I have a automation setup to create rows in a 2nd table when a new record is created in the 1st table, this triggers correctly and adds the rows into the 2nd table. I’ve setup another automation to watch for new rows in the 2nd table to send email notifications, but this never triggers.
Are automations unable to watch for data changes made by an automation or will I have to build the notifications into the 1st automation some how?
Yes, currently we are not able to trigger the workflow by the another workflow. My understanding is purely because of avoiding the possivle infinite loop? if i m correct.
Sounds like this is definitely new feature under new Automation, to trigger process (worflow) by another process (workflow)…
In this sample topics on this thread, I asume table 1 and Table 2 should not have any direct connectoin and links. But do we need to pass “process input” when we call process 1 ? to start bot?s
We need to pass value or leave them blank? As the passed value from process 1 to 2 will not be used.
This question is actuallly involving another issue and qusetion i have, so let me invite @prithpal here as well
Chaining process and process is kinda of difficult things under Automation for me (us)
Passing value from table 1 to table 2
Let me asssume a case like this
Simply the process one is to change value in column A in table 1. This process returns column A value.
Then we call this proces 1 to start child process. Column A value will be used in child process, so we pass as Process input.
Q : Im not testing this with own sample app, but process 2 will be starting by making sure process 1 is RETURNING value ? I worrry about process 2 is going to start and the previous value in colun A will be passed on to process 2. Is tehre any concept asyn behind the scene?
Process 1 will do a job to add new row to table 1. Then this row have VC to count the number of the row in this table. At the end this table returns VC column value.
To get the new/refreshed VC value could take a longer time as it will run the calc overall the tables, so my question is
Q : like case 1, the Process input would surely capture the value of VC taking into account the valeu change (add row) made by Proces 1?
The last question. When we chain multiple process in one bot, what is actualy triggering the 2nd process? The event of the bot? Or the fact the 1st process finishing the job?
When an appropriate event (say event_A: “new row added to table_1”) fires, the bot (say bot_A) will trigger the process (say process_A) passing in the new row of table_1 to process_A. All the process_A steps/tasks/actions have access to that row of data.
If you want to trigger process_B directly:
Create process_B (which has an input entity of table_2)
Call process_B from process_A (“Call process” step in process_A) and pass whatever values you want to process_B.
process_B could have a step to update table_2
Answer to your Case 1 Q:
Process 1 (parent process) will not complete execution till Process 2 (child process) has finished executing first. Child process doesn’t wait for the parent process its the other way round and Process 1 (Parent process) will not return anything till Process 2 (Child process) has finished executing.
Answer to your Case 2 Q:
Can you share a screen shot of your process (with the different steps and their configs) for case 2? Its tough to understand what you are trying to express without it.
Answer to your last Q: When you chain multiple process in one bot…as far as the bot is concerned, its only triggering ONE process, in our case process_A. The bot has no knowledge of any other process. process_A is the one that triggers process_B directly by calling it through a “Call a process step”.
Thanks you for your explanation, now i m sort of understanding it. The lessons learned we had so far with my teams is it could be better to wrap the task by process. Then on the bot, we set steps to call process for each step rather thna chain the steps with tasks only.
we see some odd stuffs when we pushed (naked) tasks (not wrapped by process) into the step, which overwrite the previous process jobs etc etc, but thank to you explnation, now we are getting more of tips .
In tems of my Q2 question, to get and pass valiable (VC value) , your tips solve the problem.
One of our lesson learned.
when you mix the process and task, something unexpected to happen. Bot is working fine, but we need to understand how Automation is gonna work. Super simple case on one BOT. 1st step will call a process to change value of the row. 2 nd step to call a process to change another value of the same row. Then 3 rd step, pure Task. Samely it update the same row in another value. But the last step (task) will make the preceeding processed null and void. The last task will reset the value of fields which was updated by 1/2nd process to be initialized.
We had difficulities why it does happens, but we reach out to the reasons.