Troubleshooting migration tool (UI version) during Matillion update process

Matillion ETL
1

Anyone have any tips on troubleshooting the Matillion migrate tool? We're in the process of upgrading from 1.40 to 1.51.5 but can't seem to get the migrate tool to connect to the new instance. We enter URL/credentials for the new instance, select everything hit "migrate" and the dialog just sits on "Waiting for Response - Project Groups".

 

A few things to note:

  • The EC2 instance for the updated Matillion has exactly the same settings as the old instance:
    • VPC
    • Subnet
    • Security Group
    • IAM User
  • We're able to SSH into the old server and telnet on port 80 to the new instance with no problem so both EC2 instances can communicate with each other (at least over HTTP)
  • We've changed the Security Group inbound settings to include all TCP traffic. This didn't make a difference

 

The documentation hints at needing to open up incoming TCP/IP connections from the incoming instance but I would assume that doing this via Security Group config would cover this - https://documentation.matillion.com/docs/2966007

 

Any guidance we can get on this would be much appreciated.

Thanks!

 

- Terence

 

 

2

Hi @TMcDe​ (Terence),

That's definitely interesting that you can't make that connection give both instances are in the same VPC, have the same subnet, security group, etc. One thing to keep in mind is that you may not be working against port 80 depending on if you have HTTPS setup on either instance. Under the hood the migration tool is just making calls to Matillion's API on both servers. I would check this document and see if anything stands out as a possible issue: https://documentation.matillion.com/docs/2916124

One diagnostic approach I would suggest is to check the server logs on both the source and destination instances after trying to use the migration tool to see if you get any more detailed information about the failure.

The other diagnostic approach I would suggest is getting Postman setup on any machine that has access to both instances and try making a simple REST call to both. That will tell you if it's an issue with connectivity between the two instances or if it's simply an issue accessing the API on instance or another. It should also point you to a more precise error or issue.

I hope this helps. Report back and let us know if you need more help or have something offer on your particular scenario.