#CraigRobertWhite-run-report

1 messages · Page 1 of 1 (latest)

edgy joltBOT
quaint sequoia
astral ivy
#

Ah, okay. Glad to hear it's working!

quaint sequoia
#

I may have been mistaken. While we were able to create a ReportRun after several times of it giving us an error, it's now stuck in a pending state again. I'll check in in a little while with a status update.

astral ivy
#

CraigRobertWhite-run-report

quaint sequoia
#

I think I can safely say that these ReportRuns are demonstrating the same "stuck" state that the ones yesterday encountered. Just this morning these same reports took seconds to generate.

If it helps at all, we're in the test environment using "live" data (as in not pre-filled data).

#

The explanation we received yesterday was that there were insufficient resources allocated to reporting yesterday and that your teams were trying to resolve the issue. Maybe because we were doing reports this morning before a lot of users started coming in, there were more resources available?

astral ivy
#

Ah, okay. It looks like there may have been a reporting delay yesterday that has since been resolved. These reports aren't instantaneous, so some delay is expected.

#

Are you experiencing really long wait-times? If so, what's the size of the report you're running?

quaint sequoia
#

connected_account_payout_reconciliation.itemized.3
For a month's worth of data. Again, it was previously taking seconds to reach a succeeded status. However, these ones have been going for an hour or so.

#

There must have been some kind of outage just recently because we have
1 pending
5 failed
1 pending
(which the failed ones were what my original question was about)

astral ivy
#

How much data is in each report? Are you still getting failed reports today?

#

For a month's worth of data.
This is entirely subjective and could be any amount from 0 to some astronomical upper bound

quaint sequoia
#

In this environment for this account these reports only have 90 rows. Again, these are the ones we were testing earlier this morning. The only reason we have been running them multiple times is for development purposes.

astral ivy
#

When was "earlier this morning"? We're global so that could mean anytime up to 24 hours ago, depending on where you're located

#

(sorry for all the questions, I just want to make sure I have a full grasp of what's going on)

quaint sequoia
#

10 AM Eastern US

astral ivy
#

Hmmm, okay. Can you try running once more now and let me know what the delay is?

quaint sequoia
#

Of course, stand by
(don't actually stand by)

#

It looks like whatever was going on unclogged itself. We're getting them within seconds again. It must be a capacity in the mornings with people generating daily reports or something like that. I'm not sure. It's a little discomforting that this is happening at all for such a small amount of data.

Thank you for your help.

astral ivy
#

That's fair, though I don't think that small reports would take an hour, though there is definitely some fluctuation in timing that are due to factors outside of that. Mostly I would recommend building workflows that don't rely on any specific timing, but which are aware of the timing (e.g. will notify you of reports that error, timeout, or take longer than expected)

quaint sequoia
#

Understood. As it is right now, our app polls for the updated state of the report which might be annoying to the user having to wait a while. It will be fine though, thanks for everything.