Log-based validation of FIX applications...

, ,
1

Imported from previous forum

2

Hi All,

I’m curious how many people are using existing/captured FIX logs as a validation template for testing their systems; just loading saved logs and then replaying those logs against a target system with the target system’s responses being evaluated against the log data…

Is this something you are doing at your firm? Is it an integral part of the testing and Q/A process? If so, what are some of the challenges that you’ve encountered doing this?

Thanks,

Russ

3

Hi All,

I’m curious how many people are using existing/captured FIX logs as a
validation template for testing their systems; just loading saved logs
and then replaying those logs against a target system with the target
system’s responses being evaluated against the log data…

Is this something you are doing at your firm? Is it an integral part of
the testing and Q/A process? If so, what are some of the challenges that
you’ve encountered doing this?

Thanks,

Russ

We do. But not as standard operating procedure…
We do that only when we are on a tight schedule and can’t do actual testing with the client.

It has its own drawbacks - specially the fact that it can never cover as many real-time scenarios that can be covered while testing with the client.

4

[ original email was from Christiaan de Vries - chrisdus@gmail.com ]
Same here; It’s hard to do it as integral part of the testing process time wise, but there are some cool tools being developed at the moment which make that a lot easier… :slight_smile: I am working with a software developer who is currently developing a complete package to do every kind of test you’d want to do… :slight_smile:

Regards,
Chris

Hi All,

I’m curious how many people are using existing/captured FIX logs as a
validation template for testing their systems; just loading saved logs
and then replaying those logs against a target system with the target
system’s responses being evaluated against the log data…

Is this something you are doing at your firm? Is it an integral part
of the testing and Q/A process? If so, what are some of the challenges
that you’ve encountered doing this?

Thanks,

Russ

We do. But not as standard operating procedure… We do that only when we
are on a tight schedule and can’t do actual testing with the client.

It has its own drawbacks - specially the fact that it can never cover
as many real-time scenarios that can be covered while testing with
the client.