Closed Trabendo-daze closed 2 years ago
Have you gone in to set the environment variables for this piece of content on RStudio Connect, as shown here ("Vars (Environment Variables)")? You might want to check that there are really two variables called QUALTRICS_API_KEY
and QUALTRICS_BASE_URL
saved on that published content on RStudio Connect.
Yes, I should have clarified that those variables are set, and that I have double and triple-checked their values. I've gotten a different error when those are not filled in, so I'm fairly certain I've entered the variables correctly.
I just tested an .Rmd
on our demo RStudio Connect and it appears to be working. I made a very simple file and published it, then added the two environment variables.
One thing you might try is republishing the doc so that it only does this:
Sys.getenv("QUALTRICS_API_KEY")
Sys.getenv("QUALTRICS_BASE_URL")
knitr::knit_exit()
That should only print out your credentials (hopefully successfully) so you can see they are there, but then exit before running the rest of the file.
Thanks for looking into this. A colleague has reported similar issues, so might be something with our RSC instance? I'll look into this more after Thanksgiving and follow up on here.
Alright, so I ran the code chunk in your comment on 11/23 and I found something really weird.
When I knit the chunk locally, it acts as intended, printing my api key and base url. When I deploy the document with source code, it knits instantly, meaning that it doesn't error out once before I add the variables to that document. Furthermore, it prints credentials that aren't mine.
When I do add my own variables, it still prints the other credentials out. It seems like RSC is somehow referencing some sort of globally stored API key and base url variables? Maybe I can solve this by just renaming my variables, but this is definitely an issue with our RSC instance and probably has nothing to do with qualtRics. Gonna post something in our R slack channel.
Renaming the variable something other than QUALTRICS_API_KEY
solves the issue, but this shouldn't be a problem, applications should be independent from each other, right?
It sounds like someone has put Qualtrics credentials in some kind of global configuration file (.Rprofile
or .Renviron
, perhaps) on the machine where you have RStudio Connect installed. It is possible to do this with a supervisor script, like these folks are talking about.
Yes. I wish you could use an env var across pieces of content because I've had to change one in 10 different places before.
On Mon, Nov 29, 2021 at 2:56 PM Brendon Kaufman @.***> wrote:
Renaming the variable something other than QUALTRICS_API_KEY solves the issue, but this shouldn't be a problem, applications should be independent from each other, right?
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ropensci/qualtRics/issues/242#issuecomment-981966386, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABXAL4RMYE7YEQO564JGZUDUOPLHLANCNFSM5IRV66IQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
Yeah, I've requested that feature from our RSC reps as well. At Indeed, we have this on our internal Python-based data science platform and it's really nice!
Closing this because it's our problem, not yours. Thanks!
I'm able to access survey data locally in RStudio, but when I deploy the code to RStudio Connect, I'm hit with authentication errors. I have been able to deploy this same code in the past (as recently as a month ago) with no issues, but this problem has popped up recently.
Example code:
Example error (only on RStudio Connect):
I've tried resetting my API key, but no dice. Any ideas?