HI! I’m introducing myself into Wait Steps, i tried using both options, wait for “A period of time” and “A condition” and none of them worked correctly.
For example I tried first setting a 5min timeout and really took a lot of seconds:
@AleksiAlkio I 'm quite frustrated with this issue. I just realized there’s a bug.
The only condition is a 5-minute wait, which is 300 seconds, but the automation monitor clearly shows it’s not triggering as it should. None of the triggers are happening on time.
This feature seems to be entirely malfunctioning. I’m not even sure who to report this to. And frankly, reporting it to support seems pointless, as they often don’t grasp these issues.
The execution timeout is set for an additional 300 seconds, so technically, we’re looking at a 300-second wait plus a 300-second execution timeout, totaling 600 seconds. In the worst-case scenario, it shouldn’t exceed 900 seconds. However, as highlighted in my previous post, the actual execution time is significantly longer, which is a clear indication of a bug
I made a test with a simple Bot that has been working just fine. I added a 5 minutes delay and it took more than one hour to send the email. Definitely a bug.
I’ve noticed that bots using a wait-step are failing to return values for quite some time. They don’t write to the database, yet they show as completed in the automation monitor. This needs fixing. Also, the audit log doesn’t show any data about the wait step returns, only basic trigger details, which aren’t helpful.
I noticed this and a couple of other things that do not work when the app is not deployed. In fact, it seems that pretty much none of the automations work as they should when the app is not deployed. Have you guys tried deploying your apps? In my case this has resolved 90% of the automation issues that I have found in the past. I guess the automations are the very last part that we are supposed to develop, and that is why it is so dysfunctional until we actually deploy. Tell me if this helps at all!
They keep making the same mistake, not reading details properly. This way, real bugs aren’t reaching the developers because of these support desk inefficiency.
I reported this issue with the thread a while ago, hoping that the AppSheet support team would take care of it. It’s been quite a few days now, so I really hope they’ve gotten around to fixing it. I’m eager to test it out and see if the problem has been resolved.