add time budget for integrity checks#20714
Merged
sudeepdino008 merged 15 commits intomainfrom Apr 23, 2026
Merged
Conversation
…; log exhaustion as info
…avy runs honor budget
sudeepdino008
requested review from
AskAlexSharov and
Giulio2002
as code owners
sudeepdino008
changed the title
AskAlexSharov
approved these changes
Apr 22, 2026
Contributor
There was a problem hiding this comment.
Pull request overview
Adds an optional wall-clock time budget to the erigon seg integrity command so integrity runs can be capped and (when budgeted) checks are executed in a cheap→heavy order with per-check time slices.
Changes:
- Add
--integrity.budget <duration>and run integrity checks sequentially with per-check timeouts derived from remaining budget. - Make long-running integrity routines more responsive to cancellation/deadlines (
BlockReader.IntegrityTxnIDandCheckKvi). - Refactor
doPublishableto acceptdatadir.Dirsinstead of*cli.Context.
Reviewed changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| db/snapshotsync/freezeblocks/block_reader.go | Add context.Context to IntegrityTxnID and poll for cancellation periodically. |
| db/integrity/integrity_kvi.go | Add periodic ctx polling in the hot skip path and propagate ctx.Err() on cancellation. |
| db/integrity/integrity_action_type.go | Reorder FastChecks and add SortChecksByCost helper for budgeted runs. |
| cmd/utils/app/snapshots_cmd.go | Add --integrity.budget, implement per-check budgeting/timeout behavior, and adjust doPublishable signature/call sites. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
… on perCheck log)
sudeepdino008
added a commit
that referenced
this pull request
Apr 27, 2026
Add `--integrity.budget <duration>` to `erigon seg integrity` that caps total wall-clock time for an integrity run. - When set, checks are sorted cheap → heavy (based on `integrity.FastChecks` order) and each receives a per-check deadline of `remaining / remaining_checks`, so finishing early on a cheap check rolls budget forward to the heavy tail. - On per-check timeout: log `[integrity] budget exhausted, moving on` at INFO and continue to the next check. Not treated as an integrity failure — `--failFast` only reacts to actual integrity errors. - When not set (default): no change in behavior — user-supplied order preserved, no deadline applied. - `BlockReader.IntegrityTxnID` now takes `ctx` and polls every 1000 segments so `BlocksTxnID` honors the budget. - `CheckKvi` producer now checks `ctx` every 100k keys, so skip-heavy runs (e.g. `CommitmentKvi` with `SampleRatio=0`) don't overshoot their slice by minutes. - `doPublishable` takes `datadir.Dirs` directly instead of `*cli.Context`. follow up: - pick to release/3.4 - change automation script to allot 30min budget --------- Co-authored-by: Alex Sharov <AskAlexSharov@gmail.com>
wmitsuda
added a commit
that referenced
this pull request
Apr 27, 2026
… (#20838) Cherry-picks the integrity-area chain from `main` onto `performance`, in order: 1. #20435 — `integrity: restore StateProgress check` 2. #20680 — `integrity, stagedsync: cap state collation at block snapshots progress` 3. #20714 — `add time budget for integrity checks` 4. #20835 — `cmd/utils/app: skip commitment history integrity checks when disabled` #20714 is the structural refactor of `doIntegrity` (`errgroup.Go(switch{...})` → `runCheck := func{}` + budget loop). Conflicted with performance-only Bloatnet customizations on the same function (#20806, #20810, #20813); resolved by adopting #20714's shape and re-applying the Bloatnet handling for `CommitmentKvDeref`, `CommitmentHistVal`, and `StateRootVerifyByHistory`. `failFast` arg on `CheckCommitmentHistAtBlkRange` preserved (matches performance's signature). Submodule pointer for `execution/tests/execution-spec-tests` left unchanged. `make erigon` + `make lint` clean.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add
--integrity.budget <duration>toerigon seg integritythat caps total wall-clock time for an integrity run.integrity.FastChecksorder) and each receives a per-check deadline ofremaining / remaining_checks, so finishing early on a cheap check rolls budget forward to the heavy tail.[integrity] budget exhausted, moving onat INFO and continue to the next check. Not treated as an integrity failure —--failFastonly reacts to actual integrity errors.BlockReader.IntegrityTxnIDnow takesctxand polls every 1000 segments soBlocksTxnIDhonors the budget.CheckKviproducer now checksctxevery 100k keys, so skip-heavy runs (e.g.CommitmentKviwithSampleRatio=0) don't overshoot their slice by minutes.doPublishabletakesdatadir.Dirsdirectly instead of*cli.Context.follow up: