Repository navigation
Release CLI 3.4.3 and Docker v0.0.10 - #1172
Merged
Merged
Conversation
Bump helix-cli to 3.4.3 and its default local image to v0.0.10, move current-version references in CLI tests, READMEs, docs and the published-image parity example, and add the v0.0.10 / 3.4.3 release notes. The local server guide now warns that v0.0.10 upgrades a persisted database's index storage version, which v0.0.9 and earlier then refuse to open.
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Automations to automatically generate PRs for you. |
counts_over_range_intersections_apply_every_filter (#1169) seeds a 600-entry write batch on the default 2 MiB test thread, which overflows its stack in Linux debug builds and fails the workspace test jobs on main. Run it through run_high_stack_contract like the file's other heavy contracts.
matthewsanetra
added a commit
that referenced
this pull request
Oct 5, 2026
The v0.0.10 image release ([run 37354464092](https://github.com/HelixDB/helix-db/actions/runs/37354464092)) did not publish. Its `Workspace and tooling quality` job hit `timeout-minutes: 30` partway through `Server coverage`, so `Publish tested image` was skipped. The main push run for #1172 timed out the same way. | Step | Duration | | --- | --- | | Workspace tests | 26m31s (passed) | | Workspace doctests | 51s (passed) | | Server coverage | cancelled at the 30-minute limit | | Clippy, Formatting, Shell syntax | skipped | Before #1172's stack fix, the job failed early in `production_contracts`, so it never reached the limit. This raises only the quality job's timeout to 60 minutes. The image, benchmark and publish timeouts are unchanged. CLI 3.4.3 is already released and defaults to v0.0.10, so this blocks fresh `helix start` until the image publishes. After merging, re-dispatch `docker-image.yml` from main with `release_version=v0.0.10`. <!-- greptile_comment --> <!-- greptile_summary --> <p><a href="https://app.greptile.com/api/retrigger?id=75063203"><picture><source media="(prefers-color-scheme: dark)" srcset="https://greptile-static-assets.s3.amazonaws.com/badges/RetriggerDark.svg?v=2"><source media="(prefers-color-scheme: light)" srcset="https://greptile-static-assets.s3.amazonaws.com/badges/Retrigger.svg?v=2"><img alt="Retrigger" src="https://greptile-static-assets.s3.amazonaws.com/badges/Retrigger.svg?v=2" align="right"></picture></a><br clear="all"></p> <!-- greptile-risk --> The PR appears safe to merge. <h3>Summary</h3> Raises the Docker image workflow’s quality-job timeout from 30 to 60 minutes so the release gate has more time to finish. Image, benchmark, and publish timeouts are unchanged. <sub>Reviews (1) · Last reviewed commit: ["ci(docker-image): raise the quality job ..."](https://github.com/helixdb/helix-db/commit/eab93854b1f9efba10272b53f68475151a252548)</sub> <!-- /greptile_comment -->
nepalisagun
pushed a commit
to nepalisagun/helix-db
that referenced
this pull request
Oct 6, 2026
Workspace tests now take about 26.5 minutes and doctests about one more, so the 30-minute limit cancels the job during server coverage. It cancelled the main push run for HelixDB#1172 and the v0.0.10 release dispatch, which skipped publication. Clippy, formatting and shell syntax still follow coverage.
xav-db
added a commit
that referenced
this pull request
Oct 6, 2026
![https://greptile-static-assets.s3.amazonaws.com/badges/RetriggerDark.svg?v=2"><source](https://greptile-static-assets.s3.amazonaws.com/badges/RetriggerDark.svg?v=2"><source){kind=link}
![https://greptile-static-assets.s3.amazonaws.com/badges/Retrigger.svg?v=2"><img](https://greptile-static-assets.s3.amazonaws.com/badges/Retrigger.svg?v=2"><img){kind=link}
Main now includes the range-intersection count contract's 16 MiB stack fix through #1172. Merge current main to resolve the duplicate change and retain the explanation that the debug future must be constructed and polled on the dedicated thread. The remaining diff is two comment lines; the fixture and every assertion match main. ```text libtest → dedicated 16 MiB thread → Tokio → unchanged contract ``` Validated the merged branch on fresh ARM64 EC2: - All 58 production contracts passed, plus the count contract with a 1 MiB outer test stack. - Workspace Clippy (`-D warnings`) and formatting passed. - Reviewed the diff and confirmed it changes only the explanatory comment. The original fix was also validated with rebuilt-image packaging, native-volume persistence, S3/Compose suites, full workspace tests and doctests before main incorporated it.
xav-db
added a commit
that referenced
this pull request
Oct 7, 2026
Release PR for **CLI 3.4.4** and **Docker image v0.0.11**, following the same shape as #1172. ```diff crates/cli/Cargo.toml version 3.4.3 → 3.4.4 (and Cargo.lock) crates/cli/src/config.rs DEFAULT_LOCAL_IMAGE_TAG v0.0.10 → v0.0.11 CLI tests, README, CLAUDE.md v0.0.10 → v0.0.11 docs default image, v0.0.11 release notes, llms-full.txt docker-image/README.md published release v0.0.10 → v0.0.11 CONTRIBUTORS.md CLI image v0.0.10 → v0.0.11 sdks/tests/parity/COVERAGE.md published-image parity example v0.0.10 → v0.0.11 ``` The local-server storage warning now reads "v0.0.10 or later", because v0.0.10 is the version that moved to index storage version 5. The v0.0.10 release notes are unchanged. ## Order 1. Merge this. The `Docker image` publish job requires `DEFAULT_LOCAL_IMAGE_TAG` on main to equal `release_version`. 2. Dispatch the `Docker image` workflow from main with `release_version=v0.0.11`. It reruns the full suite and publishes only if it passes. 3. Once v0.0.11 is published, run `cli.yml` from main to tag and publish v3.4.4. ## What's in v0.0.11 / 3.4.4 - **Strong text search keeps working during ingest** (#1176). It has its own bound, 512 MiB by default, configurable with `HELIX_STRONG_TEXT_SEARCH_MAX_ANALYSIS_BYTES`, and it reuses pending-text analyses across searches. 20K–40K-document bursts gave 0 rejections, against 106 and 234 before. - **Strong searches decode pending work once per request**, off the async workers, and a strong vector bound is added, set with `HELIX_STRONG_VECTOR_SEARCH_MAX_PENDING_BYTES` (#1174). 4 searches in one request went from 5.0 s to 2.2 s. - **Queue admission counts each operation's real memory**: encoded size + 576 B (#1178). - **Writer open streams the queue backlog**: 25 s / 4.5 GB → 5 s / 1.3 GB on a 2.26 GB backlog (#1177). - **Linear publication drains** (#1175). - No storage changes since v0.0.10. ## Validation of main (cfc5fca and 7372487) on EC2 c7i.8xlarge - **Checks:** fmt, clippy, `cargo check --all-features`, db lib and doc tests, server tests, and all release production suites (212 tests). - **Image:** `docker-image/test.sh` and every `index_queue_contracts` scenario (correctness on disk and S3, member limit, benchmark) pass. - **Upgrade from v0.0.9:** searches stay exact after the upgrade, and v0.0.9 then refuses the store with "Unsupported index storage version 5". - **Backfills** (100K docs, 100 writes/s), time until caught up, no 429s: | Index | Caught up | |---|---| | text | 215 s | | vec128 | 704 s | | vec768, disk | 1,907 s | | vec768, S3 | 1,852 s | - **Mixed workload** (100K docs, 768-d vector + text, 10 min), against v0.0.9 at the same write load: reads are unchanged (point read p99 7 ms, strong vector p99 69 vs 80 ms). Writes went from 406 / 708 ms with 16% transaction conflicts to 57 / 107 ms with none. - **This branch:** `cargo check --workspace --locked`, `cargo test -p helix-cli` (except `typescript_runtime`, which needs `npm ci` in the SDK), CLI clippy, rustfmt, and the docs checks (`check-docs`, `generate-llms(-full) --check`, `check-openapi`). <!-- greptile_comment --> <!-- greptile_summary --> <p><a href="https://app.greptile.com/api/retrigger?id=76201413"><picture><source media="(prefers-color-scheme: dark)" srcset="https://greptile-static-assets.s3.amazonaws.com/badges/RetriggerDark.svg?v=2"><source media="(prefers-color-scheme: light)" srcset="https://greptile-static-assets.s3.amazonaws.com/badges/Retrigger.svg?v=2"><img alt="Retrigger" src="https://greptile-static-assets.s3.amazonaws.com/badges/Retrigger.svg?v=2" align="right"></picture></a><br clear="all"></p> <!-- greptile-risk --> The PR appears safe to merge with the stated Docker-first publication order. <h3>Summary</h3> Releases CLI `3.4.4` with Docker image `v0.0.11` as its default. - Updates image references and matching CLI tests. - Adds release notes for search and queue improvements. - Keeps saved project tags and storage version unchanged. - `xav-db` acknowledges the merge-before-publication order: publish Docker first, then CLI. The temporary pre-publication state is intentional. <sub>Reviews (1) · Last reviewed commit: ["Release CLI 3.4.4 and Docker v0.0.11"](7e15f74) · [Reviewed by Greptile](https://www.greptile.com/?utm_source=greptile_expert&utm_medium=github&utm_campaign=code_reviews&utm_content=helixdb/helix-db)</sub> <!-- /greptile_comment -->
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
1 participant
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.
Release PR for CLI 3.4.3 and Docker image v0.0.10, following the same shape as #1161.
Version-history references stay as written:
crates/db.The TypeScript README's
package-smokecommands stay on v0.0.5, because they mirror CI's digest-pinned compatibility check intypescript-sdk.yml.The
COVERAGE.mdparity pin was bumped together with the CLI default in v0.0.5 (620fb543) and missed afterwards. v0.0.5 can't serve thesearch_consistencyparity fixtures added in #1156.Order
Docker imageworkflow's publish job requiresDEFAULT_LOCAL_IMAGE_TAGon main to equalrelease_version, so the image can only be published after this lands.Docker imageworkflow from main withrelease_version=v0.0.10. It reruns the full suite and publishes only if it passes.cli.ymlfrom main to tag and publish v3.4.3.What's in v0.0.10 / 3.4.3
unsupported_index_storage_version.tag" instructions.search_consistency: strong | eventual. Strong is the default and exact.index_backpressure(429, retryable) is returned in three cases: past 1 GB or 250K pending entities per index, when a whole-index strong search is behind more than 800 changes, and when strong text analysis would exceed one publication's budget.index_operation_batch_too_large(400) is returned when one write stages more than 8 MiB for one index.blocked_index_entity_countand retried about once a minute.missing simhashknown issue (feat(db): asynchronous vector and text index publication #1156 C1, fix(db): hold back only the index entity whose publication keeps failing #1165). Re-embeds keep their HNSW reverse-link locators. Indexes damaged by earlier versions still need a drop and recreate (HEL-952).search_consistencyover HTTP until then. chore(deps): bump http-cache-semantics from 4.2.0 to 4.3.0 in /docs in the npm_and_yarn group across 1 directory #1166 is a docs dependency bump.Testing
cargo check --workspace --lockedpasses, so the hand-editedCargo.lockis consistent.cargo test --locked -p helix-clipasses: lib 241,e2e_cli13,runtime_commands20,typescript_runtime, and the other targets. The Docker-onlye2e_runtimetests are ignored by design, because they need the unpublished v0.0.10 image.--edition 2024 --checkis clean on the edited Rust files.check-docs,generate-llms --check,generate-llms-full --checkandcheck-openapipass.helixdb/helixdb:v0.0.xreference outside release notes is v0.0.10, except the CI-pinned v0.0.5 TypeScript smoke.The PR appears safe to merge, though the published-image parity instructions should include an image pull step.
Summary
This release PR updates the CLI default to Docker v0.0.10, bumps the CLI to 3.4.3, and updates image references and release guidance. The published-image parity instructions need a pull step before their new image pin can work without a local copy.
Reviews (1) · Last reviewed commit: "chore(release): CLI 3.4.3 and Docker ima..."
Also: fix main's workspace tests
counts_over_range_intersections_apply_every_filter(#1169) overflows the default 2 MiB test-thread stack in Linux debug builds. That failsWorkspace tests,DB all targetsandWorkspace and tooling qualityon main, and the Docker release'squalityjob runs the same tests.92c15755runs it through the file's existing 16 MiBrun_high_stack_contract, like the other heavy contracts. It passes locally even withRUST_MIN_STACK=524288.