How to detect failed to load Atlassian Connect app

We have had a couple of uninstall feedbacks from users of our app. Reviewing the logging didn’t reveal any problems.

Is there a way to be informed whenever Jira Cloud fails to load the app. Is there a rest endpoint or another monitoring event that can be watched for?


Hey @francis, can you elaborate a little on the feedback you received? It unclear what you’re specifically asking for here.

So what happens is

  • A user has one of our apps installed
  • The user wants to access the app
  • The app doesn’t respond due to some reason
  • This leaves a bad impression and uninstalls with the feedback ‘Bugs in the addon’

Reviewing our logging, we cannot see when this happened or for what reason.

To further RCA the problem, we would need to have more details.

The question is how this can be done. How can we detect that an App fails to load.

Hi @francis
Unfortunately iframe security applies both ways, so we have no visibility into what is happening inside an app’s iframe from our end. Many vendors implement some kind of frontend monitoring (sentry, new relic etc.) in order to detect frontend errors

If you suspect the cause is outside of the iframe, then please DM me, and I’m happy to work with you to investigate further

Lets first have sentry (or similar in place)
It would be a good thing if there is a way that a vendor can also know when Jira has trouble to access the app.

Not sure if such feature request has been raised already and if not, in what project it should be raised.

1 Like

Hi @francis ,
This is an interesting idea. I searched around and I couldn’t find any previous post in community or a ticket for that.
After internal discussions, we’d like to give developers a chance to discuss requests and use cases on this thread in order to capture everything.
The problem seems very clear but if you have any more comments or ideas, please share them here. I’m also curious to hear if other developers have any more suggestions or comments.

1 Like

Probably not what you were thinking of, but this thread triggered me into the idea that it would be absolutely awesome if we could somehow share a sentry DSN or something which would allow you to push AP / Connect bridge errors to our Sentry instance.

We have Sentry installed in our app so we catch things like AP not available (from what we can tell this happens due to user opening up the iframe in another tab).

The one thing we are getting outside of that are the wonderful firewalls or ad blockers which sometimes blocks us (highly recommend using .com, .net, .org - ie not “fancy” domains to serve connect apps - it reduces the blocks). These don’t hit our sentry (since the iframe doesn’t load) however we’re able to spot things when the user reports it to us from the screenshots (broken images). We’ve put out a page at Access check which we direct our end users to when they are having issues accessing the app which really makes firewall access checking much faster.

It would be “neat” if Atlassian had a “troubleshooting” link in the “we can’t load the app” inline message which pops up a dialog which tries to load a “health” module or just the app logo (the module would be better since it would trigger the add blockers) with messaging directing the user to either contact the app’s support team OR check their local settings.



+1 for this suggestion.