I have an email programmed to be sent when a scheduled refresh fails, however since I access quicksight via a 3rd party, I do not have a direct login to quicksight, so when I click on View Dataset, it prompts me for a login I do not have.
The 3rd party advised I should post on here for support or a feature request to make these emails more descriptive about which dataset is affected.
Hello @uwindsordata, are you accessing Quick through a custom application that embeds the console experience? Also, do you normally have access to datasets, but it breaks when you try to click this link?
I am curious about your account permissions and what features you are able to engage with outside of this specific issue. I am also curious how often you are getting these failed to refresh alerts. Maybe there is something that can be resolved on the dataset side to reduce these failures altogether.
Hi Dylan:
Yes, its a third party, which I have a link that loads the console experience. I have full access to the datasets when I access the console through the third party, but when I click on the link it prompts for a login, which I assume is not the login I use for the third party
I have started to get the emails, and my goal is to identify which dataset is failing and to fix it, however because of my poor naming strategy, I have multiple sets which use “New Custom SQL” and its timeconsuming to go through each dataset and check the refresh status.
Hey @uwindsordata, I was going to suggest going to the dataset directly to see what error it shows instead of using the link, but I see how that could be an issue with your naming conventions.
If you have the ability to edit the datasets, I would definitely recommend going through them and updating the names so you can more easily figure out where the data is failing. Since you are accessing via a third-party, I am assuming you don’t have any AWS console access. You could potentially speed up the renaming process by doing it programmatically, but that can’t be done without a direct connection.
I think you may just be running into a weird bug. Not sure if it is coming from Quick directly, or if it has to do with the way the application is being embedded and authenticated. Is there a team that manages the embedded platform that you are able to communicate with? I am curious if they have looked into this issue at all.
Hi Dylan:
Their support team referred me here and said it was up to AWS to customize those emails. 
. If there is an email I can contact, I would be happy to take this off-forum.
My guess is the embedding/authentication workflow is preventing the “view dataset” button from working properly.
Thanks
Hi @uwindsordata,
I noticed it’s been awhile since last communication took place on this thread; what if you copy the link and paste it in a window/tab that is already logged in to that environment, will the link work then without additional login required?
I agree that renaming your datasets, while it may be somewhat time consuming, may be your best immediate action to better note which dataset is failing.
In terms of additional information included to better identify the dataset, what type of information would you recommend that would be helpful in identifying the dataset difference? I can tag this as a feature request with the idea of capturing additional identifying information.
Hi Brett:
If I copy the link from the email and paste it in a tab thats already logged in to quick, it still shows the login prompt ( my guess is that my quick instance shows up as an iframe).
The emails just show “New Custom SQL”, but don’t actually show what dataset this custom SQL is located in, that would be the first piece of information desired.
Thanks
Hi @uwindsordata,
Thanks for the additional information! I know there are times where you apply customSQL and it will adjust the dataset name to match…but what you’re saying is that you’re getting an failed message in relation to the sql, not the dataset. I’m surprised that it doesn’t include but definitely agree that would extremely helpful!
I will mark this as a feature request to promote visibility to the support team!
