Skip to main content

I have some system flows with more steps. If there is an issue with the system flow, it’s hardly undoable to debug. There’s no logging at all. I now created some logging in the application self, but with steps with no code attached it’s not possible to track the system flow in the debugger somewhere. 

Hi Edwin,

Which steps in particular require the logging?

At the moment we are implementing an Environment monitor in which all integration is monitored; such as File storage writes and read, Web and HTTP connectors, Application and DB connectors, and many more. Anything that connects to something outside the platform will be logged here. Anything that takes place inside the platform, i.e. Decisions / Process procedure logic, that is not logged in this.

Does that suffice?


Maybe it’s an idea to have a log depth or log intensity parameter which you can put on ‘normal’ or ‘debug’. In debug mode it should show all login of the system flows including the variables and the starts of the specific process steps, so that you’re able to follow what happens. It’s not possible to follow a system flow with the proces flow monitor and that’s the requirement. If the parameter is on normal, it can show the start and stop of the proces flow.

I’m now having an issue that the system flow starts (according IAM) but nothing happens and I can’t see anything except for my own logging, but that doesn’t work either now as well.


NewWorking on it!

We are also implementing the possibility for you to invoke system flows manually. This will allow you to test system flows in development by manually invoking it. That also will result in the system flow being mentioned in the Process flow monitor, in which you can monitor its progress and see all steps and their results.

So this idea we are actually working on 😄


Hi Edwin,

Which steps in particular require the logging?

At the moment we are implementing an Environment monitor in which all integration is monitored; such as File storage writes and read, Web and HTTP connectors, Application and DB connectors, and many more. Anything that connects to something outside the platform will be logged here. Anything that takes place inside the platform, i.e. Decisions / Process procedure logic, that is not logged in this.

Does that suffice?

Are you, by any chance, also registering the whole chain in which requests follow each other - as in a HTTP request results in a Indicium log entry(error) which in turn results into which SQL statement was provided to the SQL Server with its parameters?


Hey @Ricky,

Are you, by any chance, also registering the whole chain in which requests follow each other - as in a HTTP request results in a Indicium log entry(error) which in turn results into which SQL statement was provided to the SQL Server with its parameters?

We currently intent to only log the date time in UTC, the process action, the provider (if any), the status code, the http response code, and we added an Extra info field for Indicium to fill in with any relevant data.

Logging responses is quite the security risk as it may contain private information, so we opted not to include that in the current implementation. 


Hello,

We are also interested in logging of system flows. I think it is great to look at the integrations part, but not having logging on the internal functions is imho even more important. We made our system flows into subflows so they can be scheduled, but also be called from the development environment placing an interactive task with a small process flow above it, which calls the system flow (works nice btw). So now we have a system flow that works correctly and can be monitored in the development environment. After deploying it to the test environment however, where it actually uses the scheduled system flow configuration, it becomes it black box and nothing can be seen and, unfortunately, here the systemflow runs into some error which is only seen by not getting the expected result.

Is there by any chance some way to have a look in the flow handling “internals” of indicium to see what is going on? (The IAM scheduled flow log and the Indicium logging don’t show anything concerning this task).

 


Working on it!Completed