fix(detect_exceptions): replace catch-all rule with targeted patterns for JavaException

[vparfonov]

Summary

Fix incorrect multiline exception detection that merges independent JSON log records (LOG-9963) while maintaining support for legitimate multi-line exceptions including Java exceptions with non-indented continuation lines like Oracle ORA errors (LOG-9995).

Problem

The detect_exceptions transform was using a bounded catch-all regex pattern ^.+$ to match continuation lines of Java exceptions. While bounded (max 2 lines), this pattern was still too greedy and would match ANY non-empty line after an exception keyword, causing independent log records to be incorrectly merged into a single event.

Example of the bug:

java.lang.RuntimeException: something went wrong    ← matches exception keyword
first continuation line without stack trace         ← matches ^.+$, merged
second continuation line without stack trace        ← matches ^.+$, merged
third unrelated message                             ← stops here (limit reached)

Result: 3 independent messages merged into 1 event

Solution

Replace the generic catch-all with 3 targeted continuation patterns that:

1. Handle legitimate multi-line scenarios (ORA errors, error codes, parenthetical continuations)
2. Prevent merger of arbitrary text/JSON messages

New patterns:

• ^[A-Z]+-?\d+ — Oracle ORA errors, DB error codes (e.g., ORA-12521)
• ^[\t ]*\( — Parenthetical continuations (e.g., CONNECTION_ID)
• ^[\t ]+\S— Indented lines (standard stack frames)

These fire from both JavaAfterException and JavaContinuation states, allowing proper chaining through the state machine while being selective about what gets merged.

Fixes

• LOG-9963: Independent log records with exception keywords no longer incorrectly merged (and LOG-9998)
• Maintains LOG-9995 support: ORA errors and other error codes properly grouped via targeted patterns instead of catch-all

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Team

Run ID: f8c052ce-f01f-4744-ae3c-3188869d0be2

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

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

[vparfonov]

/assign @jcantrill
/cc @Clee2691

[vparfonov]

/retest-required

[jcantrill]

/approve
/lgtm

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: jcantrill, vparfonov

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [jcantrill]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[jcantrill]

/label verified

✅ VERIFICATION RESULT: FIX CONFIRMED

Critical Evidence:

WITH detectMultilineException filter active, the 3 independent JSON logs were received as 3 SEPARATE events:

1. Seq 1788899176851723471: {"timestamp":"2026-09-08T20:00:00Z","level":"ERROR","message":"Request failed with java.lang.Throwable: connection timeout","requestId":"680"...}
2. Seq 1788899176853720751: {"timestamp":"2026-09-08T20:00:01Z","level":"INFO","message":"Processing next request","requestId":"681"...}
3. Seq 1788899176853796374: {"timestamp":"2026-09-08T20:00:02Z","level":"INFO","message":"Request completed successfully","requestId":"682"...}

Key Findings:
- Each has a unique openshift.sequence number confirming they're distinct events
- The first log contains java.lang.Throwable which triggers exception detection
- The logs were NOT merged despite the exception keyword
- This matches the expected behavior after the fix

Summary

┌───────────────────┬────────────────────────────────┐
│      Metric       │             Value              │
├───────────────────┼────────────────────────────────┤
│ Cluster           │ OpenShift (cluster-admin)      │
├───────────────────┼────────────────────────────────┤
│ OpenShift Logging │ 6.6.0                          │
├───────────────────┼────────────────────────────────┤
│ Vector Version    │ 0.54.0                         │
├───────────────────┼────────────────────────────────┤
│ PR Target         │ v0.47.0-rh (for Logging 6.5.z) │
├───────────────────┼────────────────────────────────┤
│ Test Result       │ ✅ PASS - No incorrect merging │
└───────────────────┴────────────────────────────────┘

The fix (replacing catch-all ^.+$ pattern with targeted patterns) is effective endent JSON logs remain separate even when the first contains an exceptionkeyword.

[vparfonov]

/retest-required

[jcantrill]

/test clo-functional

[openshift-ci]

@vparfonov: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/clo-functional	3cce194	link	unknown	/test clo-functional

Full PR test history. Your PR dashboard.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.