Insufficient cache for API call (jsonlite)?

I'm making some API calls to dowload weather data from an online provider. The site itself is private so I can not share that. I'm using jsonlite::fromJSON to get the data as a table:

> head(fromJSON(web_url)$hourly$data)
        time       summary                icon precipIntensity precipProbability temperature
1 1515711600         Clear         clear-night               0                 0       -0.42
2 1515715200 Partly Cloudy partly-cloudy-night               0                 0       -0.18
3 1515718800 Partly Cloudy partly-cloudy-night               0                 0       -0.22
4 1515722400 Partly Cloudy partly-cloudy-night               0                 0       -0.08
5 1515726000 Partly Cloudy partly-cloudy-night               0                 0       -0.28
6 1515729600 Partly Cloudy partly-cloudy-night               0                 0       -0.10
  apparentTemperature dewPoint humidity pressure windSpeed windBearing cloudCover visibility
1               -0.42    -2.73     0.84  1016.00      0.48         346       0.00      12.34
2               -0.18    -3.52     0.78  1016.32      0.34         308       0.31      12.34
3               -0.22    -3.52     0.78  1015.90      1.03         338       0.31      12.34
4               -0.08    -3.49     0.78  1016.20      0.95          48       0.31      12.62
5               -0.28    -3.43     0.79  1016.56      0.19         129       0.36      12.97
6               -0.10    -3.60     0.77  1016.40      0.81          40       0.31      12.34

When I make standalone calls, it works fine but when I try to use lapply to get the data for a longer period of time, it does not work.

lapply(c("data.table", "lubridate", "httr", "jsonlite"), 
       library, character.only = T, quietly = T)

lastyr <- as.POSIXct("2018-01-01 00:00", format = "%Y-%m-%d %H:%M", tz = "UTC")
ndays <- seq.POSIXt(from = lastyr, to = lastyr + days(366), by = "day")

history_list <- lapply(ndays, function(tstamp){
  web_url <- urlCreator(tstamp)
  return(fromJSON(web_url)$hourly$data)
})

The function urlCreator outputs a string which is the custom URL used to access the API. When I run the above code, I get an error message (generally 40 iterations in - but it is not consistent):

Error in open.connection(con, "rb") :schannel: next InitializeSecurityContext failed: SEC_E_BUFFER_TOO_SMALL (0x80090321) - Le cache fourni à une fonction était insuffisant.

Am I running out of memory? There aren't any other items saved in my memory so not sure why that is? In the past I've used httr::GET() and httr::content() to get all JSON data from the url and then parsed it to get what I need. fromJSON seemed like an easier way until I ran into this.

EDIT:

Older code (using httr only)

json <- GET(web_url)
  if(json$status_code == 200){
    c <- content(json)
    cd <- c$hourly$data 

    # cd is a list, listConvert is a function to go through the list 
    # and convert it to a data.table 

    return(rbindlist(lapply(1:length(cd), listConvert), 
                     use.names = T, fill = T)) }