Bound multipart text/plain field materialization in Jersey AppSec by jandro996 · Pull Request #11944 · DataDog/dd-trace-java

jandro996 · 2026-07-14T11:09:28Z

What Does This Do

Replaces the unbounded bodyPart.getValue() call in MultiPartHelper.collectBodyPart (jersey2 and jersey3 AppSec modules) with readContent(bodyPart), which reads the part through the existing byte-capped MultipartContentDecoder (bounded by MAX_CONTENT_BYTES, backed by Config.get().getAppSecMaxFileContentBytes()).
Caps the total number of values accumulated across the bodyMap (not just distinct field names) at MAX_FILES_TO_INSPECT, symmetric to the existing cap on filesContent. Repeating the same field name across multiple parts no longer bypasses the limit — every collected value counts against the cap.
Derives the bodyMap key via FormDataContentDisposition.getName() (a safe field accessor obtained through the existing malformed-disposition guard) instead of the unguarded bodyPart.getName(), which re-parses the disposition header and can throw on malformed input outside that guard.
Splits the charset-handling helper into readContent() (used only by the bodyMap/text-field path, defaults undeclared charset to UTF-8 to match Jersey's own getValue() behavior) and readFileContent() (used only by the filesContent path, keeps MultipartContentDecoder's own charset fallback for parity with file content in every other multipart helper). The UTF-8 default is applied locally in Jersey's helper only — the shared MultipartContentDecoder fallback (Charset.defaultCharset()) is untouched, since it's also used by Tomcat, Netty, Vert.x, RESTEasy and commons-fileupload.
Applies all of the above identically to both dd-java-agent/instrumentation/jersey/jersey-appsec/jersey-appsec-2.0 (javax.ws.rs) and jersey-appsec-3.0 (jakarta.ws.rs) — the two MultiPartHelper.java files remain byte-for-byte identical modulo the namespace.
Updates MultiPartHelperTest.groovy in both modules to mock bodyPart.getEntityAs(InputStream) instead of bodyPart.getValue(), and adds test cases for: value truncation beyond MAX_CONTENT_BYTES, the total-value cap (MAX_FILES_TO_INSPECT) blocking further accumulation regardless of distinct field names, and charset defaulting for undeclared-charset text fields vs. file content.

Motivation

Jersey 2/3 AppSec instrumentation materialized text/plain multipart field values via the unbounded getValue() and accumulated an unlimited number of values into the body map before any WAF limit was applied. A crafted multipart request with a very large text/plain field, or a large number of parts (even repeating the same field name), could exhaust heap/CPU before hitting any bound.

Additional Notes

Scope is intentionally limited to Jersey. RESTEasy's MultipartFormDataReaderInstrumentation has a related unbounded-read pattern (getBodyAsString() on all parts) that is out of scope for this PR and should be addressed as a follow-up.
The charset default for undeclared-charset text/plain fields went through a short-lived regression: an early attempt changed the shared MultipartContentDecoder's fallback to UTF-8 directly, which would have silently altered production behavior for Tomcat, Netty, Vert.x, RESTEasy and commons-fileupload (that fallback is Charset.defaultCharset() by deliberate design, per PR Add server.request.body.files_content AppSec address for Tomcat and Netty multipart #11198). This was reverted; the UTF-8 default is now applied only within Jersey's own helper.

Contributor Checklist

Format the title according to the contribution guidelines
Assign the type: and (comp: or inst:) labels in addition to any other useful labels
Avoid using close, fix, or any linking keywords when referencing an issue
Use solves instead, and assign the PR milestone to the issue
Update the CODEOWNERS file on source file addition, migration, or deletion
Update public documentation with any new configuration flags or behaviors

Jira ticket: APMSP-3237

Note:Once your PR is ready to merge, add it to the merge queue by commenting /merge./merge -c cancels the queue request. /merge -f --reason "reason" skips all merge queue checks; please use this judiciously, as some checks do not run at the PR-level. For more information, see this doc.

Jersey 2/3 AppSec instrumentation read text/plain multipart field values via the unbounded getValue() and accumulated an unlimited number of distinct field names, allowing a crafted multipart request to exhaust heap/CPU before WAF limits apply. Reuse the existing byte-capped MultipartContentDecoder and the configured file-content limits (MAX_CONTENT_BYTES, MAX_FILES_TO_INSPECT) to bound both the size of each field value and the number of distinct field names collected.

jandro996 · 2026-07-14T11:09:42Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit:124e9188ab

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".

datadog-datadog-us1-prod · 2026-07-14T11:23:57Z

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 69.73% (+12.53%)

_{This comment will be updated automatically if new data arrives.

🔗 Commit SHA: d11f0a2 | Docs | Datadog PR Page | Give us feedback!}

dd-octo-sts · 2026-07-14T11:30:48Z

🟢 Java Benchmark SLOs — All performance SLOs passed

Suite	Status
Startup	🟢 pass

SLO thresholds are defined here based on automatically generated metrics. A warning is raised when results are within 5% of the threshold.

PR vs. master results

Scenario	Candidate	master	Δ (95% CI of mean)
startup:insecure-bank:iast:Agent	13.99 s	13.98 s	[-0.8%; +0.9%] (no difference)
startup:insecure-bank:tracing:Agent	12.90 s	12.98 s	[-1.3%; +0.0%] (no difference)
startup:petclinic:appsec:Agent	16.98 s	16.87 s	[-0.4%; +1.7%] (no difference)
startup:petclinic:iast:Agent	17.04 s	17.04 s	[-0.8%; +0.7%] (no difference)
startup:petclinic:profiling:Agent	16.80 s	16.83 s	[-1.3%; +0.9%] (no difference)
startup:petclinic:sca:Agent	17.02 s	16.77 s	[+0.6%; +2.4%] (maybe worse)
startup:petclinic:tracing:Agent	15.76 s	16.11 s	[-6.4%; +2.0%] (no difference)

Commit:d11f0a24 · CI Pipeline · Benchmarking Platform UI

Load and DaCapo benchmarks can be triggered manually in the GitLab pipeline. Results will appear in the Benchmarking Platform UI after completion.

…d getName() - Cap now bounds total accumulated values in the body map, not just distinct field names, closing a bypass where repeating a field name skipped the limit. - Derive the body map key from FormDataContentDisposition.getName() (a safe field accessor) instead of FormDataBodyPart.getName(), which re-parses the disposition header and can throw on malformed input. Addresses Codex review findings 1 and 3 on PR #11944.

…charset Charset.defaultCharset() depends on the JVM/platform locale and can differ from the UTF-8 default Jersey's own getValue() used, corrupting text captured by AppSec on JVMs whose platform charset isn't UTF-8. Addresses Codex review finding 2 on PR #11944.

jandro996 · 2026-07-14T12:04:16Z

@codex review

chatgpt-codex-connector · 2026-07-14T12:09:03Z

Codex Review: Didn't find any major issues. You're on a roll.

Reviewed commit:e25666acf5

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".

…latform charset" This reverts commit e25666a.

…ultipartContentDecoder MultipartContentDecoder is shared by Tomcat, Netty, Vert.x, RESTEasy and commons-fileupload. Its Charset.defaultCharset() fallback was a deliberate, reviewed decision (Manuel, PR #11198) - changing it in e25666a silently altered production behavior for all those integrations, not just Jersey. Revert the shared decoder change and instead default to UTF-8 only in Jersey's MultiPartHelper, matching Jersey's own getValue() default, by appending charset=UTF-8 to the contentType before calling the shared decoder.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit:06683b8439

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".

datadog-datadog-us1-prod

More details

Validated charset defaulting and cap boundary: contentTypeWithDefaultUtf8 correctly appends UTF-8 when no charset is declared and preserves explicit charsets unchanged; the totalBodyMapValues cap blocks ALL additions once the total reaches MAX_FILES_TO_INSPECT — including repeated field names — which is more restrictive than the PR description's "distinct-field-name" framing, but intentional per the in-code comment and test assertions.

Was this helpful? React 👍 or 👎

_{📊 Validated against 28 scenarios · Open Bits AI session}

_{🤖 Datadog Autotest · Commit 06683b8 · What is Autotest? · Any feedback? Reach out in #autotest}

readContent() served both the bodyMap (text field) and filesContent call sites, so the Jersey-local UTF-8 default leaked into file content too. Split it into readContent() (UTF-8 default, text fields) and readFileContent() (unmodified content type, parity with MultipartContentDecoder's fallback used by file content in every other multipart helper).

dd-octo-sts · 2026-07-24T11:38:31Z

/merge

gh-worker-devflow-routing-ef8351 · 2026-07-24T11:38:35Z

View all feedbacks in Devflow UI.

2026-07-24 11:38:35 UTC ℹ️ Start processing command /merge

2026-07-24 11:38:40 UTC ℹ️ MergeQueue: pull request added to the queue

The expected merge time in master is approximately 1h (p90).

2026-07-24 12:33:09 UTC ℹ️ MergeQueue: This merge request was merged

jandro996 added type: bug fix Bug fix comp: asm waf Application Security Management (WAF) labels Jul 14, 2026

chatgpt-codex-connectorBot reviewed Jul 14, 2026
View reviewed changes

jandro996 added 2 commits July 14, 2026 13:54

jandro996 added 2 commits July 14, 2026 14:16

Revert "fix: default multipart content decoding to UTF-8 instead of p…
26f0d25
…latform charset" This reverts commit e25666a.

jandro996 marked this pull request as ready for review July 14, 2026 13:28

jandro996 requested a review from a team as a code owner July 14, 2026 13:28

jandro996 requested review from ValentinZakharov, claponcet, manuel-alvarez-alvarez and vandonr and removed request for a teamJuly 14, 2026 13:28

chatgpt-codex-connectorBot reviewed Jul 14, 2026
View reviewed changes

Comment thread...c/jersey-appsec-2.0/src/main/java/datadog/trace/instrumentation/jersey2/MultiPartHelper.java Outdated

datadog-datadog-us1-prodBot reviewed Jul 14, 2026
View reviewed changes

jandro996 added 2 commits July 15, 2026 10:05

Merge branch 'master' into APMSP-3237
d11f0a2

manuel-alvarez-alvarez reviewed Jul 23, 2026
View reviewed changes

Comment thread...c/jersey-appsec-2.0/src/main/java/datadog/trace/instrumentation/jersey2/MultiPartHelper.java

manuel-alvarez-alvarez approved these changes Jul 24, 2026
View reviewed changes

jandro996 added this pull request to the merge queue Jul 24, 2026

github-merge-queueBot removed this pull request from the merge queue due to failed status checks Jul 24, 2026

gh-worker-dd-mergequeue-cf854dBot merged commit 84de0a8 into masterJul 24, 2026
586 checks passed

gh-worker-dd-mergequeue-cf854dBot deleted the APMSP-3237 branch July 24, 2026 12:33

github-actionsBot added this to the 1.65.0 milestone Jul 24, 2026

Uh oh!

Conversation

jandro996 commented Jul 14, 2026• edited by atlassian BotLoading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What Does This Do

Motivation

Additional Notes

Contributor Checklist

Uh oh!

jandro996 commented Jul 14, 2026

Uh oh!

chatgpt-codex-connectorBot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

datadog-datadog-us1-prodBot commented Jul 14, 2026• edited by datadog-official BotLoading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dd-octo-stsBot commented Jul 14, 2026• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🟢 Java Benchmark SLOs — All performance SLOs passed

Uh oh!

jandro996 commented Jul 14, 2026

Uh oh!

chatgpt-codex-connectorBot commented Jul 14, 2026

Uh oh!

chatgpt-codex-connectorBot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

datadog-datadog-us1-prodBot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

dd-octo-stsBot commented Jul 24, 2026

Uh oh!

gh-worker-devflow-routing-ef8351Bot commented Jul 24, 2026• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

jandro996 commented Jul 14, 2026•
edited by atlassian Bot
Loading

datadog-datadog-us1-prodBot commented Jul 14, 2026•
edited by datadog-official Bot
Loading

dd-octo-stsBot commented Jul 14, 2026•
edited
Loading

gh-worker-devflow-routing-ef8351Bot commented Jul 24, 2026•
edited
Loading