Skip to content

id(sync): bound data-phase batches by bytes as well as count (#180) - #2215

Merged
dezren39 merged 4 commits into
mainfrom
queue/2971-180-sync-batch-bytes
Oct 30, 2026
Merged

dezren39 merged 4 commits into
mainfrom
queue/2971-180-sync-batch-bytes

Conversation

@dezren39

@dezren39 dezren39 commented Oct 29, 2026 •

Copy link
Copy Markdown
Member

Fixes #180 (Phase 3: sync data phase batches bounded by count, not bytes), per the design pinned by triage task 2023. Queue task 2971.

Change (pkgs/id/src/event_store/sync.rs only)

  • New BATCH_BYTES = range::DEFAULT_FRAME_BUDGET (4 MiB) alongside BATCH_EVENTS (256).
  • receive_data tracks the pending batch's encoded byte total (record.len(), captured before decode) and flushes the batch before an event that would push it past BATCH_BYTES. Whichever bound fills first wins.
  • An empty batch always takes the event, so a single event larger than BATCH_BYTES is decided alone — never split, never an error on size alone (still bounded by FRAME_CAP, 16 MiB).
  • No wire, sender, BATCH_EVENTS, FRAME_CAP, or ReadBudget changes. Small-event behaviour is unchanged.

Tests (CI)

  • test_data_phase_batches_are_bounded_by_bytes_as_well_as_count: ~1.5 MiB events (byte bound binds at 2 per batch, count bound nowhere near); a committed-cut after 4 records leaves exactly 4 events in the receiver — two whole byte-bound batches, where a count-only receiver commits none.
  • test_an_event_larger_than_the_byte_bound_is_decided_alone: a 5 MiB event plus small events sync to completion; all arrive and are accepted.
  • Existing sync tests cover preserved batch semantics (whole-batch commit reporting, fork-in-batch isolation, interrupted-session behaviour).

Verification is CI-first per fleet policy: no builds were run on the sandbox or laptop.


View with [code]smith Autofix with [code]smith
Need help on this PR? Tag @codesmith-bot with what you need.

receive_data decided batches bounded only by BATCH_EVENTS (256), so one
write transaction could buffer up to 256 maximum-size frames. Add
BATCH_BYTES = range::DEFAULT_FRAME_BUDGET (4 MiB): track the pending
batch's encoded byte total (record length, captured before decode) and
flush before an event that would push the batch past the bound. An
empty batch always takes the event, so a single event larger than the
bound is decided alone — never split, never an error on size alone
(it remains bounded by FRAME_CAP). Count-only behaviour for small
events is unchanged.

Tests: a loopback cut test with ~1.5 MiB events proves two byte-bound
batches of two commit where a count-only receiver would commit none;
a second test syncs a single event larger than BATCH_BYTES plus small
events to completion.

Queue task 2971; design pinned by triage task 2023 on #180.
@dezren39
dezren39 requested a review from a team as a code owner October 29, 2026 09:37
@coderabbitai

coderabbitai Bot commented Oct 29, 2026 •

Copy link
Copy Markdown

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 24 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available.

Learn how review limits work.

Review configuration:

Run configuration
  • Source mode: Auto
  • Repository: Unavailable
  • Branch: Unavailable
  • File: No configuration file from this repository applied
  • Commit: Unavailable
  • Layers: defaults
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 1f482d0a-8202-4829-8368-12aa97547972
Commits

Reviewing files that changed from the base of the PR and between 01cad93 and 26fa8fa.

Files selected for processing: 2
  • pkgs/id/src/event_store/sync.rs
  • pkgs/id/src/world/actor.rs
  • Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 29, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-30T05:18:54.226766Z 26fa8fa New commits
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a10c24f6d3

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread pkgs/id/src/event_store/sync.rs Resolved
Codex P1 on PR 2215 was right: chain::sign commits the body by hash, so the old fixtures encoded to ~186-byte records and the byte bound never fired (CI test-unit: left 0, right 4). The largest valid record a fixture can build is ~10 KiB (tag.clear with 256 claims, measured), so the production 4 MiB bound is unreachable by signed fixtures. Thread the bound through run()/receive_data as test plumbing (same pattern as Cut/CommitSignal; production passes BATCH_BYTES unchanged) and rewrite both tests at fixture scale: batches of two records under a two-record bound, and solo batches under a sub-record bound, accounting for a batch flushing only when the next overflowing record arrives.
@dezren39
dezren39 added this pull request to the merge queue Oct 30, 2026
@github-merge-queue
github-merge-queue Bot removed this pull request from the merge queue due to a conflict with the base branch Oct 30, 2026
@dezren39
dezren39 merged commit 6add000 into main Oct 30, 2026
@dezren39
dezren39 deleted the queue/2971-180-sync-batch-bytes branch October 30, 2026 05:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Phase 3: sync data phase batches are bounded by count, not bytes

1 participant