Hi all, We’ve been running the open-source version...
# ask-community-for-troubleshooting
v
Hi all, We’ve been running the open-source version of Airbyte (v1.1) on our own AWS EC2 infrastructure for some time without issues. However, the instance suddenly stopped working today. We’ve since created a completely new EC2 instance, but we’re unable to deploy any version of Airbyte — every attempt fails. Has anyone experienced similar issues or can provide guidance on troubleshooting this?
k
I'm sorry to hear you're having trouble redeploying Airbyte on your new EC2 instance. Based on the knowledge sources, here are some troubleshooting steps and common issues that may help: 1. Instance Type and Resources Ensure your EC2 instance has sufficient resources. Airbyte is known to perform poorly or even freeze on small instance types like t3.micro or t2.medium. The documentation and community recommend at least a t2.medium for testing and t2.large for production. Also, make sure you have enough disk space (100GB is recommended) as Airbyte stores logs and its internal database on disk. Insufficient resources can cause instability, freezing, or data loss, as reported by other users running on smaller instances or with limited disk space [Deploy EC2: Slow runtime on t3.micro](https://discuss.airbyte.io/t/2334), [Crashing, freezing, corruption / lost data](https://discuss.airbyte.io/t/3715). 2. Docker and abctl Setup Double-check that Docker is installed and running, and that your user is added to the Docker group. Follow the official steps for installing Airbyte with
abctl
on EC2, including using the
--host
flag if you want to access Airbyte from outside the instance. Also, ensure your security group allows inbound traffic on the port Airbyte is running on (default 8000) [Using an EC2 Instance with abctl](https://docs.airbyte.com/platform/deploying-airbyte/abctl-ec2). 3. Common Deployment Errors - If you see errors like "pod airbyte-abctl-airbyte-bootloader failed" or "timed out waiting for the condition," these are often related to Docker or Kubernetes issues. Try running the failing Docker commands manually to get more detailed error messages. - If you get errors about not being able to ingress direct IP addresses, make sure you are using a recent version of
abctl
(>=0.15.0), which supports exposing IPs [Troubleshooting abctl](https://docs.airbyte.com/platform/1.7/deploying-airbyte/troubleshoot-deploy). 4. Database and Migration Issues If you are using an external database (like AWS RDS) instead of the default local Docker database, be aware that some users have reported migration failures specific to RDS setups. If possible, test with the default local database to isolate the issue [Upgrade migration failing](https://discuss.airbyte.io/t/4667). 5. Logs and Further Debugging Collect logs from the Airbyte server and containers (
docker logs airbyte-server > server.log
) to help identify the root cause. If the UI or API is unresponsive, check for resource exhaustion or errors in the logs [Crashing, freezing, corruption / lost data](https://discuss.airbyte.io/t/3715), [Airbyte not responding via UI or API](https://github.com/airbytehq/airbyte/issues/46554). If you can provide specific error messages or logs from your failed deployments, I can help you further narrow down the issue. Otherwise, reviewing the above points should help you identify the most common causes of deployment failures on EC2.
v
@kapa.ai 1. We're using t3.2xlarge which should be more that enough. 2. Yes, the docker is installed, and abctl is also installed. We're using abctl v0.30.2 4. We're trying with local DB and also with AWS RDS, both approaches unsuccessful
k
A message from kapa.ai
Thank you for the additional details. Based on the knowledge sources provided, here are some focused troubleshooting steps and observations: 1. Instance Type Your use of a t3.2xlarge instance is more than sufficient, so resource constraints are unlikely to be the cause. 2. Docker and abctl You confirmed Docker and
abctl
(v0.30.2) are installed. This matches the recommended setup for EC2 deployments. The official documentation outlines the steps for installing and running Airbyte with
abctl
on EC2, including ensuring Docker is running and your user is in the Docker group. Double-check that you followed all steps, including restarting your shell after adding your user to the Docker group, as this is sometimes overlooked and can cause permission issues [Using an EC2 Instance with abctl](https://docs.airbyte.com/platform/1.8/deploying-airbyte/abctl-ec2). 3. Database (Local and RDS) Issues - Local DB: There are known permission issues with the local Postgres volume when using
abctl
. If you see errors like
chown: /var/lib/postgresql/data/pgdata: Operation not permitted
or
initdb: error: could not access directory "/var/lib/postgresql/data/pgdata": Permission denied
, a workaround reported by users is to manually set the ownership of the Airbyte DB volume directory:
Copy code
bash      sudo chown 70:70 ~/.airbyte/abctl/data/airbyte-volume-db
(UID 70 is used by the Postgres container) [Permission issue while installing Airbyte with abctl](https://github.com/airbytehq/airbyte/issues/62093). - AWS RDS: Several users have reported issues connecting to RDS, including timeouts and migration failures. Even when connectivity is confirmed, Airbyte may fail during migrations or with transaction errors. This has been observed in both forum posts and GitHub issues, and sometimes the problem is specific to RDS (not seen with local Docker DB). There is no definitive solution in the knowledge sources, but it is a known pain point [Error deploying Airbyte with an external Postgres database (AWS RDS)](https://discuss.airbyte.io/t/2179), [Upgrade migration failing](https://discuss.airbyte.io/t/4667). 4. General Troubleshooting - Ensure your security groups and network ACLs allow traffic between your EC2 instance and RDS (if using RDS). - Check logs for specific errors (
docker logs <container_name>
), especially for the
airbyte-server
,
airbyte-db
, and
airbyte-temporal
containers. - If you have run
abctl local uninstall
and then reinstalled, you may need to manually clean up the data directories to avoid permission issues. Summary: - For local DB issues, try the
chown
workaround above. - For RDS, double-check connectivity and be aware that some migration/transaction issues are known but not fully resolved in the community. - If you can share specific error messages from your logs, it may help pinpoint the issue further. If these steps do not resolve your problem, please provide the exact error logs you are seeing during deployment for more targeted assistance. The knowledge sources do not contain a universal fix for your scenario, but the above steps address the most common and documented issues.
v
@kapa.ai Is there a dedicated page for deploying Airbyte on AWS EC2 machine?
k
A message from kapa.ai
Yes, there is a dedicated page for deploying Airbyte on an AWS EC2 machine. The official documentation provides a guide specifically for using an EC2 instance with
abctl
, which covers installing Docker, setting up
abctl
, and running Airbyte on EC2. This guide also includes important notes about configuring the security group and accessing Airbyte from outside the instance by using the
--host
flag with
abctl
 [Using an EC2 Instance with abctl](https://docs.airbyte.com/platform/next/deploying-airbyte/abctl-ec2). If you need step-by-step instructions, you can refer directly to that page for the most up-to-date and detailed process.
v
@kapa.ai We've followed everything but still cannot deploy Airbyte
k
A message from kapa.ai
I'm sorry to hear that you're still unable to deploy Airbyte on your EC2 instance despite following the official instructions. Based on the knowledge sources, here are the recommended next steps: 1. Check Common Errors and Logs: The documentation lists several common errors when deploying with
abctl
, such as issues with the Airbyte bootloader, Docker command failures, kubeadm initialization errors, and ingress/DNS problems. For each error, the docs recommend: - Running the failing Docker command manually to get more detailed error output. - Verifying that Docker can run containers in general (e.g.,
docker run hello-world
). - Checking for permission issues or port conflicts. - Reviewing logs for the Airbyte services to identify the root cause of the failure. 2. Version Compatibility: If you encounter an error about "must be a DNS name, not an IP address," ensure you are using
abctl
version >=0.15.0, as this version added support for exposing Airbyte on an IP address rather than just a DNS name. 3. Security Groups and Network: Make sure your EC2 security group allows inbound traffic on the port Airbyte is running on (default 8000, or the port you specified with
--port
). Also, ensure there are no network restrictions blocking access to required resources, especially if using an external database. 4. Further Support: If you have tried all troubleshooting steps and are still unable to deploy, the documentation suggests reaching out for help via: - The [GitHub Airbyte Forum](https://github.com/airbytehq/airbyte/discussions/categories/questions) - The [Airbyte Community Slack](https://airbytehq.slack.com/archives/C021JANJ6TY) - Submitting a [GitHub issue](https://github.com/airbytehq/airbyte/issues) with detailed logs and error messages The knowledge sources do not provide a universal fix for deployment failures beyond these steps. If you can share specific error messages or logs from your deployment attempts, the community or support channels may be able to provide more targeted assistance. For reference, see the troubleshooting and EC2 deployment guides: - [Troubleshooting abctl](https://docs.airbyte.com/platform/next/deploying-airbyte/troubleshoot-deploy) - [Using an EC2 Instance with abctl](https://docs.airbyte.com/platform/next/deploying-airbyte/abctl-ec2)
b
when you launch the ec2, dont select amazon linux, select ubuntu, that might help
v
I've tried deploying on Ubuntu but I'm still struggling. I’ve tested both v2.0 and v1.8.5, but neither was successful.
b
Create a brand new Ubuntu, then put the exact commands you run into a text file (just the commands not the results). Then paste the result of the last command, the one that fails and you get stuck on
v
At least v1.8.5 created 66 tables in RDS, seems that connection with the external DB is now an issue anymore
But after that:
Copy code
SUCCESS  Installed Helm Chart nginx/ingress-nginx:
            Name: ingress-nginx
            Namespace: ingress-nginx
            Version: 4.13.3
            AppVersion: 1.13.3
            Release: 1
  INFO    No existing Ingress found, creating one
 SUCCESS  Ingress created
  DEBUG   Received unexpected event: *v1.Status
  DEBUG   Event watcher completed.
  ERROR   Timed out waiting for ingress
  ERROR   Unable to install Airbyte locally
  ERROR   browser liveness check failed: context deadline exceeded
b
Yeah I hate Linux. In the end I just paid a dev ops guy I know to just give me the exact commands to run to do what I needed
v
I will try with CentOS or Fedora as well
b
Steps: spin up a new ec2, run these exact 10 commands in this order, and you’re done. That’s what I needed and got sick of working for days to get it.
I mean once it’s up it’s great but getting to that point with this infrastructure is a pita
v
Ubuntu alternatives
Another thing that chat suggested is that the kernel should be > 4.19
There is nothing in the official documentation about this
b
Yeah I just submitted a pr to that doc, you have to set the ec2 max hops property to 3
v
Ok, I will set that now and will test with a new machine and will let you know
b
But what I mean is this: https://gist.github.com/dossy/ecb0fa4a4e620fc166cd6e07d9d915b9 https://gist.github.com/dossy/bade960542bc7c6fccfac8e2c822d340 I tell him exactly what I need to do, and a few hours later I get one of these from him that just works.
Trick is getting as far as you can so you can define a clear greenfield ask that he can do without needing more explanation or access to your systems
v
Copy code
RuntimeError

  The Poetry configuration is invalid:
    - Additional properties are not allowed ('group' was unexpected)
b
now how about every single command you ran on the brand new ec2 before you got that error message?