gitnexus/test/unit/cross-platform-shard.test.ts
View file

@ -29,6 +29,33 @@ const allShards = (files: readonly string[], total: number): readonly (readonly
Array.from({ length: total }, (_unused, i) => shardFiles(files, i + 1, total));
describe('cross-platform shard partition', () => {
it('does not recreate the overloaded Windows shard from the #3190 CI run', () => {
// Run 34014266125: these serialized DB suites were missing or undercharged
// in the scheduling table. Keep the observed profile independent of the
// table so deleting a weight cannot make this regression pass again.
const observed: Readonly<Record<string, number>> = {
'test/integration/skills-e2e.test.ts': 444,
'test/unit/incremental-index-extension-dml-gate.test.ts': 414,
'test/integration/fts-extension-e2e.test.ts': 146,
'test/integration/analyze-wal-checkpoint-failure.test.ts': 86,
};
const floor = weightOf('test/unmeasured.test.ts');
const loads = allShards(ALL_CROSS_PLATFORM, SHARD_TOTAL).map((files) =>
files.reduce((sum, file) => sum + Math.max(weightOf(file), (observed[file] ?? 0) + floor), 0),
);
// This is a deterministic replay of recorded weights, not a timing test.
// The runner's existing watchdog remains 20 minutes.
expect(Math.max(...loads)).toBeLessThan(20 * 60);
const shards = allShards(ALL_CROSS_PLATFORM, SHARD_TOTAL);
const heavyweightLocations = [
'test/integration/cli-e2e.test.ts',
'test/integration/skills-e2e.test.ts',
'test/unit/incremental-index-extension-dml-gate.test.ts',
].map((file) => shards.findIndex((files) => files.includes(file)));
expect(heavyweightLocations).not.toContain(-1);
expect(new Set(heavyweightLocations).size).toBe(SHARD_TOTAL);
});
it('covers every file exactly once, with no overlap between shards', () => {
const shards = allShards(ALL_CROSS_PLATFORM, SHARD_TOTAL);
const seen = shards.flatMap((s) => [...s]);
@ -48,7 +75,7 @@ describe('cross-platform shard partition', () => {
expect(Math.max(...weights)).toBeLessThanOrEqual(ideal * 1.34);
});
it('never puts the two heaviest suites on the same shard', () => {
it('keeps the CLI and worker-pool suites on different shards', () => {
// The exact shape of the outage: cli-e2e and worker-pool are 621 s and
// 222 s, so together they are most of a shard's budget before anything else
// is scheduled.