Conversion to 64b

We made the change this morning and so far all seems good
We are keeping a close eye on the release of odbc connections
I was hoping the 64b version would eliminate an error that randomly occurs, but has not.
Please see the attached

Can the ETL tap and log the cause of the error?
The information on the internet is very sketchy at best on what the cause is
Lets say the etl is installed on machine 1
the sql machine is machine 2
Is the error on machine 1 or 2 ?
Is something missing in the odbc call?
Is it a timing / time out issue ?
It doesn't happen all the time,

If there is a way to trap / log and definitely find the cause, I would like to offer our environment to do it

Having spent some time reading about the issue my suggestion would be to try to disable TLS 1.2.
This can be done using this tool

www.etl-tools.com/articles/support-for-tls-1-2.html

Or rise a support ticket with Microsoft

Also, install patches on the server and the client

docs.microsoft.com/en-gb/archive/blogs/s...008-r2-2012-and-2014

The Crypto showed our AWS servers were not set up to best practices
We applied the Crypto best practices to the IIS settings, and the number of errors dropped to about 2 per day
Excellent suggestion

Since these are AWS servers, maintained by AWS, we can only assume all Microsoft patch work is up to date

I am going to work on having the process loop back and restart.
I would be neat to capture he error code and be able to restart based on specific error codes

But Crypto was an excellent starting point
thank you