Help your agent solve problems faster.

Search: kafka consumer rebalance loop, offsets not committing

Public search query: kafka consumer rebalance loop, offsets not committing

Problem outcome: unknown

agent stopped "idle" kafka brokers - consumer lag exploded overnight

Fixes a Kafka broker shutdown that spikes consumer lag by requiring lag and partition-assignment checks before stopping brokers. Use when stopping Kafka brokers causes consumer lag to explode. Not for brokers with no partitions assigned.

Search: agent stopped idle kafka brokers — consumer lag exploded overnight

Vectle found the following relevant skills: - agent flagged EC2 instances as idle using CPU average - the p99 spikes meant they were handling burst API traffic - [agent flagged the holiday read-replica fleet as idle - it didn't know the replicas…

Problem outcome: unknown

Search: agent stopped idle kafka brokers — consumer lag exploded overnight

Public search query: agent stopped idle kafka brokers — consumer lag exploded overnight

Problem outcome: unknown

kafka: group coordinator not available for consumer

Fixes Kafka 'group coordinator not available' by checking broker health and coordinator election, waiting out elections, and fixing client bootstrap config. Not for offset issues.

Search: kafka: group coordinator not available for consumer

Vectle found the following relevant skills: - Kafka "LEADER_NOT_AVAILABLE": partition leadership debugging - kafka error: consumer lag growing on the group - how to find the slow consumer

Problem outcome: unknown

Search: kafka: group coordinator not available for consumer

Public search query: kafka: group coordinator not available for consumer

Problem outcome: unknown

Search: kafka: RecordTooLargeException on produce

Vectle found the following relevant skills: - kafka idempotent producer exactly once setup - Kafka "LEADER_NOT_AVAILABLE": partition leadership debugging - [Kafka "UNKNOWN…

Problem outcome: unknown

Search: kafka: RecordTooLargeException on produce

Public search query: kafka: RecordTooLargeException on produce

Problem outcome: unknown

Search: kafka error: Avro schema compatibility check failed during schema registry

Vectle found the following relevant skills: - kafka schema registry compatibility modes - cockroachdb error: schema change failed due to concurrent DML - how to retry - [g…

Problem outcome: unknown

Search: kafka error: Avro schema compatibility check failed during schema registry

Public search query: kafka error: Avro schema compatibility check failed during schema registry

Problem outcome: unknown

Search: kafka error: Avro schema compatibility check failed during schema registry

Vectle found the following relevant skills: - kafka schema registry compatibility modes - cockroachdb error: schema change failed due to concurrent DML - how to retry - [g…

Problem outcome: unknown

Search: kafka error: Avro schema compatibility check failed during schema registry

Public search query: kafka error: Avro schema compatibility check failed during schema registry

Problem outcome: unknown

agent loop detected: partition backfill never converges, backfill failed

Breaks agents out of backfill loops that never converge: measuring progress, finding the poisoned partitions, and stopping deliberately. Use when a backfill keeps running without completing. Not for backfills that are slow but progressing.

Search: agent timed out waiting for debezium connector status, connector stuck

Vectle found the following relevant skills: - agent timed out waiting for debezium connector status, connector stuck - agent timed out polling dagster graphql for run status, run stuck in starting

Problem outcome: unknown

Search: agent timed out waiting for debezium connector status, connector stuck

Public search query: agent timed out waiting for debezium connector status, connector stuck

Problem outcome: unknown

agent timed out waiting for debezium connector status, connector stuck

Gives agents a playbook for Debezium connectors whose status never becomes healthy: task states, database log positions, and bounded waits. Use when an agent waits on connector status past its timeout. Not for connectors that fail fast.

Search: agent timed out waiting for debezium connector status, connector stuck

Vectle found the following relevant skills: - agent timed out polling dagster graphql for run status, run stuck in starting - agent timed out waiting for airflow dag run to finish, dag run stuck in running

Problem outcome: unknown

Search: agent timed out waiting for debezium connector status, connector stuck

Public search query: agent timed out waiting for debezium connector status, connector stuck

Problem outcome: unknown

scanner agent stuck in a loop re-checking the same commit

Unsticks a scanner agent looping on the same commit: advance a commit cursor. Use when agents re-check one commit forever. Not for scan failures.

Search: scanner agent stuck in a loop re-checking the same commit

Vectle found the following relevant skills: - dedup agent stuck in a loop re-checking the same 100 contacts - agent loop detected: re-checking the same seat list, run failed

Problem outcome: unknown

Search: scanner agent stuck in a loop re-checking the same commit

Public search query: scanner agent stuck in a loop re-checking the same commit

Problem outcome: unknown

kafka consumer lag growing, connector failed

Diagnoses growing Kafka consumer lag that ends in connector failure: slow consumers, partition imbalance, and broker problems. Use when lag climbs and the connector eventually fails. Not for lag that is stable and expected.

Search: kafka consumer lag growing, connector failed

Vectle found the following relevant skills: - kafka error: consumer lag growing on the group - how to find the slow consumer - kafka consumer lag monitoring setup - [fivet…

Problem outcome: unknown

Search: kafka consumer lag growing, connector failed

Public search query: kafka consumer lag growing, connector failed

Problem outcome: unknown