fix(generations): a sealed segment may only declare the generations it holds
Diagnosis of the "packed history is damaged" narration that fires on every
run of the affected stores. It is a WRITER defect, and the reader's refusal
was the symptom rather than the cause.
A sealed segment declares one contiguous range [firstGeneration,
lastGeneration], and every reader treats that range as containment:
coveringSegment is an interval test, hasGeneration returns true for anything
inside it, and open() seeds committedRanges from it.
repackHistory handed fold() a SPARSE batch. Three filters punch holes in its
candidate list mid-run — a generation absent from committedRanges never
appears, one still in the pending buffer is skipped, one whose tx.json will
not read is skipped — and fold() then computed the range from the first and
last survivor, claiming every generation in between. The next open merged
that mis-declared range back into committedRanges, re-admitting the hole as
committed history, so the following auto-compaction pass asked the packed
tier for a frame that was never written and failed. Re-merged at every open,
which is why it repeated on every run.
Confirmed against a forensic fixture: generation directories 1..2503 present
except exactly one, 1416; and its fact-log segment already showed the tell —
seg-...1410.bfl declaring 1410..1940 (531 generations) while recording 530
facts.
Three changes:
- repackHistory folds each contiguous RUN as its own segment
(`contiguousRuns`), so ranges describe exactly what the segments contain.
- fold() REFUSES a non-contiguous batch, naming the gap and its width. The
density law is now mechanical, so no future caller can reintroduce it. A
refusal loses nothing: the generations stay live and readable.
- Stores already carrying the damage heal instead of wedging. A segment
whose declared span exceeds its frame count is SPARSE; `actualRanges()`
reads the real generation list from its sidecar so open() never re-admits
the holes, and readFrame reports such a hole as unpacked with a narration
naming the segment, rather than throwing. A DENSE segment missing a frame
is still loud damage — that one means the manifest and sidecar disagree.
Pins: nine unit cases (refusal and its message, honest ranges for separately
folded runs, a reconstructed pre-fix sparse segment serving its real frames
while reporting holes as unpacked, holes excluded from actualRanges, and the
dense-segment damage path still throwing) plus an end-to-end case that
deletes a generation directory and drives the real sequence — ordinary
close()-time repacking folds over the hole, then reopen and compact must both
complete. Verified red without the fix: the segment declared an
11-generation span while holding 10 frames.
(cherry picked from commit 9a888c37e9)
This commit is contained in:
parent
c99308710a
commit
a963a744cc
4 changed files with 389 additions and 13 deletions
|
|
@ -147,6 +147,60 @@ export class GenerationSegmentStore {
|
|||
return this.coveringSegment(gen) !== null
|
||||
}
|
||||
|
||||
/**
|
||||
* @description True when `meta` declares more generations than it holds
|
||||
* frames — a segment sealed by a writer that folded across a hole. The
|
||||
* manifest records `frames` at fold time, so this is an O(1) comparison
|
||||
* against the declared span and needs no I/O.
|
||||
*/
|
||||
private isSparse(meta: SegmentMeta): boolean {
|
||||
return meta.lastGeneration - meta.firstGeneration + 1 !== meta.frames
|
||||
}
|
||||
|
||||
/**
|
||||
* @description The generations this tier ACTUALLY holds, as coalesced
|
||||
* ascending intervals — not what the segments declare.
|
||||
*
|
||||
* Dense segments (every one a current writer produces) contribute their
|
||||
* declared range with no I/O. A SPARSE segment — one sealed before the
|
||||
* density law was enforced, whose declared range spans generations it has
|
||||
* no frame for — has its real generation list read from its sidecar and
|
||||
* contributed instead, with the discrepancy narrated once.
|
||||
*
|
||||
* This is what keeps a store that already carries the damage from wedging.
|
||||
* `open()` seeds `committedRanges` from these intervals, so a hole is never
|
||||
* re-admitted as a committed generation, and the auto-compaction pass that
|
||||
* used to fail on every run with "packed history is damaged" simply never
|
||||
* asks for the missing frame.
|
||||
*
|
||||
* @returns Ascending, non-overlapping `[first, last]` intervals.
|
||||
*/
|
||||
async actualRanges(): Promise<Array<[number, number]>> {
|
||||
const out: Array<[number, number]> = []
|
||||
for (const meta of this.manifest.segments) {
|
||||
if (!this.isSparse(meta)) {
|
||||
out.push([meta.firstGeneration, meta.lastGeneration])
|
||||
continue
|
||||
}
|
||||
const missing = meta.lastGeneration - meta.firstGeneration + 1 - meta.frames
|
||||
prodLog.warn(
|
||||
`[GenerationSegments] sealed segment ${meta.file} declares generations ` +
|
||||
`${meta.firstGeneration}..${meta.lastGeneration} but holds only ${meta.frames} ` +
|
||||
`frame(s) — ${missing} generation(s) in that span were never folded into it. ` +
|
||||
`Serving the frames it actually holds; the declared span is not treated as ` +
|
||||
`committed history. (Written by a pre-density-law writer that folded across a ` +
|
||||
`gap; the segment itself is intact and no record is lost.)`
|
||||
)
|
||||
const idx = await this.sidecarFor(meta)
|
||||
for (const [gen] of idx.generations) {
|
||||
const last = out[out.length - 1]
|
||||
if (last !== undefined && gen === last[1] + 1) last[1] = gen
|
||||
else out.push([gen, gen])
|
||||
}
|
||||
}
|
||||
return out
|
||||
}
|
||||
|
||||
/**
|
||||
* Fold consecutive generations into ONE new sealed segment + sidecar and
|
||||
* append it to the manifest atomically. Caller guarantees: `gens` is
|
||||
|
|
@ -164,6 +218,38 @@ export class GenerationSegmentStore {
|
|||
throw new Error('[GenerationSegments] fold() input must be strictly ascending')
|
||||
}
|
||||
}
|
||||
// THE DENSITY LAW, MADE MECHANICAL.
|
||||
//
|
||||
// A sealed segment declares a CONTIGUOUS range [firstGeneration,
|
||||
// lastGeneration] and every reader treats that range as containment:
|
||||
// `coveringSegment` is an interval test, `hasGeneration` returns true for
|
||||
// anything inside it, and `open()` seeds committedRanges from it. So a
|
||||
// segment folded from a SPARSE input silently claims generations it does
|
||||
// not hold, and the first read of one of those holes throws
|
||||
// "inside sealed segment ... but has no frame — packed history is damaged".
|
||||
//
|
||||
// That is exactly how the damage was produced. `repackHistory` skipped
|
||||
// generations mid-batch — ones absent from committedRanges, ones still in
|
||||
// the pending buffer, ones whose tx.json would not read — and handed the
|
||||
// survivors here, where the range was computed from the first and last of
|
||||
// them. Worse, the mis-declared range was then merged back into
|
||||
// committedRanges at the next open, which is what turned a quiet hole into
|
||||
// a repeating auto-compaction failure on every subsequent run.
|
||||
//
|
||||
// Callers now split at discontinuities; this refusal is what keeps any
|
||||
// future caller from reintroducing the class. A refusal here loses
|
||||
// nothing — the generations stay in the live tier, readable, and the next
|
||||
// pass folds them correctly.
|
||||
for (let i = 1; i < gens.length; i++) {
|
||||
if (gens[i].generation !== gens[i - 1].generation + 1) {
|
||||
throw new Error(
|
||||
`[GenerationSegments] fold() input is not contiguous: ${gens[i - 1].generation} → ` +
|
||||
`${gens[i].generation} skips ${gens[i].generation - gens[i - 1].generation - 1} ` +
|
||||
`generation(s). A sealed segment declares a dense range, so folding a sparse ` +
|
||||
`batch would claim generations it does not hold. Split the batch at the gap.`
|
||||
)
|
||||
}
|
||||
}
|
||||
const last = this.manifest.segments[this.manifest.segments.length - 1]
|
||||
if (last && gens[0].generation <= last.lastGeneration) {
|
||||
throw new Error(
|
||||
|
|
@ -364,12 +450,37 @@ export class GenerationSegmentStore {
|
|||
return this.decodeFrame(payload)
|
||||
}
|
||||
}
|
||||
// In the covering range but not present: the packed tier is dense by
|
||||
// construction (fold packs every generation it is handed, including
|
||||
// record-less ones) — absence inside a sealed range is damage.
|
||||
// Inside the covering range but with no frame. Two very different causes,
|
||||
// and conflating them is what made this class wedge every maintenance pass
|
||||
// on the affected stores.
|
||||
//
|
||||
// (1) A SPARSE SEGMENT — the manifest's own `frames` count is smaller than
|
||||
// the span it declares. That segment was sealed by a writer that
|
||||
// folded across a hole (the class this file's density law now bars).
|
||||
// The segment is INTACT and nothing is lost; it simply never held this
|
||||
// generation. Answering "not packed" is the honest answer, and it lets
|
||||
// the caller's two-tier read decide what a genuinely absent generation
|
||||
// means, instead of every compaction pass dying on a repeating throw.
|
||||
// `actualRanges()` keeps such holes out of committedRanges at open, so
|
||||
// in a healed store nobody asks this question in the first place.
|
||||
//
|
||||
// (2) A DENSE SEGMENT missing a frame it says it has — the manifest and
|
||||
// the sidecar disagree about a segment that claims to be complete.
|
||||
// That IS damage, and it stays loud.
|
||||
if (this.isSparse(meta)) {
|
||||
prodLog.warn(
|
||||
`[GenerationSegments] generation ${gen} falls inside sealed segment ${meta.file}'s ` +
|
||||
`declared range ${meta.firstGeneration}..${meta.lastGeneration}, but that segment ` +
|
||||
`holds ${meta.frames} frame(s) for a ${meta.lastGeneration - meta.firstGeneration + 1}` +
|
||||
`-generation span — it was sealed across a gap and never held this generation. ` +
|
||||
`Reporting it as unpacked rather than as damage; no record is lost.`
|
||||
)
|
||||
return null
|
||||
}
|
||||
throw new Error(
|
||||
`[GenerationSegments] generation ${gen} is inside sealed segment ${meta.file}'s declared ` +
|
||||
`range but has no frame — packed history is damaged`
|
||||
`range but has no frame, and that segment declares a complete ${meta.frames}-frame ` +
|
||||
`span — the manifest and the sidecar disagree; packed history is damaged`
|
||||
)
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -96,6 +96,35 @@ export const FOLD_CHECKPOINT_PATH = '_system/fold-checkpoint.json'
|
|||
/** Storage-root-relative prefix of the per-generation record directories. */
|
||||
export const GENERATIONS_PREFIX = '_generations'
|
||||
|
||||
/**
|
||||
* @description Split an ascending list of fold candidates into maximal
|
||||
* CONTIGUOUS runs — `[7,8,9,12,13]` becomes `[[7,8,9],[12,13]]`.
|
||||
*
|
||||
* A sealed segment declares one dense range `[firstGeneration,
|
||||
* lastGeneration]`, and every reader treats that range as containment. So a
|
||||
* batch with a hole in it must never become one segment: it would claim a
|
||||
* generation it does not hold, and the first read of that hole reports the
|
||||
* packed history as damaged. One run, one segment — the ranges then describe
|
||||
* exactly what the segments contain.
|
||||
*
|
||||
* @param gens - Fold candidates, strictly ascending by generation.
|
||||
* @returns One array per contiguous run, in ascending order. Empty in, empty out.
|
||||
*/
|
||||
export function contiguousRuns(gens: FoldGeneration[]): FoldGeneration[][] {
|
||||
const runs: FoldGeneration[][] = []
|
||||
let run: FoldGeneration[] = []
|
||||
for (const g of gens) {
|
||||
const prev = run[run.length - 1]
|
||||
if (prev !== undefined && g.generation !== prev.generation + 1) {
|
||||
runs.push(run)
|
||||
run = []
|
||||
}
|
||||
run.push(g)
|
||||
}
|
||||
if (run.length > 0) runs.push(run)
|
||||
return runs
|
||||
}
|
||||
|
||||
/**
|
||||
* @description Phases of the {@link GenerationStore.commitTransaction} commit
|
||||
* protocol at which a test-only fault injector can simulate a process crash.
|
||||
|
|
@ -784,9 +813,15 @@ export class GenerationStore {
|
|||
if (storageSupportsFactLog(this.storage)) {
|
||||
this.segments = new GenerationSegmentStore(this.storage)
|
||||
await this.segments.open()
|
||||
const packedRanges = this.segments
|
||||
.segments()
|
||||
.map((s): [number, number] => [s.firstGeneration, Math.min(s.lastGeneration, this.committed)])
|
||||
// ACTUAL ranges, not declared ones. A segment sealed by a pre-density-law
|
||||
// writer can declare a span wider than the frames it holds; seeding
|
||||
// committedRanges from the declared span re-admits those holes as
|
||||
// committed generations, and every later maintenance pass then asks for a
|
||||
// frame that was never written. `actualRanges()` reads the real
|
||||
// generation list from the sidecar for exactly those segments (and does
|
||||
// no I/O for the dense ones, which is all of them on a healthy store).
|
||||
const packedRanges = (await this.segments.actualRanges())
|
||||
.map((r): [number, number] => [r[0], Math.min(r[1], this.committed)])
|
||||
.filter(([lo, hi]) => lo <= hi)
|
||||
if (packedRanges.length > 0) {
|
||||
// Merge packed (older) + live (newer) interval sets — both ascending;
|
||||
|
|
@ -3121,13 +3156,26 @@ export class GenerationStore {
|
|||
foldInput.push({ generation: gen, timestamp: delta.timestamp, delta, records })
|
||||
}
|
||||
if (foldInput.length === 0) continue
|
||||
await segments.fold(foldInput)
|
||||
segmentsCreated++
|
||||
// Segment + manifest durable → the live copies retire.
|
||||
for (const g of foldInput) {
|
||||
await this.storage.removeRawPrefix(`${GENERATIONS_PREFIX}/${g.generation}`)
|
||||
// SPLIT AT DISCONTINUITIES. `eligible` is NOT contiguous — three
|
||||
// filters above punch holes in it: a generation missing from
|
||||
// committedRanges never appears, one still in the pending buffer is
|
||||
// skipped, and one whose tx.json will not read is skipped. A sealed
|
||||
// segment declares a DENSE range, so folding across such a hole makes
|
||||
// the segment claim a generation it does not hold; the next open
|
||||
// merges that mis-declared range into committedRanges, and every
|
||||
// subsequent auto-compaction pass then asks for the missing frame and
|
||||
// fails with "packed history is damaged". Fold each contiguous RUN as
|
||||
// its own segment instead — same bytes, honest ranges.
|
||||
for (const run of contiguousRuns(foldInput)) {
|
||||
if (deadline !== undefined && Date.now() >= deadline) break
|
||||
await segments.fold(run)
|
||||
segmentsCreated++
|
||||
// Segment + manifest durable → the live copies retire.
|
||||
for (const g of run) {
|
||||
await this.storage.removeRawPrefix(`${GENERATIONS_PREFIX}/${g.generation}`)
|
||||
}
|
||||
folded += run.length
|
||||
}
|
||||
folded += foldInput.length
|
||||
}
|
||||
if (folded > 0) {
|
||||
prodLog.info(
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue