Skip to main content

There are a variety of ways to connect to your Benchling Data Warehouse. But sometimes, connecting to your warehouse doesn’t go smoothly. If you experience issues connecting to your warehouse, this guide is for you. The goal of this guide is to:

  1. Provide some general guidelines on how to diagnose connection issues

  2. Provide a list of commonly encountered issues and suggested solutions

  3. Provide a space for community members to help each other solve warehouse connectivity issues! If you’re running into an issue yourself, chances are, other members of the community may have run into the same one.

Note on Third Party Tools

Please note that while we provide the Data Warehouse for use however you would like, Benchling does not provide guarantees on third party tools you may use to connect. Some tools include Power BI, Tableau, and ETLeap. If you are running into issues with these tools, we encourage you to first check out the documentation specific to that tool and then ask for help here from the Community.

Here are some links to documentation for some third party tools:

Tips on Diagnosing Issues

First, take a look at the list of frequently encountered issues below. If none of the resolutions provided work, please ask your firewall/networking IT team to assist you with connecting. We very frequently find that connection issues are due to firewalls blocking the connection to the warehouse. Please ask your firewall IT team to monitor the system while you try to connect; more likely than not, they can provide logs that describe why the connection failed. These logs usually point to the source of the problem.

If the logs do not allow you to fully solve the problem, then please post here so that other Community members may be able to help out.

Frequently Encountered Issues

Here’s a list of issues customers have encountered in the past along with potential solutions. If your issue is not on this list, please post a comment below to ask for help from the Community!

  • SSL SYSCALL errors. If you receive an SSL SYSCALL error, make sure you are using the correct credentials and password, and pay close attention to the following error text.

    • If the following text mentions “EOF detected,” it is likely that your warehouse has experienced a tiny blip in availability, causing a temporary reboot. Since this issue is transient, please try connecting again to the warehouse after a few moments. Example:

      port 5432 failed: SSL SYSCALL error: EOF detected connection to server at "postgres-warehouse.<subdomain>.benchling.com" (<IP Address>), port 5432 failed: FATAL
    • If the following text mentions “Connection Reset By Peer,” it is exceedingly likely that a firewall on your side is preventing the warehouse connection. The “Peer” here is referring to your firewall. Please ask your networking/firewall IT team to help diagnose this issue. Usually, when you attempt to connect to the warehouse, the firewall will produce logs that will help you all understood the root problem. Example:

      (<IP Address>), port 5432 failed: server closed the connection unexpectedly This probably means the server terminated abnormally before or while processing the request. SSL SYSCALL error: Connection reset by peer
    • If the following text mentions “FATAL: no pg_hba.conf entry for host..” it is very likely that you are not connecting with SSL/TLS. Please see our documentation on how to connect with SSL/TLS: Overview & Getting started. It is also possible that you may be attempting to connect with an SSL/TLS connection, but your firewall or VPN is not allowing the SSL/TLS to go through. As such, please work with your networking/firewall IT team if you believe you’re connecting with SSL/TLS. We have even seen cases where an ODBC tool is claiming to connect with SSL/TLS but in reality is not. Ultimately, the “peer” here may be referring to the warehouse or your firewall.

      SSL SYSCALL error: Connection reset by peer connection to server at "postgres-warehouse.<subdomain>.benchling.com" (<IP Adress>), port 5432 failed: FATAL: no pg_hba.conf entry for host "<IP Address>", user "u$example", database "warehouse", no encryption
  • An error of the form “no pg_hba.conf entry for host..”

    • Please see above the section referring to this under the “SSL SYSCALL” section above--it’s possible that you receive this error without an accompanying “SSL SYSCALL” text. However, the resolution is very likely the same: please check that you are connecting with SSL/TLS.

  • When you try to connect, the connection attempt seems to take a while and then ultimately fails.

    • Your Benchling tenant and warehouse may have IP allowlisting involved; see our documentation there: Self-Serve IP Allowlist Configuration. Please double check with your Benchling admins if this is the case. If so, make sure you’re attempting to connect to the warehouse via an allowed IP address.

  • You know you use a firewall to manage connections to the data warehouse. However, you use the IP address of the warehouse in the firewall itself, and this IP address seems to have changed, causing the firewall to disallow connections to the warehouse.

    • Because of our data warehouse architecture (i.e. that fact that it is built on AWS), the IP address of the warehouse itself may change at any time due to factors outside of our control. Please use the URL of the warehouse (e.g. “postges-warehouse.<subdomain>.benchling.com”) in your firewall instead.

Ask the Community!
Have any questions or issues not addressed above? Or have you successfully solved a connection issue for your own team? Please comment below so we can help each other out!

Be the first to reply!

Reply