Repository navigation
Update module github.com/sigstore/rekor to v1.5.2 [SECURITY] (main) - #1875
renovate[bot] wants to merge 1 commit into
Conversation
|
|
<!-- fullsend:risk-assessment --> DetailsRe-review anchoring preserves moderate score of 2: bot-authored single-line security dependency bump in go.mod with Tier 1 and Tier 2 signals identical to prior assessment (12 commits/30d, 4 authors/90d), and no new signals provide articulable reason to deviate. Previous run<!-- fullsend:risk-assessment --> DetailsRe-review anchoring preserves moderate score of 2: bot-authored single-line security dependency bump in go.mod with Tier 1 and Tier 2 signals identical to prior assessment (12 commits/30d, 4 authors/90d), and no new signals provide articulable reason to deviate. Previous run (2)<!-- fullsend:risk-assessment --> DetailsRe-review anchoring preserves moderate score of 2: bot-authored single-line security dependency bump in go.mod with Tier 1 and Tier 2 signals identical to prior assessment (12 commits/30d, 4 authors/90d), and no new signals provide articulable reason to deviate. Previous run (3)<!-- fullsend:risk-assessment --> DetailsBot-authored single-line security dependency bump in go.mod with Tier 1 and Tier 2 signals identical to prior assessment (12 commits/30d, 4 authors/90d); re-review anchoring preserves the moderate score of 2. Previous run (4)<!-- fullsend:risk-assessment --> DetailsBot-authored single-line security dependency bump in go.mod with Tier 1 and Tier 2 signals identical to prior assessment (12 commits/30d, 4 authors/90d); re-review anchoring preserves the moderate score of 2. Previous run (5)<!-- fullsend:risk-assessment --> DetailsBot-authored single-line security dependency bump in go.mod with unchanged Tier 1 signals from prior assessment; moderate churn hotspot (12 commits/30d, 4 authors/90d) and absent test coverage sustain a moderate score consistent with the prior evaluation. Previous run (6)<!-- fullsend:risk-assessment --> DetailsBot-authored single-line security dependency bump in go.mod with unchanged Tier 1 signals from prior assessment; moderate churn (12 commits/30d) and coupling gap (go.sum and acceptance files absent) sustain a moderate score consistent with the prior evaluation. Previous run (7)<!-- fullsend:risk-assessment --> DetailsBot-authored security dependency bump touching only go.mod, with moderate recent churn (10 commits/30d, 4 distinct authors/90d) and coupling signals showing go.sum and acceptance module files are typically co-modified but absent from this PR. Previous run (8)<!-- fullsend:risk-assessment --> DetailsBot-authored single-file go.mod security bump; Tier 1 signals identical to prior assessment (1.75 sub-score) and Tier 2 signals unchanged (11 fix/security commits in 90d, 10 churn in 30d, sub-score 2.0); composite 0.62×1.75 + 0.38×2.0 = 1.85 rounds to 2, preserving the prior moderate score per re-review anchoring. Previous run (9)<!-- fullsend:risk-assessment --> DetailsXS bot-authored single-line go.mod security bump (sigstore/rekor); Tier 1 unchanged from prior assessment; Tier 2 shows elevated fix/security commit volume (11 in 90d, up from 6) and increased churn (10 in 30d, up from 5), but composite (0.62×1.25 + 0.38×2.14 = 1.59) still rounds to 2, preserving the prior moderate score per re-review anchoring. Previous run (10)<!-- fullsend:risk-assessment --> DetailsXS bot-authored single-line go.mod bump addressing a known CVE (sigstore/rekor v1.5.0→v1.5.2); Tier 1 signals unchanged from prior assessment, and Tier 2 confirms go.mod remains a moderate-churn file (5 commits/30d) with 3 recent authors and a pattern of 6 security/fix commits in 90 days, keeping the score at 2 (moderate) consistent with re-review anchoring. Previous run (11)Risk Assessment: moderate (2/5) DetailsXS bot-authored dependency bump of a single go.mod line addressing a known CVE; the change itself is minimal and low-risk, but go.mod is a high-churn file with multiple recent authors and a pattern of security patches, lifting the score to moderate. Previous run (12)Risk Assessment: moderate (2/5) DetailsExtremely small bot-authored dependency bump (1 file, 2 lines) with no protected or security-sensitive paths changed; a moderate Tier 2 score (2.0) driven by multi-author contention, coupling gaps (go.sum and acceptance modules not included in PR), and slight churn elevates the composite above low. Previous run (13)Risk Assessment: low (1/5) DetailsBot-authored (renovate[bot]) XS security bump of github.com/sigstore/rekor v1.5.0 to v1.5.2 touching only go.mod (1 file, 2 lines, blast_radius=small, 0 protected paths, 0 security-sensitive files, no CI workflow changes). Tier 2 git history shows go.mod is a well-precedented, high-churn Renovate target with many prior SECURITY bumps merged on the same pattern. Tier 3: no formal linked issue, but authorization derives from the established Renovate configuration and the CVE-2026-48702 fix rationale; rekor is an indirect dependency, further limiting runtime blast radius. Signals are unchanged from the prior assessment, so re-review anchoring preserves the score of 1. Previous run (14)Risk Assessment: low (1/5) DetailsBot-authored XS security bump of github.com/sigstore/rekor (1 file, 2 lines) following an established Renovate pattern; Tier 1 signals unchanged from prior assessment and Tier 2 confirms a well-precedented, low-churn file; re-review anchoring preserves score of 1. Previous run (15)Risk Assessment: low (1/5) DetailsBot-authored XS security bump of github.com/sigstore/rekor (1 file, 2 lines) following an established renovate pattern; Tier 1 signals unchanged from prior assessment and Tier 2 confirms a well-precedented, low-churn file; re-review anchoring preserves score of 1. Previous run (16)Risk Assessment: low (1/5) DetailsBot-authored XS dependency bump (1 file, 2 lines) from renovate for a security advisory fix, with no protected paths or CI changes touched; composite score is driven to low by the automated, well-precedented pattern of go.mod updates in this repo. Previous run (17)Risk Assessment: low (1/5) DetailsSingle go.mod line change by Renovate bot for a security update to sigstore/rekor with no protected paths, no CI workflow changes, XS blast radius, and low git history churn (4 commits in 90d, 1 routine fix) — composite of ~1.44 preserved at 1 per anchoring rule since Tier 1 signals are unchanged from prior assessment. Previous run (18)Risk Assessment: low (1/5) DetailsSingle go.mod line change by a Renovate bot with no protected paths, no CI workflow changes, minimal blast radius (XS), and low git history churn — dependency file change scores 3 on that dimension but is offset by all other signals near 1, yielding a composite of 1.37 which rounds to low. Previous run (19)Risk Assessment: low (1/5) DetailsSingle-file, 2-line indirect dependency security bump by a trusted bot with no blast radius and a well-established Renovate pattern in this repo; composite score (Tier1 62% × 1.5 + Tier2 38% × 1.0 = 1.31) rounds to 1. Previous run (20)Risk Assessment: low (1/5) DetailsSingle-line indirect dependency version bump by authorized renovate bot resolving a CVE, with no protected paths, no logic changes, and a consistent low-risk history of equivalent renovate PRs in this repo. Previous run (21)Risk Assessment: low (1/5) DetailsBot-authored single-file indirect dependency security bump (go.mod only) with minimal churn, no protected or security-sensitive paths, and small blast radius yields composite 1.32 (0.62x1.25 + 0.38x1.43), consistent with the prior assessment score of 1. Previous run (22)Risk Assessment: low (1/5) DetailsBot-authored single-file indirect dependency bump (go.mod only) with minimal 30-day churn, no protected or security-sensitive paths, no CI changes, and zero revert history yields a composite score of 1.28 (0.62x1.25 + 0.38x1.33), rounding to 1. Previous run (23)Risk Assessment: low (1/5) DetailsBot-authored, single-file dependency bump with 2 lines changed, no protected paths, no security-sensitive code paths, and small blast radius; only the dependency file dimension slightly elevates Tier 1, yielding a composite of 1.37 that rounds to low. |
|
Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous runLooks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (2)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (3)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (4)Looks good to me Labels: Go dependency update PR follows established convention of including the go label alongside dependencies and renovate. <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (5)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (6)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (7)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (8)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (9)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (10)Looks good to me <!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (11)ReviewFindingsCritical
Next steps:
Previous run (12)ReviewFindingsCritical
Next steps:
Previous run (13)ReviewFindingsHigh
Next steps:
Previous run (14)ReviewFindingsCritical
Next steps:
Previous run (15)ReviewFindingsHigh
Next steps:
Previous run (16)ReviewFindingsCritical
Labels: Renovate-authored Go dependency bump. Next steps:
Previous run (17)ReviewFindingsCritical
Next steps:
Previous run (18)ReviewFindingsInfo
<!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJmaW5kaW5ncyI6W119 --> Previous run (19)ReviewFindingsHigh
Next steps:
Previous run (20)ReviewFindingsHigh
Next steps:
Previous run (21)ReviewFindingsHigh
Next steps:
Previous run (22)ReviewFindingsHigh
Next steps:
Previous run (23)ReviewFindingsHigh
Next steps:
|
4724e40 to
fa1df9a
Compare
fa1df9a to
561131b
Compare
561131b to
c58a906
Compare
c58a906 to
afda8b1
Compare
afda8b1 to
f36ad0d
Compare
Superseded by updated review
f36ad0d to
cb89133
Compare
Superseded by updated review
Superseded by updated review
fc3b1d5 to
3507f6e
Compare
Superseded by updated review
3507f6e to
ef98638
Compare
Superseded by updated review
ef98638 to
9a43040
Compare
Superseded by updated review
9a43040 to
11b4d2c
Compare
Superseded by updated review
11b4d2c to
9e98994
Compare
Superseded by updated review
9e98994 to
a3c3f9e
Compare
Superseded by updated review
a3c3f9e to
a3cde58
Compare
Superseded by updated review
a3cde58 to
4a3be36
Compare
Superseded by updated review
4a3be36 to
7049e11
Compare
Superseded by updated review
This PR contains the following updates:
v1.5.0→v1.5.2Rekor has an OOM Condition due to Unbounded gzip Decompression in Alpine APK Parsing Logic
CVE-2026-48702 / GHSA-47q9-m4ww-924m
More information
Details
Description
The
Package.Unmarshal()function inpkg/types/alpine/apk.godecompresses the signature and control gzip members of an APK file into in-memory buffers without bounding the total decompressed size. The existingmax_apk_metadata_sizecheck (default 1MB) is only applied to individual tar entry header sizes after decompression completes, so it does not prevent a decompression bomb from consuming unbounded heap memory.An attacker can craft a gzip stream that compresses at a ~1000:1 ratio (e.g., 2MB compressed zeros → 2GB decompressed). When submitted as spec.package.content in an Alpine
ProposedEntry, the server decompresses the full payload into memory during request processing, triggering a fatal Go runtime out-of-memory error or OS OOM-kill that cannot be caught by the server's recover() middleware.This is reachable via two unauthenticated endpoints:
Both invoke
V001Entry.Canonicalize()→fetchExternalEntities()→apk.Unmarshal(packageData), which performs the unbounded decompression.Workarounds
There is no effective workaround. Setting
max_request_body_sizereduces but does not eliminate exposure due to the ~1000:1 compression ratio (a 1MB body limit still allows ~1GB heap allocation). Settingmax_apk_metadata_sizehas no effect on this vulnerability since the check is applied after decompression.Severity
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:HReferences
This data is provided by the GitHub Advisory Database (CC-BY 4.0).
Release Notes
sigstore/rekor (github.com/sigstore/rekor)
v1.5.2Compare Source
Features
Bug Fixes
Improvements
v1.5.1Compare Source
Features
Bug Fixes
Configuration
📅 Schedule: (UTC)
🚦 Automerge: Enabled.
♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about this update again.
This PR was generated by Mend Renovate. View the repository job log.