1

I am working on placing a rpivotTable inside of a Shiny app. When I try on test data (a data frame with 1000 rows) I am able to run my app from the command line, and others may access the app given my ip:port as enjoyed. However, when I up the size of the data frame being fed into rpivotTable, the app will 'grey' out and I'm not able to serve the app to others.

I have also, successfully tested this same app, spinning up an EC2 instance, and upping the instance type, but the same thing would happen. I was getting a similar error as shown in this post ERROR: [on_request_read] connection reset by peer in R shiny and on this github issue https://github.com/rstudio/shiny/issues/1469. "ERROR: [on_request_read] connection reset by peer"

My syntax is pretty straightforward in terms of calling and rendering the rpivotTable, but as the size of the data frame increases, my app doesn't work. My suspicion is that this is a timeout parameter in the javascript widget?

egolinko
  • 21
  • 1
  • 2
  • Are you able to post some code to give us a better idea of what may be going on? – Ryan Morton Mar 06 '18 at 23:18
  • Sure, the code is pretty straight forward for the pivot table and rendering, see below, it's just that the file that is being fed in is about 200,000 rows, and only 6 columns, it works fine for test data, so I'm not convinced it's a shiny thing – egolinko Mar 07 '18 at 13:30
  • `tabPanel("name of my tab", rpivotTable::rpivotTableOutput("cash") )` – egolinko Mar 07 '18 at 13:30
  • `output$cash <- renderRpivotTable({ rpivotTable::rpivotTable(ex)} )` – egolinko Mar 07 '18 at 13:31
  • Where `ex` is the data frame, apologies for the multiple comments, noob at posting here, though use stack overflow everyday – egolinko Mar 07 '18 at 13:38

1 Answers1

0

I had the same problem, and had to upgrade from t3a.medium to t3a.large. That's more than I wanted, but it works now.

Steve Powell
  • 1,646
  • 16
  • 26