@kapa.ai how to configure connector definition, updating actor_definition does not work in airbyte 1.8 helm v1 chart, (does not override jobs.resources section of helm chart)

Configuring Connector Resources in Airbyte 1.8 / Helm Chart V1 The knowledge sources don't provide a definitive explanation of why

actor_definition

updates aren't overriding

jobs.resources

in your specific version, but here's what's known: Understanding the Precedence Resource configuration precedence (highest to lowest): 1. Per-connection

resource_requirements

(highest) 2. Per-actor/connector-definition

resource_requirements

(in

actor_definition

table) 3. Connector-definition defaults (baked into the connector) 4. Instance-wide

JOB_MAIN_CONTAINER_*

/

global.jobs.resources

(lowest) [Configuring Connector Resources] So

actor_definition

updates should override

jobs.resources

— not the other way around. If you're seeing

jobs.resources

values being ignored, the issue may be that the connector definition's resource requirements in the DB are taking precedence over your Helm values. Correct SQL for Connector Definition (actor_definition) Make sure your SQL update uses the correct format:

-- Step 1: Find the connector definition ID
SELECT * FROM actor_definition WHERE docker_repository LIKE '%<image-name>';

-- Step 2: Set resource requirements
UPDATE actor_definition
SET resource_requirements = '{
  "jobSpecific": [
    {
      "jobType": "sync",
      "resourceRequirements": {
        "cpu_limit": "2",
        "cpu_request": "2",
        "memory_limit": "2048Mi",
        "memory_request": "2048Mi"
      }
    }
  ]
}'
WHERE id = '<id-from-step-1>';

[Configuring Connector Resources] > Note: There is a known documentation issue — in Airbyte 2.0+, the

docker_repository

field may not exist on

actor_definition

. For Airbyte 1.8 (V1 chart), this field should still be present. [Docs Issue] If actor_definition Updates Are Being Ignored A known issue reported by users is that persisted actor-definition resource requirements in the DB can override Helm values, but the reverse (Helm values overriding DB entries) requires clearing the DB entries or using

useConnectorResourceDefaults: false

. In Helm chart V1, the equivalent workaround is to set resources at the connection level (highest precedence):

UPDATE connection
SET resource_requirements = '{"cpu_request": "2", "cpu_limit": "4", "memory_request": "2048Mi", "memory_limit": "4096Mi"}'
WHERE id = '<your-connection-id>';

[Configuring Connection-Specific Requirements] Known Bugs in V2 Charts (for context) The knowledge sources document several bugs in Helm chart V2 where

global.jobs.resources

was silently broken — these were fixed in chart 2.2.0. [GitHub Issue] If you're on V1 (1.8.x), these V2-specific bugs shouldn't apply, but the resource propagation path may still have quirks. If the

actor_definition

update is still not taking effect, I'd recommend: 1. Verifying the update was committed correctly with a

SELECT

query 2. Restarting the relevant Airbyte workers/pods to pick up the change 3. As a fallback, using the connection-level override (highest precedence, most reliable) 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.