You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
97 lines
2.8 KiB
97 lines
2.8 KiB
require 'json'
|
|
|
|
# Parses rbtrace memory dumps
|
|
# Taken from https://samsaffron.com/archive/2015/03/31/debugging-memory-leaks-in-ruby
|
|
# Instructions:
|
|
# 0.- Prepare your local environment for production mode (since it should be more faithful):
|
|
# Add `SslRequirement.disable_ssl_check = true` to `config/environments/production.yml`.
|
|
# Edit `config/initializers/carto_db.rb` making `use_https?` return `false`.
|
|
# 1.- Start server in production mode enabling memory profiling:
|
|
# `MEMORY_REPORTING=true RAILS_ENV=production bundle exec rails s -p 3000`
|
|
# DO NOT run with MEMORY_REPORTING=true in production environments
|
|
# 2.- Use the features that you want to profile as much as possible.
|
|
# 3.- Memory dump. Find server PID and request the dump:
|
|
# `ps xah | grep ruby`, for example
|
|
# `bundle exec rbtrace -p <PID> -e 'Thread.new{CartoDB::memory_dump("/tmp/dump")}'`
|
|
# 4.- Run the Analyzer to get some insights about objects:
|
|
# `ruby lib/cartodb/analyzer.rb /tmp/dump > /tmp/dump.analysis.csv`
|
|
|
|
require 'json'
|
|
class Analyzer
|
|
def initialize(filename)
|
|
@filename = filename
|
|
end
|
|
|
|
def analyze
|
|
data = []
|
|
File.open(@filename) do |f|
|
|
f.each_line do |line|
|
|
data << (parsed=JSON.parse(line))
|
|
end
|
|
end
|
|
|
|
data.group_by { |row| row["generation"] }
|
|
.sort { |a, b| a[0].to_i <=> b[0].to_i }
|
|
.each do |k, v|
|
|
puts "generation #{k} objects #{v.count}"
|
|
end
|
|
end
|
|
end
|
|
|
|
class Analyzer2
|
|
def initialize(filename)
|
|
@filename = filename
|
|
end
|
|
|
|
def analyze
|
|
(1..10).each do |i|
|
|
bottom = 1 + (i - 1) * 10
|
|
top = i * 10
|
|
puts "-------- #{bottom}-#{top}"
|
|
data = []
|
|
File.open(@filename) do |f|
|
|
f.each_line do |line|
|
|
parsed=JSON.parse(line)
|
|
range = bottom..top
|
|
data << parsed if range.include?(parsed["generation"].to_i)
|
|
end
|
|
end
|
|
data.group_by { |row| "#{row['file']}:#{row['line']}" }
|
|
.sort { |a, b| b[1].count <=> a[1].count }
|
|
.each do |k, v|
|
|
puts "#{k} * #{v.count}"
|
|
end
|
|
end
|
|
end
|
|
end
|
|
|
|
class AnalyzerClasses
|
|
def initialize(filename)
|
|
load(filename)
|
|
end
|
|
|
|
def load(filename)
|
|
@classes = {}
|
|
File.open(filename + '.classes') do |f|
|
|
@classes = JSON.load(f)
|
|
end
|
|
|
|
@data = []
|
|
File.open(filename + '.dump') do |f|
|
|
f.each_line do |line|
|
|
@data << JSON.parse(line)
|
|
end
|
|
end
|
|
end
|
|
|
|
def analyze
|
|
@data.group_by { |row| "#{row['type']}:#{row['class']}" }.each do |_, v|
|
|
memsize = v.inject(0) { |s, x| s + x['memsize'].to_i }
|
|
class_name = v[0]['class'].nil? ? '' : @classes[(v[0]['class'].hex / 2).to_s]
|
|
puts "#{v[0]['type']},#{v[0]['class']},#{class_name},#{v.count},#{memsize}"
|
|
end
|
|
end
|
|
end
|
|
|
|
AnalyzerClasses.new(ARGV[0]).analyze
|