I have a not very pleasent feedback unfortunatelly. deploying airbyte is being such a struggle. I tried literally 10 different chart versions and variations of flags and each one of them throw a different error. This prevents so many people from trying and trust on the platform. I'm a user for more than 3 years and from what I recall, there was not a single instance where deploying it on a new VM works without so much time spent and stress. thats sad because its an amazing product I've been using for a long time, but its hard to justify spending so much time and effort just to install it. I appreciate different environments have different complexities, but installing airbyte through abctl on a fresh azure VM with ubuntu seems pretty vannila and a scenario I would expect tests to be done before release. hope it improves on this front. seems to be a portion of the software that in general would benefit from some attention.

Hello Everyone. Anyone using the clickhouse connector? I'm currently syncing from a mysql db but Airbyte is changing my table definitions and engines (forcing it to replacingmergetree) Anybody encountered such and can help? Thank you

Hi everyone, I have deployed Airbyte on a GKS cluster and configured connections for GA4, Google Analytics, WooCommerce, etc., loading data into BigQuery. I’m facing an issue with scheduled syncs. The cron job runs successfully at 6 AM UTC, but in the timeline it shows 0 records loaded, and the latest data is not appearing in BigQuery. I have deployed 3 separate Airbyte instances, and all of them are showing the same issue. Even when I run the sync manually, I get the same behavior. Sometimes the stream status also shows incomplete data for certain streams. Additionally, my streams are loading data at a very slow rate. Sync jobs are taking 1–2 hours to complete, which is another issue I’m facing. Could you please help me understand what might be causing these issues? I would really appreciate your support. Thank you.

Could you, @[DEPRECATED] Marcos Marx, hook up your public repositories with https://www.coderabbit.ai/ for free PR checks?

Hi Airbyte team, when I deploy latest Helm chart,

airbyte/webapp:2.0.1

is missing in Docker Hub right now, it would be great to add the missing Docker image, thanks! ☺️ https://github.com/airbytehq/airbyte/issues/75448

hi there i am getting an error: This organization has no available sources available to configure. I was able to find documentation that we shoudl be able to click on authentication module but i dont see that in teh agent engine? i dont see authentication module at all following this documentation: https://docs.airbyte.com/ai-agents/platform/authenticate/build-auth/authentication-module I dont see options to enable connectors in my organization either? According to the documentation there shoudl be a connectors tab? https://docs.airbyte.com/ai-agents/platform/enable-connector but i dont see that at all?

Is there anyone who can help me installing airbyte? I ran "abctl local install -v" and what I got is as below. DEBUG helm: Pod airbyte-abctl-bootloader running DEBUG helm: Add/Modify event for airbyte-abctl-bootloader: MODIFIED DEBUG helm: Pod airbyte-abctl-bootloader running DEBUG helm: Add/Modify event for airbyte-abctl-bootloader: MODIFIED ERROR Failed to install airbyte/airbyte Helm Chart DEBUG looking at airbyte-abctl-bootloader Failed() DEBUG found logs: Unsetting empty environment variable 'AB_INSTANCE_ DEBUG looking at airbyte-db-0 Running() DEBUG found logs: The files belonging to this database system will b ERROR Unable to install Airbyte locally DEBUG Received unexpected event: *v1.Status DEBUG Event watcher completed. ERROR unable to install airbyte chart: unable to install helm: failed pre-install: 1 error occurred: * pod airbyte-abctl-bootloader failed

@Support Team @support I know this probably isn't the place to do this, but I don't know where else to turn. We attempted to set up SSO via Entra and accidentally did it incorrectly and now we are locked out. We direly need to get back into our account for production jobs running today. I have submitted a ticket #17219. If this reaches anyone and can help me faster than through email support, I would greatly appreciate it.

Hi @Support Team @support I’m facing an issue connecting the Amazon Seller Partner connector. Is there someone who can help me with this?

Hello, sorry if this is not the proper channel. I'm trying to make Airbyte stop adding ANSI escape codes in the log level of its log messages. Example:

2026-04-16 13:29:31,353 [main]	^[[34mINFO^[[0;39m	c.z.h.HikariDataSource(<init>):79 - config-pool - Starting...

You can see it's not

INFO

, it's ``^[[34mINFO^[[0;39m` . Seems to be hardcoded, if I didn't miss anything. Thank you 🙂

Hello Airbyte Team. I was reading this article comparing Airbyte vs. dltHub on the official webpage, and I noticed that the section 'About dltHub' has the same content as the Qlink comparison.

hey @Support Team Just noticed an issue with the connector builder. My organization is set to EU but when i test a connection in the connector builder it seems to use one of the US ips (34.106.127.139) That’s a problem because only the EU ones are whitelisted with one of my clients. Makes me wonder if that also happens with active connections?

Hello, We're new to airbyte so forgive me if this is stupid 🙂 We started using the marketing connectors to sync data from the ads platform and found it quite surprising that there was no easy way to configure some extra metadata columns on the destination table and a way to add an extra little lambda function/ sql/ whatever to populate them. Working around this is actually quite painful so just wanted to put it on your radar and check with you if we're thinking about this the wrong way. Great experience other than that overall thanks a lot for all your efforts building the product!!

Hello, Not sure if right place, but today a lot of still relevant Airbyte

issues

have been arbitrarily

closed

by devin-ai-integration 😕 A few examples: • https://github.com/airbytehq/airbyte/issues/12437 • https://github.com/airbytehq/airbyte/issues/64156 • https://github.com/airbytehq/airbyte/issues/11157 • https://github.com/airbytehq/airbyte/issues/27067 • https://github.com/airbytehq/airbyte/issues/14064 • https://github.com/airbytehq/airbyte/issues/44858 Please could you reopen and prioritize them ? Notably for Helm Chart & Destination GCS connector which should have been considered a long time ago 🙏 Thank you 🙂

Hello Team, We are using clickhouse connector present in airbyte to load the data but it seems the airbyte only supports a single node, do we have any timeline to when we can expect the connector to push the data to a clickhouse cluster. please let me know if i am missing something.

Hi everyone! We're running Airbyte 2.0.0 self-hosted on AWS RDS and planning a PostgreSQL upgrade for our platform database (currently on PG 15). We'd love to understand the version boundaries before we proceed. A few questions: 1. What is the maximum PostgreSQL version officially supported for the Airbyte platform database (the internal DB, not connectors)? 2. Has anyone successfully run Airbyte 2.x against PostgreSQL 17 or 18? Any known issues? 3. We also use Temporal as part of the stack on the same instance — are there any known conflicts between Airbyte 2.0.0, Temporal, and newer Postgres versions? We noticed the docs don't explicitly state a min/max supported version (this issue (https://github.com/airbytehq/airbyte/issues/48542) seems to confirm the gap), so any real-world experience would be really helpful.

Hi! Working with the Meta Marketing Source on Airbyte Cloud. The

ad_creatives

stream returns

image_hash

and

video_id

for many creatives instead of full-resolution URLs. We currently post-process via a custom worker calling

/adimages?hashes=[...]

(200/batch) and

/?ids=v1,v2,...&fields=picture,thumbnails

(50/batch) after each sync. Two questions : 1. Is there a built-in or planned source config to resolve these natively (e.g.

resolve_creative_thumbnails: true

) so the URL lands directly in the`ad_creatives`stream? 2. Any best-practice pattern in the Airbyte ecosystem for post-sync enrichment that hits the source API for derived fields?

Hi Once i run Airbyte on my local, i am seeing the following error.

{
  "i18nKey": "errors.http.badGateway",
  "i18nParams": {
    "status": 502
  },
  "name": "HttpError",
  "requestId": "b9H3Q3Pg6aZKn4LMM313xE",
  "request": {
    "url": "/api/v1/web_backend/config",
    "method": "GET"
  },
  "status": 502,
  "response": "<!DOCTYPE html>\n<html>\n<head>\n<title>Error</title>\n<style>\nhtml { color-scheme: light dark; }\nbody { width: 35em; margin: 0 auto;\nfont-family: Tahoma, Verdana, Arial, sans-serif; }\n</style>\n</head>\n<body>\n<h1>An error occurred.</h1>\n<p>Sorry, the page you are looking for is currently unavailable.<br/>\nPlease try again later.</p>\n<p>If you are the system administrator of this resource then you should check\nthe error log for details.</p>\n<p><em>Faithfully yours, nginx.</em></p>\n</body>\n</html>\n"
}

Hi 👋 We're using the Facebook Marketing source (connector v5.2.6) with a ClickHouse destination. We're trying to retrieve the playable video URL (MP4

source

) for ad creatives so we can display hover-to-play previews, similar to how we already handle image creatives. We're running into an issue with a large number of ads that are partnership/branded-content ads. In these cases, the video belongs to a Facebook Page that isn't administered by the connected account (for example, an influencer's Page). What we've observed so far: • We can retrieve video IDs from ad creatives. • The`Videos`stream (Meta's`/act_X/advideos`) only returns videos in the ad account's*own uploaded video library*— it never includes the*Page-post videos*that the ads actually reference (we measured*0 overlap*between ad creative`video_id`s and the`videos`stream),even for an account that administers the owning Page. • The ad creatives streams provide thumbnails, but not the original video URL. • Direct Graph API requests to retrieve the video`source`fail due to permissions because the connected account doesn't manage the Page that owns the video (error #10). We found we can resolve the source via a Page access token (Graph API on the video node, using a token from

/me/accounts

) — but only when the connected user administers the owning Page. So our real question is whether the connector can fetch the video

source

via page tokens / the page node, rather than only the ad-account

/advideos

library. A few questions: • Has anyone encountered this limitation when working with Facebook Marketing ads? • Is there a stream, field, or connector configuration that exposes the playable video URL for these ads? • Or is this simply a Meta API permission restriction, meaning the video source is inaccessible unless the connected account has admin access to the owning Page? Any insights would be greatly appreciated. 🙏

Anybody saw this warning

Maximum capacity currently reached, additional jobs will be queued until capacity is available

Appeared out of the blue at 8pm yesterday

Airbyte Cloud has been slow for weeks. It takes like 20 seconds to open any page, and it's taking more than 10 minutes to be able to create a connector.

Hi all, anyone facing issue with Airbyte selfhosted salesforce connector ? While Authenticating after adding client id and secret. it shows this error. Thanks in Advance for help

A lot of 409 errors lately from Airbyte.. anybody else noticed it?

Hello, not letting the user choose the cursor field for "Full refresh | Overwrite + Deduped" is an issue, considering the cursor is used for the deduplication. It makes this option basically unusable for most connections

Hello! Just curious if there are any existing functionalities (or planned ones) that allow for adding descriptions to sources/connections? We have hundreds of connections and it would be useful if we can add extra comments or notes (e.g., who are the stakeholders, what are the downstream dependencies, etc)

Hello! 👋 we'd love your input on how to further speed up our Meta syncs. Our setup: • We sync our clients' ad data on a recurring schedule (every 6 hours). • We're on AirbyteCloud, using the Meta / Facebook Marketing source connector. • One of our clients is a large account: the initial sync took ~12h, and recurring syncs now run ~1h30–1h45. What we've already done to lighten the sync: • Field selection: we only sync the fields we actually use on the

insights

and

ad_creatives

streams, instead of the full schema. • Capped the insights lookback window to 14 days (

insights_lookback_window

). • Rolling

start_date

(we pull the last 12 months of history, not all-time). Our question: given the above, what else can we do on the Airbyte side to make recurring syncs faster and more reliable on large accounts? A few specific things we're wondering about: • Any recommended settings for the Meta connector to reduce sync duration on high-volume accounts ? • Any best practices to avoid Meta API timeouts / rate-limit stalls during long syncs? • Would incremental sync tuning or a different sync mode help here? • Cloud vs self-hosted: would running self-hosted Airbyte give us meaningfully better performance or more control for large accounts like this? Thanks a lot 🙏

Have thoughts about Airbyte Data Replication tools? We currently have a survey out and want to hear from you! 100 respondents will receive a $25 gift card. Please use your company email to ensure you're entered for a chance to win. docs.google.com/forms/…/viewform

Hi @airbyte team This morning I have just realised that the ability to do the Manual Shopify Authentication option is no longer available on Airbyte Cloud. Given his is critical to us and if it is not supported we have to consider churning as its our no1 use case on the platform. Thanks and I have also raised a Support ticket which is support.airbyte.com/hc/en-us/requests/19166

Hi everyone, I’ve submitted a feature request to Airbyte for configuring CPU and memory resources per connection directly from the UI. Airbyte already supports connection-specific resource requirements internally, but today this typically requires changing the internal database manually. In environments with many connections, this makes resource management harder than it should be. The main use case is that not every connection has the same workload. For example, a lightweight CDC connection may only need a small amount of CPU and memory, while another connection doing a large initial sync or heavy batch load may need much more. Being able to configure these values per connection from the UI would help avoid both over-provisioning and under-provisioning. If this would be useful in your environment as well, please consider upvoting the feature request. More upvotes can help show the Airbyte team that there is broader demand for this capability. github.com/airbytehq/airbyte/discussions/85061

Hello Everyone, We are using Airbyte OSS 2.0.1 with ClickHouse destination connector 2.1.24. After upgrading ClickHouse from 26.6.1.1193 to 26.8 LTS, syncs fail before inserting records with:

Failed to get table schema

Non-null columnName and columnType are required

TableSchemaParser.readTSKV

The connector bundles ClickHouse Java client-v2 0.9.4. ClickHouse 26.8 changed the precedence of the

X-ClickHouse-Format

header, causing the client’s

DESCRIBE TABLE ... FORMAT TSKV

response to be returned as RowBinary and incorrectly parsed as TSKV. The issue is fixed upstream in clickhouse-java #3069, but Airbyte connector versions up to 2.1.30 do not appear to include the updated client. ClickHouse 26.6 works correctly. can you please share your thoughts.