Borehole · Project · free tier
databendlabs/databend
Free reading: not for reliance
c471f50 · 34,902 commits · 7,204 files
Read 2026-11-03

databendlabs/databend: technical due diligence, read from its repository

What Borehole read in databendlabs/databend's code and history at c471f50: its licence, security advisories, releases, review practice, tests, dependencies and operations. Every finding cites the file and line it rests on. The full report also covers the team and asks the questions the code cannot answer.

DimensionRating
Architecture & Codebase — including AI provenancestrong
Engineering Process & SDLC Maturityadequate
Product & Engineering Maturitystrong
Security & Reliability Postureadequate
Observability & Operationsadequate
Fit with the Acquirernot assessable

What the repository says

  • Releases are cut from another branch than the one surveyed
  • LICENSE: Relicensed: Apache-2.0 → AGPL-3.0 (2020-10-14) → Apache-2.0 (2021-02-15) → Elastic-2.0 (2023-04-28) → Apache-2.0 (2026-10-15)
  • src/meta/ee/Cargo.toml: 41 files are named as enterprise-edition code
  • AGENTS.md:32: states a policy on AI-generated contributions

Architecture & Codebase — including AI provenance

strong
01.6

src/meta/schema-api-test-suite/src/schema_api_test_suite.rs is 10,512 lines, 89× this codebase's median file

adequate

The median source file here is 117 lines. 99 files exceed 1,170 lines, the largest being 10,512. Size is not a defect on its own — some problems genuinely live in one place — but these are where merge conflicts, review fatigue and single-owner knowledge concentrate, and they are the first thing that slows an inheriting engineer down.

Evidence
src/meta/schema-api-test-suite/src/schema_api_test_suite.rs:1 — 10,512 lines (89× median)
src/query/ast/src/parser/statement.rs:1 — 7,185 lines (61× median)
src/query/config/src/config.rs:1 — 4,308 lines (36× median)
src/query/service/src/interpreters/access/privilege_access.rs:1 — 3,705 lines (31× median)
src/query/functions/src/scalars/variant.rs:1 — 3,626 lines (30× median)

Engineering Process & SDLC Maturity

adequate
02.24

8% of recent commits change more than 2,000 lines

adequate

205 of 2615 commits in the last 24 months changed over 2,000 lines of code, not counting reformatting, renames, vendored or generated code. A change that size is rarely reviewed line by line. Ask how they were reviewed, and what made them that large.

Evidence
205 commits over 2,000 lines

Product & Engineering Maturity

strong
04.62

Releases are cut from another branch than the one surveyed

adequate

The newest pre-release, v1.2.911-rc6, is not on the branch this survey read; the newest release on it is v1.2.881. The shipped product may differ from what this report describes. Survey the release branch too.

Evidence
tag v1.2.911-rc6

Security & Reliability Posture

adequate
05.6

6 Dockerfiles run the application as root

adequate

No USER instruction switches away from root, so the process runs with full privileges inside the container. It is a one-line fix and an ordinary expectation, which is why its absence is worth asking about: it usually says more about whether anyone has reviewed the deployment than about this container specifically.

Evidence
.oss-scanner/Dockerfile — runs as root
docker/Dockerfile — runs as root
docker/build-tool/debian/Dockerfile — runs as root
docker/build-tool/dev/Dockerfile — runs as root
05.27

The build runs third-party code it does not pin

adequate

11 third-party CI actions are referenced by a tag or branch rather than a commit, so they run whatever that tag points to on the day; 2 steps pipe a download straight into a shell. Tags have been moved to malicious commits in real supply-chain attacks, and a piped script is whatever the server returns that minute. Pinning to a commit is the fix, and it is small.

Evidence
.github/workflows/build_tool.yml:33 — action pinned by tag
.github/workflows/cloud.yml:155 — action pinned by tag
scripts/ci/ci-setup-chaos-meta.sh:53 — download piped into a shell
05.26

52 locked dependencies have published vulnerabilities, 13 high or critical

adequate

52 of 1,434 locked dependencies match 101 published advisories in OSV, 13 of them rated high or critical. A lockfile says which version is installed, not whether the flawed code is reachable from this product, so each needs a look. Most are fixed by an upgrade, and the advisory names the version. None of the high or critical ones is in what the product ships: they are optional extras of a package, installed only on request. This lockfile pins development tools alongside what ships and does not say which is which, so some of these may never reach a user.

Evidence
Cargo.lock — wasmtime 27.0.0: GHSA-897f-g5gm-fmxm (critical), fixed in 36.0.19
Cargo.lock — zip 4.6.1: GHSA-m665-v7pm-7grj (critical), fixed in 9.0.5
src/bendpy/uv.lock — pyarrow 21.0.0: GHSA-rgxp-2hwp-jwgg (high), fixed in 23.0.1; an optional extra
05.60

Relicensed: Apache-2.0 → AGPL-3.0 (2020-10-14) → Apache-2.0 (2021-02-15) → Elastic-2.0 (2023-04-28) → Apache-2.0 (2026-10-15)

adequate

The root licence changed 4 time(s). Code taken under earlier terms stays under them; check which terms any fork or customer relied on.

Evidence
LICENSE — AGPL-3.0 from 2020-10-14
LICENSE — Apache-2.0 from 2021-02-15
LICENSE — Elastic-2.0 from 2023-04-28
LICENSE — Apache-2.0 from 2026-10-15
05.61

41 files are named as enterprise-edition code

adequate

Files named `.ee.` are, by the common open-core convention, the paid edition's code, usually under a licence that is not open source even where the rest is. What the buyer may do with each part differs.

Evidence
src/meta/ee/Cargo.toml — enterprise-edition file
src/meta/ee/src/lib.rs — enterprise-edition file
src/query/ee/Cargo.toml — enterprise-edition file

Observability & Operations

adequate
06.8

One request cannot be followed through the logs

adequate

Nothing in the service carries a request or correlation id, and there is no tracing. When a customer reports a failure, the log lines that belong to their request cannot be picked out from everyone else's.

Evidence
no request id, correlation id or tracing found
06.9

No error reporter is wired in

adequate

Nothing in the service initialises an error reporter (Sentry, Rollbar, Bugsnag, Honeybadger or an APM agent). An exception in production reaches a person only if someone is reading the logs, which usually means a customer reports it first.

Evidence
no error reporter initialised
06.35

Has a reporting host, telemetry.databend.com; its default could not be read

adequate

src/common/building/src/lib.rs names telemetry.databend.com as a place to report to, and no setting read here shows whether reporting is on as shipped. For a buyer running it inside their own product, ask whether copies report by default, what is sent, how often, and whether the buyer's customers were told.

Evidence
src/common/building/src/lib.rs:130 — names telemetry.databend.com

Fit with the Acquirer

not assessable

Not rated: structurally not assessable from a repository.

Read more, or read your own

The full report on databendlabs/databend: every dimension, the questions for the room, and the audit log.

Survey a repository: free on public repositories. Private code is read on your own machine by a collector whose source you can read first.