In toph-allen/interceptr: Intercept Environment Errors Before They Happen
knitr::opts_chunk$set(
warning = FALSE,
message = FALSE,
echo = FALSE
)
options(digits = 3)
library(tidyverse)
library(lubridate)
library(pins)
library(interceptr)
Please note:
- The data represents only the subset of Connect users who interacted with support and submitted diagnostic bundles.
- These bundles are not de-duplicated to account for multiple bundles submitted from the same user.
discard_after_dash <- function(x) {
x1 <- str_split(x, pattern = "-")
map_chr(x1, 1)
}
version_major <- function(x) {
x1 <- str_split(x, pattern = "\\.")
x2 <- map_chr(x1, 1)
}
version_minor <- function(x) {
x1 <- str_split(x, pattern = "\\.|-")
x2 <- map_chr(x1, ~ paste(.x[1:2], collapse = "."))
modify_if(x2, ~ str_detect(.x, pattern = "NA"), ~ NA)
}
version_micro <- function(x) {
x1 <- str_split(x, pattern = "\\.|-")
x2 <- map_chr(x1, ~ paste(.x[1:3], collapse = "."))
modify_if(x2, ~ str_detect(.x, pattern = "NA"), ~ NA)
}
patch_digit <- function(x) {
x0 <- discard_after_dash(x)
x1 <- str_split(x0, pattern = "\\.|-")
x2 <- map_chr(x1, ~ paste(.x[4], collapse = "."))
modify_if(x2, ~ str_detect(.x, pattern = "NA"), ~ "0")
}
in_gigs <- function(mem_str) {
# Returns the memory roughly in gigabytes, multiplying or dividing by
# 1000 based on the units in the mem string.
units = str_extract(mem_str, "[A-Z]")
digits = as.numeric(str_extract(mem_str, "(\\d)+"))
if (is.na(units)) {
return(NA)
}
if (units == "T") {
digits = digits * 1000
} else if (units == "M") {
digits = digits / 1000
}
return(digits)
}
env_vars <- as.list(Sys.getenv(c("CONNECT_SERVER", "CONNECT_API_KEY")))
board_register_rsconnect(name = "rsconnect", server = env_vars$CONNECT_SERVER, api_key = env_vars$CONNECT_API_KEY)
diag_pin <- pin_get(
"toph/rsc-diagnostic-bundles-parsed",
board = "rsconnect",
cache = FALSE
)
# diag_local <- read_csv("data/summary.csv")
# Some unnecessary parse_[type]() functions in here to handle read.csv() version of
# bundle summary dataset published from this repo.
dat <- diag_pin %>%
transmute(
date = as.Date(date.received),
month = floor_date(date, unit = "month"),
quarter = floor_date(date, unit = "quarter"),
version = version.connect,
version_major = version_major(version.connect),
version_minor = version_minor(version.connect),
version_micro = version_micro(version.connect),
version_patch = discard_after_dash(version.connect),
os_version = os_release.version,
os_distro = discard_after_dash(os_release.version),
cpu_count = as.numeric(cpumodel.count),
mem_str = memory.count,
mem_gigs = map_dbl(memory.count, in_gigs),
docker = docker.used == "yes",
internet = internet.available == "yes",
config.email_provider.value = recode(parse_character(config.server.emailprovider), .missing = "missing"),
config.email_provider.configured = recode(parse_character(config.server.emailprovider),
none = "none (explicit)",
print = "configured",
sendmail = "configured",
smtp = "configured",
.missing = "missing"),
config.auth_provider = parse_character(config.authentication.provider),
across(starts_with("feature_usage"), parse_logical)
) %>%
tibble
Figures
Number of reports over time
# dat %>%
# group_by(month) %>%
# summarize(n = n()) %>%
# ggplot(aes(x = month, y = n)) +
# geom_bar(stat = "identity") +
# labs(title = "Number of diagnostic bundles per month")
dat %>%
group_by(quarter) %>%
summarize(n = n()) %>%
ggplot(aes(x = quarter, y = n)) +
geom_bar(stat = "identity") +
labs(title = "Number of diagnostic bundles per quarter")
Most of the following plots will display display properties in terms of the proportion of bundles, unless otherwise stated.
Connect version
# dat %>%
# group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, version_minor, version_micro) %>%
# summarize(version_frac = n() / month_n[1]) %>%
# ggplot(aes(x = month, fill = version_micro, y = version_frac)) +
# geom_col(color = "black", size = 0.25) +
# labs(title = "Proportion of Connect micro version by month")
dat %>%
group_by(quarter) %>%
mutate(quarter_n = n()) %>%
group_by(quarter, version_minor, version_micro) %>%
summarize(version_frac = n() / quarter_n[1]) %>%
ggplot(aes(x = quarter, fill = version_micro, y = version_frac)) +
geom_col(color = "black", size = 0.25) +
labs(title = "Proportion of Connect micro version by quarter")
# dat %>%
# group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, version_minor, version_micro) %>%
# summarize(version_frac = n() / month_n[1]) %>%
# ggplot(aes(x = month, fill = version_micro, y = version_frac)) +
# geom_col(color = "black", size = 0.25) +
# facet_wrap(~ version_micro) +
# theme(legend.position = "none") +
# labs(title = "Proportion of Connect micro version by month")
#
#
# dat %>%
# group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, version_minor, version_micro, version_patch) %>%
# summarize(version_frac = n() / month_n[1]) %>%
# ggplot(aes(x = month, fill = version_patch, y = version_frac)) +
# geom_col(color = "black", size = 0.25) +
# facet_wrap(~ version_micro) +
# labs(title = "Proportion of Connect micro version by month (colored by patch)")
#
#
# dat %>%
# group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, version_minor, version_micro, version_patch) %>%
# summarize(
# patch_frac = n() / month_n[1],
# patch_digit = patch_digit(version_patch),
# ) %>%
# ggplot(aes(x = month, fill = version_patch, y = patch_frac)) +
# geom_col() +
# facet_grid(cols = vars(version_micro), rows = vars(patch_digit)) +
# theme(legend.position = "none") +
# labs(title = "Proportion of Connect version x patch for each month")
# select(dat, date, version_micro) %>%
# drop_na() %>%
# ggplot(., aes(x = date)) +
# geom_density(aes(fill = version_micro), position = "fill") +
# labs(title = "Proportion of Connect micro version over time (density plot version)")
Host OS version and distro
dat %>%
ggplot(aes(x = os_distro, fill = os_distro)) +
geom_bar(color = "black", size = 0.25) +
labs(title = "Total number of bundles submitted from each Distro")
dat %>%
group_by(quarter) %>%
mutate(quarter_n = n()) %>%
group_by(quarter, os_distro) %>%
summarize(distro_frac = n() / quarter_n[1]) %>%
ggplot(aes(x = quarter, y = distro_frac, fill = os_distro)) +
geom_col(color = "black", size = 0.25) +
labs(title = "Proportion of bundles submitted from each distro per quarter")
# dat %>%
# group_by(quarter, os_distro) %>%
# mutate(distro_n = n()) %>%
# filter(! os_distro %in% c("12", "9", "linuxmint", NA)) %>%
# group_by(quarter, os_distro, os_version) %>%
# summarize(
# version_freq = n() / unique(distro_n[1])
# ) %>%
# ggplot(aes(x = quarter, y = version_freq, fill = os_version)) +
# geom_col(color = "black", size = 0.25) +
# facet_wrap(~ os_distro) +
# labs(title = "Version as a proportion of each distro per quarter")
Host hardware
CPU Count
cpu_breaks_1 <- 2^(1:8)
cpu_breaks_2 <- 2^(1.5:8.5)
ggplot(dat, aes(x = cpu_count)) +
stat_bin(breaks = cpu_breaks_2, color = "black") +
scale_x_log10(breaks = c(cpu_breaks_1, cpu_breaks_2), labels = round) +
labs(title = "CPU core count")
Memory
mem_breaks_v1 = 2^(1:10)
mem_breaks_v2 = 2^(1.5:10.5)
mem_breaks_v3 = c(mem_breaks_v1, mem_breaks_v2)
ggplot(dat, aes(x = mem_gigs)) +
stat_bin(breaks = mem_breaks_v2, color = "black") +
scale_x_log10(breaks = mem_breaks_v3, labels = round) +
labs(title = "System memory in GB")
Environment
knitr::kable(prop.table(table(dat$docker)), caption = "Running in a Docker container", col.names = NULL)
# ggplot(dat, aes(x = docker)) +
# stat_count() +
# labs(title = "Running in a Docker container")
knitr::kable(prop.table(table(dat$internet)), caption = "Has internet connectivity", col.names = NULL)
# ggplot(dat, aes(x = internet)) +
# stat_count() +
# labs(title = "Internet connectivity")
Feature Usage
# Version 1, just using a histogram
dat_features <- dat %>%
select(month, quarter, starts_with("feature")) %>%
rename_with(~ gsub("feature_usage.", "", .x, fixed = TRUE)) %>%
gather(key = "feature", value = "used", -month, -quarter)
# dat_features %>%
# ggplot(aes(x = month, fill = used)) +
# geom_bar(position = "fill") +
# facet_wrap(~ feature) +
# labs(title = "Monthly proportion of feature usage")
#
# dat_features %>%
# filter(!is.na(used)) %>%
# ggplot(aes(x = month, fill = used)) +
# geom_bar(position = "fill") +
# facet_wrap(~ feature) +
# labs(title = "Monthly proportion of feature usage (excluding NA)")
#
# dat_features %>%
# ggplot(aes(x = quarter, fill = used)) +
# geom_bar(position = "fill") +
# facet_wrap(~ feature) +
# labs(title = "Quarterly proportion of feature usage")
dat_features %>%
group_by(quarter, feature) %>%
summarize(
used_n = sum(used, na.rm = TRUE),
used_frac = used_n / n()
) %>%
ggplot(aes(x = quarter, y = used_frac, fill = feature)) +
geom_col(position = "dodge") +
facet_wrap(~ feature) +
lims(y = c(0, 1)) +
theme(legend.position = "none") +
labs(title = "Quarterly proportion of bundles reporting usage == TRUE")
Configuration
Email provider
Configuration values.
# dat %>%
# filter(date > today() - days(91)) %>%
# select(config.email_provider.value) %>%
# table() %>%
# knitr::kable(caption = "Configured email provider in diagnostic bundles from the last quarter")
# dat %>% group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, config.email_provider.value) %>%
# summarize(
# provider_frac = n() / month_n[1]
# ) %>%
# ggplot(aes(x = month, y = provider_frac, fill = config.email_provider.value)) +
# geom_col(color = "black", size = 0.25) +
# labs(title = "Monthly proportions of email provider config")
dat %>%
group_by(quarter) %>%
mutate(quarter_n = n()) %>%
group_by(quarter, config.email_provider.value) %>%
summarize(
provider_frac = n() / quarter_n[1]
) %>%
ggplot(aes(x = quarter, y = provider_frac, fill = config.email_provider.value)) +
geom_col(color = "black", size = 0.25) +
labs(title = "Quarterly proportions of email provider config")
# dat %>% group_by(quarter) %>%
# select(quarter, config.email_provider.value) %>%
# table %>%
# knitr::kable(caption = "Quarterly email provider numbers")
We can also recode the data to show whether email provider is configured at all, set as "none", or missing from the configuration file.
# dat %>% group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, config.email_provider.configured) %>%
# summarize(
# provider_frac = n() / month_n[1]
# ) %>%
# ggplot(aes(x = month, y = provider_frac, fill = config.email_provider.configured)) +
# geom_col(color = "black", size = 0.25) +
# labs(title = "Monthly proportions of email provider configuration status")
#
dat %>%
group_by(quarter) %>%
mutate(quarter_n = n()) %>%
group_by(quarter, config.email_provider.configured) %>%
summarize(
provider_frac = n() / quarter_n[1]
) %>%
ggplot(aes(x = quarter, y = provider_frac, fill = config.email_provider.configured)) +
geom_col(color = "black", size = 0.25) +
labs(title = "Quarterly proportions of email provider configuration status")
# dat %>% group_by(quarter) %>%
# select(quarter, config.email_provider.configured) %>%
# table %>%
# knitr::kable(caption = "Quarterly numbers of EmailProvider config")
# dat %>% group_by(quarter) %>%
# mutate(quarter_n = n()) %>%
# group_by(quarter, config.email_provider.configured) %>%
# summarize(
# provider_frac = n() / quarter_n[1],
# ) %>%
# group_by(config.email_provider.configured) %>%
# mutate(
# provider_diff = provider_frac - lag(provider_frac, default = 0)
# ) %>%
# ggplot(aes(x = quarter, y = provider_diff, fill = config.email_provider.configured)) +
# geom_col(color = "black", size = 0.25) +
# facet_wrap(~config.email_provider.configured) +
# labs(title = "Quarterly change in email provider configuration status") +
# guides(fill = guide_legend(title = "Configuration")) +
# theme(legend.position = "none")
Auth provider
# dat %>%
# filter(date > today() - days(91)) %>%
# select(config.auth_provider) %>%
# table() %>%
# knitr::kable(caption = "Configured auth provider in diagnostic bundles from the last quarter")
#
# dat %>% group_by(month) %>%
# mutate(month_n = n()) %>%
# group_by(month, config.auth_provider) %>%
# summarize(
# provider_frac = n() / month_n[1]
# ) %>%
# ggplot(aes(x = month, y = provider_frac, fill = config.auth_provider)) +
# geom_col(color = "black", size = 0.25) +
# labs(title = "Monthly proportions of auth provider config")
#
dat %>%
group_by(quarter) %>%
mutate(quarter_n = n()) %>%
group_by(quarter, config.auth_provider) %>%
summarize(
provider_frac = n() / quarter_n[1]
) %>%
ggplot(aes(x = quarter, y = provider_frac, fill = config.auth_provider)) +
geom_col(color = "black", size = 0.25) +
labs(title = "Quarterly proportions of auth provider config")
# dat %>% group_by(quarter) %>%
# select(quarter, config.auth_provider) %>%
# table %>%
# knitr::kable(caption = "Quarterly auth provider numbers")
# dat %>% group_by(quarter) %>%
# mutate(quarter_n = n()) %>%
# group_by(quarter, config.auth_provider) %>%
# summarize(
# provider_frac = n() / quarter_n[1],
# ) %>%
# group_by(config.auth_provider) %>%
# mutate(
# provider_diff = provider_frac - lag(provider_frac, default = 0)
# ) %>%
# ggplot(aes(x = quarter, y = provider_diff, fill = config.auth_provider)) +
# geom_col(color = "black", size = 0.25) +
# facet_wrap(~config.auth_provider) +
# theme(legend.position = "none") +
# labs(title = "Quarterly change in auth provider share of diagnostic bundles")
toph-allen/interceptr documentation built on Jan. 30, 2021, 12:48 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
knitr::opts_chunk$set( warning = FALSE, message = FALSE, echo = FALSE ) options(digits = 3)
library(tidyverse) library(lubridate) library(pins) library(interceptr)
Please note:
- The data represents only the subset of Connect users who interacted with support and submitted diagnostic bundles.
- These bundles are not de-duplicated to account for multiple bundles submitted from the same user.
discard_after_dash <- function(x) { x1 <- str_split(x, pattern = "-") map_chr(x1, 1) } version_major <- function(x) { x1 <- str_split(x, pattern = "\\.") x2 <- map_chr(x1, 1) } version_minor <- function(x) { x1 <- str_split(x, pattern = "\\.|-") x2 <- map_chr(x1, ~ paste(.x[1:2], collapse = ".")) modify_if(x2, ~ str_detect(.x, pattern = "NA"), ~ NA) } version_micro <- function(x) { x1 <- str_split(x, pattern = "\\.|-") x2 <- map_chr(x1, ~ paste(.x[1:3], collapse = ".")) modify_if(x2, ~ str_detect(.x, pattern = "NA"), ~ NA) } patch_digit <- function(x) { x0 <- discard_after_dash(x) x1 <- str_split(x0, pattern = "\\.|-") x2 <- map_chr(x1, ~ paste(.x[4], collapse = ".")) modify_if(x2, ~ str_detect(.x, pattern = "NA"), ~ "0") } in_gigs <- function(mem_str) { # Returns the memory roughly in gigabytes, multiplying or dividing by # 1000 based on the units in the mem string. units = str_extract(mem_str, "[A-Z]") digits = as.numeric(str_extract(mem_str, "(\\d)+")) if (is.na(units)) { return(NA) } if (units == "T") { digits = digits * 1000 } else if (units == "M") { digits = digits / 1000 } return(digits) }
env_vars <- as.list(Sys.getenv(c("CONNECT_SERVER", "CONNECT_API_KEY")))
board_register_rsconnect(name = "rsconnect", server = env_vars$CONNECT_SERVER, api_key = env_vars$CONNECT_API_KEY) diag_pin <- pin_get( "toph/rsc-diagnostic-bundles-parsed", board = "rsconnect", cache = FALSE ) # diag_local <- read_csv("data/summary.csv") # Some unnecessary parse_[type]() functions in here to handle read.csv() version of # bundle summary dataset published from this repo. dat <- diag_pin %>% transmute( date = as.Date(date.received), month = floor_date(date, unit = "month"), quarter = floor_date(date, unit = "quarter"), version = version.connect, version_major = version_major(version.connect), version_minor = version_minor(version.connect), version_micro = version_micro(version.connect), version_patch = discard_after_dash(version.connect), os_version = os_release.version, os_distro = discard_after_dash(os_release.version), cpu_count = as.numeric(cpumodel.count), mem_str = memory.count, mem_gigs = map_dbl(memory.count, in_gigs), docker = docker.used == "yes", internet = internet.available == "yes", config.email_provider.value = recode(parse_character(config.server.emailprovider), .missing = "missing"), config.email_provider.configured = recode(parse_character(config.server.emailprovider), none = "none (explicit)", print = "configured", sendmail = "configured", smtp = "configured", .missing = "missing"), config.auth_provider = parse_character(config.authentication.provider), across(starts_with("feature_usage"), parse_logical) ) %>% tibble
Figures
Number of reports over time
# dat %>% # group_by(month) %>% # summarize(n = n()) %>% # ggplot(aes(x = month, y = n)) + # geom_bar(stat = "identity") + # labs(title = "Number of diagnostic bundles per month") dat %>% group_by(quarter) %>% summarize(n = n()) %>% ggplot(aes(x = quarter, y = n)) + geom_bar(stat = "identity") + labs(title = "Number of diagnostic bundles per quarter")
Most of the following plots will display display properties in terms of the proportion of bundles, unless otherwise stated.
Connect version
# dat %>% # group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, version_minor, version_micro) %>% # summarize(version_frac = n() / month_n[1]) %>% # ggplot(aes(x = month, fill = version_micro, y = version_frac)) + # geom_col(color = "black", size = 0.25) + # labs(title = "Proportion of Connect micro version by month") dat %>% group_by(quarter) %>% mutate(quarter_n = n()) %>% group_by(quarter, version_minor, version_micro) %>% summarize(version_frac = n() / quarter_n[1]) %>% ggplot(aes(x = quarter, fill = version_micro, y = version_frac)) + geom_col(color = "black", size = 0.25) + labs(title = "Proportion of Connect micro version by quarter") # dat %>% # group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, version_minor, version_micro) %>% # summarize(version_frac = n() / month_n[1]) %>% # ggplot(aes(x = month, fill = version_micro, y = version_frac)) + # geom_col(color = "black", size = 0.25) + # facet_wrap(~ version_micro) + # theme(legend.position = "none") + # labs(title = "Proportion of Connect micro version by month") # # # dat %>% # group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, version_minor, version_micro, version_patch) %>% # summarize(version_frac = n() / month_n[1]) %>% # ggplot(aes(x = month, fill = version_patch, y = version_frac)) + # geom_col(color = "black", size = 0.25) + # facet_wrap(~ version_micro) + # labs(title = "Proportion of Connect micro version by month (colored by patch)") # # # dat %>% # group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, version_minor, version_micro, version_patch) %>% # summarize( # patch_frac = n() / month_n[1], # patch_digit = patch_digit(version_patch), # ) %>% # ggplot(aes(x = month, fill = version_patch, y = patch_frac)) + # geom_col() + # facet_grid(cols = vars(version_micro), rows = vars(patch_digit)) + # theme(legend.position = "none") + # labs(title = "Proportion of Connect version x patch for each month") # select(dat, date, version_micro) %>% # drop_na() %>% # ggplot(., aes(x = date)) + # geom_density(aes(fill = version_micro), position = "fill") + # labs(title = "Proportion of Connect micro version over time (density plot version)")
Host OS version and distro
dat %>% ggplot(aes(x = os_distro, fill = os_distro)) + geom_bar(color = "black", size = 0.25) + labs(title = "Total number of bundles submitted from each Distro")
dat %>% group_by(quarter) %>% mutate(quarter_n = n()) %>% group_by(quarter, os_distro) %>% summarize(distro_frac = n() / quarter_n[1]) %>% ggplot(aes(x = quarter, y = distro_frac, fill = os_distro)) + geom_col(color = "black", size = 0.25) + labs(title = "Proportion of bundles submitted from each distro per quarter")
# dat %>% # group_by(quarter, os_distro) %>% # mutate(distro_n = n()) %>% # filter(! os_distro %in% c("12", "9", "linuxmint", NA)) %>% # group_by(quarter, os_distro, os_version) %>% # summarize( # version_freq = n() / unique(distro_n[1]) # ) %>% # ggplot(aes(x = quarter, y = version_freq, fill = os_version)) + # geom_col(color = "black", size = 0.25) + # facet_wrap(~ os_distro) + # labs(title = "Version as a proportion of each distro per quarter")
Host hardware
CPU Count
cpu_breaks_1 <- 2^(1:8) cpu_breaks_2 <- 2^(1.5:8.5) ggplot(dat, aes(x = cpu_count)) + stat_bin(breaks = cpu_breaks_2, color = "black") + scale_x_log10(breaks = c(cpu_breaks_1, cpu_breaks_2), labels = round) + labs(title = "CPU core count")
Memory
mem_breaks_v1 = 2^(1:10) mem_breaks_v2 = 2^(1.5:10.5) mem_breaks_v3 = c(mem_breaks_v1, mem_breaks_v2) ggplot(dat, aes(x = mem_gigs)) + stat_bin(breaks = mem_breaks_v2, color = "black") + scale_x_log10(breaks = mem_breaks_v3, labels = round) + labs(title = "System memory in GB")
Environment
knitr::kable(prop.table(table(dat$docker)), caption = "Running in a Docker container", col.names = NULL) # ggplot(dat, aes(x = docker)) + # stat_count() + # labs(title = "Running in a Docker container")
knitr::kable(prop.table(table(dat$internet)), caption = "Has internet connectivity", col.names = NULL) # ggplot(dat, aes(x = internet)) + # stat_count() + # labs(title = "Internet connectivity")
Feature Usage
# Version 1, just using a histogram dat_features <- dat %>% select(month, quarter, starts_with("feature")) %>% rename_with(~ gsub("feature_usage.", "", .x, fixed = TRUE)) %>% gather(key = "feature", value = "used", -month, -quarter) # dat_features %>% # ggplot(aes(x = month, fill = used)) + # geom_bar(position = "fill") + # facet_wrap(~ feature) + # labs(title = "Monthly proportion of feature usage") # # dat_features %>% # filter(!is.na(used)) %>% # ggplot(aes(x = month, fill = used)) + # geom_bar(position = "fill") + # facet_wrap(~ feature) + # labs(title = "Monthly proportion of feature usage (excluding NA)") # # dat_features %>% # ggplot(aes(x = quarter, fill = used)) + # geom_bar(position = "fill") + # facet_wrap(~ feature) + # labs(title = "Quarterly proportion of feature usage") dat_features %>% group_by(quarter, feature) %>% summarize( used_n = sum(used, na.rm = TRUE), used_frac = used_n / n() ) %>% ggplot(aes(x = quarter, y = used_frac, fill = feature)) + geom_col(position = "dodge") + facet_wrap(~ feature) + lims(y = c(0, 1)) + theme(legend.position = "none") + labs(title = "Quarterly proportion of bundles reporting usage == TRUE")
Configuration
Email provider
Configuration values.
# dat %>% # filter(date > today() - days(91)) %>% # select(config.email_provider.value) %>% # table() %>% # knitr::kable(caption = "Configured email provider in diagnostic bundles from the last quarter") # dat %>% group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, config.email_provider.value) %>% # summarize( # provider_frac = n() / month_n[1] # ) %>% # ggplot(aes(x = month, y = provider_frac, fill = config.email_provider.value)) + # geom_col(color = "black", size = 0.25) + # labs(title = "Monthly proportions of email provider config") dat %>% group_by(quarter) %>% mutate(quarter_n = n()) %>% group_by(quarter, config.email_provider.value) %>% summarize( provider_frac = n() / quarter_n[1] ) %>% ggplot(aes(x = quarter, y = provider_frac, fill = config.email_provider.value)) + geom_col(color = "black", size = 0.25) + labs(title = "Quarterly proportions of email provider config") # dat %>% group_by(quarter) %>% # select(quarter, config.email_provider.value) %>% # table %>% # knitr::kable(caption = "Quarterly email provider numbers")
We can also recode the data to show whether email provider is configured at all, set as "none", or missing from the configuration file.
# dat %>% group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, config.email_provider.configured) %>% # summarize( # provider_frac = n() / month_n[1] # ) %>% # ggplot(aes(x = month, y = provider_frac, fill = config.email_provider.configured)) + # geom_col(color = "black", size = 0.25) + # labs(title = "Monthly proportions of email provider configuration status") # dat %>% group_by(quarter) %>% mutate(quarter_n = n()) %>% group_by(quarter, config.email_provider.configured) %>% summarize( provider_frac = n() / quarter_n[1] ) %>% ggplot(aes(x = quarter, y = provider_frac, fill = config.email_provider.configured)) + geom_col(color = "black", size = 0.25) + labs(title = "Quarterly proportions of email provider configuration status") # dat %>% group_by(quarter) %>% # select(quarter, config.email_provider.configured) %>% # table %>% # knitr::kable(caption = "Quarterly numbers of EmailProvider config")
# dat %>% group_by(quarter) %>% # mutate(quarter_n = n()) %>% # group_by(quarter, config.email_provider.configured) %>% # summarize( # provider_frac = n() / quarter_n[1], # ) %>% # group_by(config.email_provider.configured) %>% # mutate( # provider_diff = provider_frac - lag(provider_frac, default = 0) # ) %>% # ggplot(aes(x = quarter, y = provider_diff, fill = config.email_provider.configured)) + # geom_col(color = "black", size = 0.25) + # facet_wrap(~config.email_provider.configured) + # labs(title = "Quarterly change in email provider configuration status") + # guides(fill = guide_legend(title = "Configuration")) + # theme(legend.position = "none")
Auth provider
# dat %>% # filter(date > today() - days(91)) %>% # select(config.auth_provider) %>% # table() %>% # knitr::kable(caption = "Configured auth provider in diagnostic bundles from the last quarter") # # dat %>% group_by(month) %>% # mutate(month_n = n()) %>% # group_by(month, config.auth_provider) %>% # summarize( # provider_frac = n() / month_n[1] # ) %>% # ggplot(aes(x = month, y = provider_frac, fill = config.auth_provider)) + # geom_col(color = "black", size = 0.25) + # labs(title = "Monthly proportions of auth provider config") # dat %>% group_by(quarter) %>% mutate(quarter_n = n()) %>% group_by(quarter, config.auth_provider) %>% summarize( provider_frac = n() / quarter_n[1] ) %>% ggplot(aes(x = quarter, y = provider_frac, fill = config.auth_provider)) + geom_col(color = "black", size = 0.25) + labs(title = "Quarterly proportions of auth provider config") # dat %>% group_by(quarter) %>% # select(quarter, config.auth_provider) %>% # table %>% # knitr::kable(caption = "Quarterly auth provider numbers")
# dat %>% group_by(quarter) %>% # mutate(quarter_n = n()) %>% # group_by(quarter, config.auth_provider) %>% # summarize( # provider_frac = n() / quarter_n[1], # ) %>% # group_by(config.auth_provider) %>% # mutate( # provider_diff = provider_frac - lag(provider_frac, default = 0) # ) %>% # ggplot(aes(x = quarter, y = provider_diff, fill = config.auth_provider)) + # geom_col(color = "black", size = 0.25) + # facet_wrap(~config.auth_provider) + # theme(legend.position = "none") + # labs(title = "Quarterly change in auth provider share of diagnostic bundles")
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.