This message was deleted.

Hi team, I have set up Airbyte OSS locally using abctl. When I try to access the API to get an access token, I receive a 400 Bad Request error, even though I have provided the correct client ID and secret. • Airbyte version: 1.8.2 • Running on port: 8080 • Web app URL: http://localhost:8080/ I am able to log in to the Airbyte UI using my email and password, and from the Applications page I can see the default application and generate an access token. However, I am not able to see the client ID and client secret there. I am only able to retrieve the client ID and secret via abctl. (In the cURL request below, I have added placeholders to hide the actual client ID and secret, but in my setup I am passing the real values retrieved from abctl) Here’s the cURL request I’m using:

curl --location '<http://localhost:8080/api/v1/applications/token>' \
--header 'Content-Type: application/json' \
--data '{"client_id":"<placeholder_client_id>","client_secret":"<placeholder_client_secret>"}

Hi! Is there a better way to see breaking changes in the API? Like a central page I can review before upgrading? I'd love to see this in the release notes if possible I read through the recent release notes and did not see any mention of the Airbyte API breaking. We learned about that after the fact from this page: https://reference.airbyte.com/reference/standalone-server-deprecation-and-migration-to-airbyte-server We are using the

/api/v1/workspaces/list

endpoint and were surprised by suddenly getting a 404 after the upgrade.

Hi airbyte team, am looking at getting twitter ads data, and saw there was a start previously, wondering if there are any plans at all to continue on that anytime soon? 🙏 https://github.com/airbytehq/airbyte/discussions/35262

Could you share a bit more about what you need with HubSpot and Airbyte? For example: setting up the HubSpot source, migrating streams, reverse ETL to HubSpot, or a specific stream/issue. A few pointers while you decide: - HubSpot destination for data activation (reverse ETL) is in early access on Airbyte Cloud and Self-Managed 1.8+, supporting Contacts, Companies, Deals, and Custom Objects upserts. [[HubSpot destination](https://docs.airbyte.com/integrations/destinations/hubspot); [Data activation](https://docs.airbyte.com/platform/move-data/elt-data-activation)] - Source HubSpot has recent changes: deprecated some v1 contacts streams in 5.0.0 and migrated marketing_emails to v3 with breaking schema changes in 6.0.0 (requires schema refresh/reset). [[HubSpot migration](https://docs.airbyte.com/integrations/sources/hubspot-migrations)] - Known topics: API key deprecation (move to Private App/OAuth), propertiesWithHistory not fetched by default, requests for custom-object associations, meetings links stream, association limits, and archived owners userIdIncludingInactive. [[API key deprecation thread](https://discuss.airbyte.io/t/3018); [PropertiesWithHistory forum](https://discuss.airbyte.io/t/3786); [Custom object associations](https://github.com/airbytehq/airbyte/issues/31509); [Meetings links stream](https://github.com/airbytehq/airbyte/issues/56360); [Deal associations limit](https://github.com/airbytehq/airbyte/issues/56447); [Archived owners fix](https://github.com/airbytehq/airbyte/issues/55818)] Tell me your goal and environment (Cloud vs OSS version), and I’ll give exact steps.

hey there! I am struggling with getting my source (Airtable) to send data through incremental sync despite clearly having cursor values defined. Where's the best place to start? I am using the pre-existing connector and haven't made any modifications to it

Hi, Does anyone have experience getting incremental syncs working properly with Microsoft Business Central Cloud OData endpoints? I'm trying to use the $filter parameter on lastModifiedDate, but I can't figure out how exactly to get that to work with Incremental Sync

Hi, We are using Airbyte - Cloud Version. we’re seeing intermittent 401 Unauthorized mid-sync (usually after a few pages) across multiple declarative HTTP sources. A full rerun later succeeds with no config changes. One source uses Bearer API key , another uses *Basic*; both have concurrency=1 with offset pagination. Any guidance on what could be the root cause for these kind of errors and the best possible solution to mitigate these kind of failures?

Hi team, I’m running self-hosted Airbyte and noticed some confusion around the API endpoints. The reference docs show Cloud endpoints like

GET /v1/sources

, but when I try those on self-hosted (e.g.

<https://my-airbyte/api/v1/sources>

), I get:

{
  "message": "Forbidden",
  "_links": {
    "self": {
      "href": "/api/v1/sources"
    }
  },
  "_embedded": {
    "errors": [
      {
        "message": "Forbidden"
      }
    ]
  }
}

If I switch to the RPC-style call (

POST /api/v1/sources/list

with a workspaceId), it works. This raises a few questions: • Are the docs at reference.airbyte.com intended only for Cloud? • For self-hosted OSS, should we always be using the RPC-style endpoints like

/api/v1/sources/list

,

/api/v1/connections/create

, etc.? • Is there a plan to unify the Cloud and OSS APIs in the future? Also, when I tried to create a Google Search Console source via API, I hit this error:

curl --location '<https://dev.airbyte.jepto.com/api/v1/sources/create>' \
--header 'Content-Type: application/json' \
--header 'Authorization: ••••••' \
--data '{
    "configuration": {
        "sourceType": "google-search-console",
        "authorization": {
            "auth_type": "Client",
            "client_id": "...redacted...",
            "client_secret": "...redacted...",
            "refresh_token": "...redacted..."
        },
        "site_urls": ["<https://www.mysite.com/>"],
        "start_date": "2021-05-01"
    },
    "name": "Sample GSC Connection",
    "sourceDefinitionId": "eb4c9e00-db83-4d63-a386-39cfa91012a8",
    "workspaceId": "48d59dac-ba6e-466f-8cff-9f2a291cfa1c"
}'

The response I got was:

{
  "message": "Internal Server Error: null",
  "exceptionClassName": "java.lang.NullPointerException",
  "exceptionStack": [],
  "rootCauseExceptionStack": []
}

Is this related to a mismatch between the Cloud docs vs the OSS API, or am I formatting the

sources/create

payload incorrectly for OSS?

#C021JANJ6TY problem with replication tables from MySQL 5.7 to Snowflake

Hi! I'm working on a custom Gmail connector and having trouble with

AddFields

transformations on a substream. The substream returns a json object as expected, but it doesn't add any fields - not even a static one. Setup: • DeclarativeStream with SubstreamPartitionRouter • Parent stream provides

composite_key

via partition_field • Simple AddFields transformation to add static

test_field

What's not working: yaml

transformations: 
  - type: AddFields
    fields:
      - path: ["test_field"]
        value: "hello world?"

The

test_field

never appears in the output records, even though: • Schema includes the field with correct type •

additionalProperties: true

is set • Field is not in

required

array • Partition router is working (API calls succeed) • I am able to use AddFields for another substream

Hi , Using the HTTP Connector Builder with a parent/child (Substream) setup: parent is date-sliced; child filters by the parent’s ID. When both run, the child stays “Queued” until the parent finishes all partitions, leading to long runs and heartbeat warnings. Questions: 1. Is there any way in Builder to have the child emit in parallel with the parent (e.g., interleaving slices or progressive handoff of parent IDs), or is the “child waits” behavior inherent? 2. If inherent, what’s the recommended pattern to run in parallel?

Hi hi, since 10th of September my Airbyte (local deployment) doesn’t run syncs. Syncs with connectors Shopify, Google sheet just got stuck since that time, before extraction and everything worked properly. I checked the logs and they look differently now. Airbyte version: 0.50.46 On the first screenshot: That’s the stuck sync, you see 89 hours. Second screenshot: normal sync from 9th of September. It has “start replication” and doesnt have “start check” Third screenshot: sync from today, which also gets stuck and looks different. It doesn’t have “start replication” but has “start check”. What is happening and what should I do to again have the data sync? Thank you in advance

I have a query API that requires users to set the unique system ID for the query to fetch. I want to be able to allow users to generate as many streams as they want during source setup by giving each the query name (to be used as the stream name) and ID that should be fetched. How do I configure a template for this?

Are people still having issues with MongoDB source incremental load after the update?

Hi, I'm using Airbyte open source to sync a Magento 2 MariaDB database via CDC. Magento 2 creates numerous temporary tables during my store operations (imports, indexing, catalog operations, etc.). These temporary tables get written to my MariaDB's binary log along with my regular tables. My Problem Flow:

1. Magento creates temp table in my database → MariaDB writes

TABLE_MAP

event to my binlog

2. Magento writes data to temp table → MariaDB writes

WRITE_ROWS

events to my binlog

3. Magento drops temp table → Table no longer exists in my database

4. My Airbyte reads binlog → Finds

WRITE_ROWS

events for table ID (e.g 223)

5. Airbyte looks for table metadata → Can't find

TABLE_MAP

event in its reading window

6. I get this error: "No TableMapEventData has been found for table id:223"

Is there a way for me to tell Airbyte to ignore those temp tables? I can't find anything in the UI. Thank you in advance.

Hi. Is it possible to get a parameter in Airbyte open source in order to kill all sync running for more than xx hours ? There is no timeout and sometimes the sync is blocked for ages without errors. (destination is bigquery and sources are mainly postgresql, gsheet, hubspot). Thanks

Hi, I’m trying to connect to ClickHouse Cloud as a destination. It worked in July, but since then I’ve had this error:

2025-09-15 15:01:37 info 
2025-09-15 15:01:37 info Connector exited, processing output
2025-09-15 15:01:37 info ----- START CHECK -----
2025-09-15 15:01:37 info 
2025-09-15 15:01:37 info Output file jobOutput.json found
2025-09-15 15:01:37 info Connector exited with exit code 0
2025-09-15 15:01:37 info Reading messages from protocol version 0.2.0
2025-09-15 15:01:37 info INFO main i.m.c.e.DefaultEnvironment(<init>):168 Established active environments: [k8s, cloud, cli, destination, connector]
2025-09-15 15:01:37 info INFO main i.a.c.AirbyteConnectorRunnable(run):33 Executing class io.airbyte.cdk.load.check.CheckOperation operation.
2025-09-15 15:01:37 warn WARN main i.a.c.l.c.CheckOperation(execute):58 Caught throwable during CHECK java.lang.IllegalArgumentException: Failed to insert expected rows into check table. Actual written: 0
	at io.airbyte.integrations.destination.clickhouse.check.ClickhouseChecker.check(ClickhouseChecker.kt:48) ~[io.airbyte.airbyte-integrations.connectors-destination-clickhouse.jar:?]
	at io.airbyte.integrations.destination.clickhouse.check.ClickhouseChecker.check(ClickhouseChecker.kt:20) ~[io.airbyte.airbyte-integrations.connectors-destination-clickhouse.jar:?]
	at io.airbyte.cdk.load.check.CheckOperation.execute(CheckOperation.kt:48) [bulk-cdk-core-load-0.1.20.jar:?]
	at io.airbyte.cdk.AirbyteConnectorRunnable.run(AirbyteConnectorRunnable.kt:34) [bulk-cdk-core-base-0.1.20.jar:?]
	at picocli.CommandLine.executeUserObject(CommandLine.java:2030) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine.access$1500(CommandLine.java:148) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine$RunLast.executeUserObjectOfLastSubcommandWithSameParent(CommandLine.java:2465) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine$RunLast.handle(CommandLine.java:2457) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine$RunLast.handle(CommandLine.java:2419) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine$AbstractParseResultHandler.execute(CommandLine.java:2277) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine$RunLast.execute(CommandLine.java:2421) [picocli-4.7.6.jar:4.7.6]
	at picocli.CommandLine.execute(CommandLine.java:2174) [picocli-4.7.6.jar:4.7.6]
	at io.airbyte.cdk.AirbyteDestinationRunner$Companion.run(AirbyteConnectorRunner.kt:289) [bulk-cdk-core-base-0.1.20.jar:?]
	at io.airbyte.cdk.AirbyteDestinationRunner$Companion.run$default(AirbyteConnectorRunner.kt:75) [bulk-cdk-core-base-0.1.20.jar:?]
	at io.airbyte.integrations.destination.clickhouse.ClickhouseDestinationKt.main(ClickhouseDestination.kt:10) [io.airbyte.airbyte-integrations.connectors-destination-clickhouse.jar:?]

Stack Trace: java.lang.IllegalArgumentException: Failed to insert expected rows into check table. Actual written: 0
	at io.airbyte.integrations.destination.clickhouse.check.ClickhouseChecker.check(ClickhouseChecker.kt:48)
	at io.airbyte.integrations.destination.clickhouse.check.ClickhouseChecker.check(ClickhouseChecker.kt:20)
	at io.airbyte.cdk.load.check.CheckOperation.execute(CheckOperation.kt:48)
	at io.airbyte.cdk.AirbyteConnectorRunnable.run(AirbyteConnectorRunnable.kt:34)
	at picocli.CommandLine.executeUserObject(CommandLine.java:2030)
	at picocli.CommandLine.access$1500(CommandLine.java:148)
	at picocli.CommandLine$RunLast.executeUserObjectOfLastSubcommandWithSameParent(CommandLine.java:2465)
	at picocli.CommandLine$RunLast.handle(CommandLine.java:2457)
	at picocli.CommandLine$RunLast.handle(CommandLine.java:2419)
	at picocli.CommandLine$AbstractParseResultHandler.execute(CommandLine.java:2277)
	at picocli.CommandLine$RunLast.execute(CommandLine.java:2421)
	at picocli.CommandLine.execute(CommandLine.java:2174)
	at io.airbyte.cdk.AirbyteDestinationRunner$Companion.run(AirbyteConnectorRunner.kt:289)
	at io.airbyte.cdk.AirbyteDestinationRunner$Companion.run$default(AirbyteConnectorRunner.kt:75)
	at io.airbyte.integrations.destination.clickhouse.ClickhouseDestinationKt.main(ClickhouseDestination.kt:10)

2025-09-15 15:01:37 info INFO main i.a.c.AirbyteConnectorRunnable(run):46 Flushing output consumer prior to shutdown.
2025-09-15 15:01:37 info INFO main i.a.c.AirbyteConnectorRunnable(run):48 Completed integration: airbyte/destination-clickhouse.
2025-09-15 15:01:37 info Checking for optional control message...
2025-09-15 15:01:37 info Optional control message not found. Skipping...
2025-09-15 15:01:37 info Writing output of ce0d828e-1dc4-496c-b122-2da42e637e48_96361554-8782-4233-b8e4-10373f363714_0_check to the doc store
2025-09-15 15:01:37 info Marking workload ce0d828e-1dc4-496c-b122-2da42e637e48_96361554-8782-4233-b8e4-10373f363714_0_check as successful
2025-09-15 15:01:37 info 
2025-09-15 15:01:37 info Deliberately exiting process with code 0.
2025-09-15 15:01:37 info ----- END CHECK -----
2025-09-15 15:01:37 info

Do you have any ideas?

There is not another solution than create views to have the same naming with small letters after replicating from mysql to snowflake ?

#C021JANJ6TY how to get the same table names from source MySQL to destination Snowflake, because after replication all tables and columns are in uppercase letters in snowflake.

Hello. I get this error with bigquery to s3 connectors : com.amazonaws.services.s3.model.AmazonS3Exception: The XML you provided was not well-formed or did not validate against our published schema @kapa.ai

Hey team 👋 I’m working with Airbyte OSS and need to fetch the required input fields for creating a source. For example — if I pass a sourceDefinitionId, I’d like to get back the list of required and optional fields needed to create that source. Is there an endpoint in OSS that provides this? I couldn’t find one in the docs

Missing Facebook Ads Purchase Data Since July 29th

I'm following up on the issue discussed in this thread: https://airbytehq.slack.com/archives/C021JANJ6TY/p1754590543220719 We are approaching almost two months since this problem was first reported, and it continues to be a critical issue for us. We're in the same position as many other users and are strongly considering a move to Fivetran if a fix isn't implemented for this problem.

Issue Summary: AsyncRetriever in DynamicSchemaLoader Missing Required stream_slicer Parameter Problem We're implementing dynamic schema generation for custom query streams that use the AsyncRetriever pattern. The schema needs to be inferred from CSV headers by executing the query during stream template generation. However, we're consistently getting this error: Encountered an error while discovering streams. Error: ModelToComponentFactory.create_async_retriever() missing 1 required keyword-only argument: 'stream_slicer' Configuration Context - Connector Type: Declarative YAML connector using low-code framework - Stream Type: Dynamic query streams that execute async jobs returning CSV data - Schema Requirement: Must dynamically infer schema from CSV headers during stream discovery/template generation Current YAML Structure (simplified): dynamic_streams: - type: DynamicDeclarativeStream name: custom_query_streams stream_template: retriever: type: AsyncRetriever # Main retriever config works fine schema_loader: type: DynamicSchemaLoader retriever: type: AsyncRetriever stream_slicer: ??? # <-- Issue is here # Rest of AsyncRetriever config for schema discovery What We've Tried 1. Empty partition_router: [] 2. stream_slicer with type: SinglePartitionRouter (doesn't exist) 3. stream_slicer with ListPartitionRouter containing single value 4. stream_slicer with CartesianProductStreamSlicer and empty array 5. Omitting stream_slicer entirely The Core Question How do we properly configure a "no-op" or single-partition stream_slicer for an AsyncRetriever that's being used inside a DynamicSchemaLoader? The retriever needs to execute the query once during discovery to get the CSV headers for schema inference, but the CDK requires a stream_slicer parameter even though we don't need actual partitioning. Technical Details - The main stream's AsyncRetriever works fine (it uses partition_router) - The issue only occurs with AsyncRetriever inside DynamicSchemaLoader - The error suggests it's a required parameter at the factory level (ModelToComponentFactory.create_async_retriever()) - Documentation doesn't clearly specify what type of stream_slicer to use for this scenario

Anyone having an issue logging into Cloud w/SSO? I'd open a ticket but can't login to reach the navigation bar!

Hey all, I am having issues with Elasticsearch. I am able to create the source and the test passes. Whenever I proceed to creating a connection though I get the following errors:

Internal message: java.lang.RuntimeException: io.airbyte.integrations.source.elasticsearch.UnsupportedDatatypeException: Cannot map unsupported data type to Airbyte data type: completion
Failure type: system_error

java.lang.RuntimeException: io.airbyte.integrations.source.elasticsearch.UnsupportedDatatypeException: Cannot map unsupported data type to Airbyte data type: completion
	at io.airbyte.integrations.source.elasticsearch.typemapper.ElasticsearchTypeMapper.lambda$formatJSONSchema$0(ElasticsearchTypeMapper.java:133)
	at java.base/java.util.Iterator.forEachRemaining(Iterator.java:133)
	at io.airbyte.integrations.source.elasticsearch.typemapper.ElasticsearchTypeMapper.formatJSONSchema(ElasticsearchTypeMapper.java:129)
	at io.airbyte.integrations.source.elasticsearch.typemapper.ElasticsearchTypeMapper.lambda$formatJSONSchema$0(ElasticsearchTypeMapper.java:131)
	at java.base/java.util.Iterator.forEachRemaining(Iterator.java:133)
	at io.airbyte.integrations.source.elasticsearch.typemapper.ElasticsearchTypeMapper.formatJSONSchema(ElasticsearchTypeMapper.java:129)
	at io.airbyte.integrations.source.elasticsearch.ElasticsearchSource.discover(ElasticsearchSource.java:78)
	at io.airbyte.cdk.integrations.base.IntegrationRunner.runInternal(IntegrationRunner.java:159)
	at io.airbyte.cdk.integrations.base.IntegrationRunner.run(IntegrationRunner.java:125)
	at io.airbyte.integrations.source.elasticsearch.ElasticsearchSource.main(ElasticsearchSource.java:37)
Caused by: io.airbyte.integrations.source.elasticsearch.UnsupportedDatatypeException: Cannot map unsupported data type to Airbyte data type: completion
	at io.airbyte.integrations.source.elasticsearch.typemapper.ElasticsearchTypeMapper.formatJSONSchema(ElasticsearchTypeMapper.java:127)
	at io.airbyte.integrations.source.elasticsearch.typemapper.ElasticsearchTypeMapper.lambda$formatJSONSchema$0(ElasticsearchTypeMapper.java:131)
	... 9 more

I understand that there's an issue with the "completion" data type. It looks like it's not being handled by the Airbyte connector. But I cannot fork the connector in the connector builder nor can I edit any configs in the source. Does anyone have any advice on what I should do? Do I need to create my own connector from scratch?

Hello community 👋, I’m running into an issue using an external S3-compatible storage (Scaleway Object Storage) with Airbyte 1.8.x and the Helm chart V1. Here’s what I’ve done: • In

values.yaml

under `global.storage`:

global:
  storage:
    type: "S3"
    secretName: "airbyte-config-secrets"
    bucket:
      log: airbyte-s3
      state: airbyte-s3
      workloadOutput: airbyte-s3
    s3:
      region: fr-par
      endpoint: <https://s3.fr-par.scw.cloud>
      authenticationType: credentials

• In the Kubernetes secret

airbyte-config-secrets

, I added my

s3-access-key-id

and

s3-secret-access-key

. • I tried forcing

S3_PATH_STYLE_ACCESS=true

to workaround DNS resolution issues, but the worker still tries to reach

<http://airbyte-s3.s3.fr-par.amazonaws.com|airbyte-s3.s3.fr-par.amazonaws.com>

and fails. My understanding: the Helm chart V1 seems designed for AWS S3, and the custom endpoint is not picked up correctly. My question: • Has anyone successfully configured Airbyte 1.8.x with Helm chart V1 to use an external S3-compatible storage directly (without an intermediate MinIO)? • If yes, could you share a working

values.yaml

and secret example or your process ? Thanks a lot for your feedback 🙏

Hey everyone! Is there a way to configure the backfillPreference to enabled via the airbyte api?

neeraj@airbyte-dev:~$ sudo abctl local install INFO Thanks for using Airbyte! Anonymous usage reporting is currently enabled. For more information, please see https://docs.airbyte.com/telemetry INFO Using Kubernetes provider: Provider: kind Kubeconfig: /root/.airbyte/abctl/abctl.kubeconfig Context: kind-airbyte-abctl SUCCESS Found Docker installation: version 28.4.0 SUCCESS Existing cluster 'airbyte-abctl' found SUCCESS Cluster 'airbyte-abctl' validation complete INFO Patching image airbyte/db:1.7.0-17 INFO Pulling image airbyte/connector-builder-server:1.8.2 INFO Pulling image airbyte/cron:1.8.2 INFO Pulling image airbyte/db:1.7.0-17 INFO Pulling image airbyte/server:1.8.2 INFO Pulling image airbyte/worker:1.8.2 INFO Pulling image airbyte/workload-api-server:1.8.2 INFO Pulling image airbyte/workload-launcher:1.8.2 INFO Pulling image temporalio/auto-setup:1.27.2 INFO Namespace 'airbyte-abctl' already exists INFO Persistent volume 'airbyte-local-pv' already exists INFO Persistent volume claim 'airbyte-storage-pvc' already exists INFO Persistent volume 'airbyte-volume-db' already exists INFO Persistent volume claim 'airbyte-volume-db-airbyte-db-0' already exists INFO Starting Helm Chart installation of 'airbyte/airbyte' (version: 2.0.12) INFO Starting Helm Chart installation of 'airbyte/airbyte' (version: 2.0.12) WARNING Encountered an issue deploying Airbyte: Pod: airbyte-db-0.1865c6e42b7a6c33 Reason: BackOff Message: Back-off restarting failed container airbyte-db-container in pod airbyte-db-0_airbyte-abctl(35dcf9e4-1736-4421-87f3-882e0fbb8922) Count: 6 Logs: chown: /var/lib/postgresql/data/pgdata: Operation not permitted chmod: /var/lib/postgresql/data/pgdata: Operation not permitted The files belonging to this database system will be owned by user "postgres". This user must also own the server process. The database cluster will be initialized with locale "en_US.utf8". The default database encoding has accordingly been set to "UTF8". The default text search configuration will be set to "english". Data page checksums are disabled. initdb: error: could not access directory "/var/lib/postgresql/data/pgdata": Permission denied WARNING Encountered an issue deploying Airbyte: Pod: airbyte-db-0.1865c6e42b7a6c33 Reason: BackOff Message: Back-off restarting failed container airbyte-db-container in pod airbyte-db-0_airbyte-abctl(35dcf9e4-1736-4421-87f3-882e0fbb8922) Count: 7 Logs: chown: /var/lib/postgresql/data/pgdata: Operation not permitted chmod: /var/lib/postgresql/data/pgdata: Operation not permitted The files belonging to this database system will be owned by user "postgres". This user must also own the server process. The database cluster will be initialized with locale "en_US.utf8". The default database encoding has accordingly been set to "UTF8". The default text search configuration will be set to "english". Data page checksums are disabled. initdb: error: could not access directory "/var/lib/postgresql/data/pgdata": Permission denied WARNING Encountered an issue deploying Airbyte: Pod: airbyte-db-0.1865c6e42b7a6c33 Reason: BackOff Message: Back-off restarting failed container airbyte-db-container in pod airbyte-db-0_airbyte-abctl(35dcf9e4-1736-4421-87f3-882e0fbb8922) Count: 8 Logs: chown: /var/lib/postgresql/data/pgdata: Operation not permitted chmod: /var/lib/postgresql/data/pgdata: Operation not permitted The files belonging to this database system will be owned by user "postgres". This user must also own the server process. The database cluster will be initialized with locale "en_US.utf8". The default database encoding has accordingly been set to "UTF8". The default text search configuration will be set to "english". Data page checksums are disabled. initdb: error: could not access directory "/var/lib/postgresql/data/pgdata": Permission denied

I built a custom connector using the connector builder and it has been running great for months and then about a month ago started timing out and I just noticed it. I see some tables are changing from status RUNNING to COMPLETE but they remain in the async queue list. Some of the tables are empty. The state manager comment looks wrong:

[ASYNC QUEUE INFO] Global: max: 593.92 MB, allocated: 10 MB (9.9977445602417 MB), % used: 0.016833345242386704 | Queue .................... | State Manager memory usage: Allocated: 9 MB, Used: -2365 bytes, percentage Used -0.000226

It just runs forever and never finishes. It used to take (consistently) ~5 minutes. I tried deleting the connection and recreating it but that did not work. I also tested each endpoint in the connection builder to see if it had any changes, but it everything works as it should. There is one table in the connection that has ~140k records so I disabled that and re-ran. It finished but took ages. I saw lots of comments like:

<gs://my_bucket/airbyte/my_table/2025/09/16/15/c29bb93c-80dc-47e7-9115-92f43ea8a485/0.csv.gz>
<gs://my_bucket/airbyte/my_table/2025/09/16/15/c29bb93c-80dc-47e7-9115-92f43ea8a485/1.csv.gz>
...

Where there is one file with 42KB of data and then there are like 900 files with 0 bytes. So, something weird is happening here but I'm not sure what it is.