class Sentry::Profiler
def compute_filename(abs_path, in_app)
TODO-neel-profiler try to fetch this from stackprof once we patch
copied from stacktrace.rb since I don't want to touch existing code
def compute_filename(abs_path, in_app) return nil if abs_path.nil? under_project_root = @project_root && abs_path.start_with?(@project_root) prefix = if under_project_root && in_app @project_root else longest_load_path = $LOAD_PATH.select { |path| abs_path.start_with?(path.to_s) }.max_by(&:size) if under_project_root longest_load_path || @project_root else longest_load_path end end prefix ? abs_path[prefix.to_s.chomp(File::SEPARATOR).length + 1..-1] : abs_path end
def in_app?(abs_path)
def in_app?(abs_path) abs_path.match?(@in_app_pattern) end
def initialize(configuration)
def initialize(configuration) @event_id = SecureRandom.uuid.delete('-') @started = false @sampled = nil @profiling_enabled = defined?(StackProf) && configuration.profiling_enabled? @profiles_sample_rate = configuration.profiles_sample_rate @project_root = configuration.project_root @app_dirs_pattern = configuration.app_dirs_pattern || Backtrace::APP_DIRS_PATTERN @in_app_pattern = Regexp.new("^(#{@project_root}/)?#{@app_dirs_pattern}") end
def log(message)
def log(message) Sentry.logger.debug(LOGGER_PROGNAME) { "[Profiler] #{message}" } end
def set_initial_sample_decision(transaction_sampled)
-
(void)
-
def set_initial_sample_decision(transaction_sampled) unless @profiling_enabled @sampled = false return end unless transaction_sampled @sampled = false log('Discarding profile because transaction not sampled') return end case @profiles_sample_rate when 0.0 @sampled = false log('Discarding profile because sample_rate is 0') return when 1.0 @sampled = true return else @sampled = Random.rand < @profiles_sample_rate end log('Discarding profile due to sampling decision') unless @sampled end
def split_module(name)
def split_module(name) # last module plus class/instance method i = name.rindex('::') function = i ? name[(i + 2)..-1] : name mod = i ? name[0...i] : nil [function, mod] end
def start
def start return unless @sampled @started = StackProf.start(interval: DEFAULT_INTERVAL, mode: :wall, raw: true, aggregate: false) @started ? log('Started') : log('Not started since running elsewhere') end
def stop
def stop return unless @sampled return unless @started StackProf.stop log('Stopped') end
def to_hash
def to_hash return {} unless @sampled return {} unless @started results = StackProf.results return {} unless results return {} if results.empty? return {} if results[:samples] == 0 return {} unless results[:raw] frame_map = {} frames = results[:frames].to_enum.with_index.map do |frame, idx| frame_id, frame_data = frame # need to map over stackprof frame ids to ours frame_map[frame_id] = idx file_path = frame_data[:file] in_app = in_app?(file_path) filename = compute_filename(file_path, in_app) function, mod = split_module(frame_data[:name]) frame_hash = { abs_path: file_path, function: function, filename: filename, in_app: in_app } frame_hash[:module] = mod if mod frame_hash[:lineno] = frame_data[:line] if frame_data[:line] frame_hash end idx = 0 stacks = [] num_seen = [] # extract stacks from raw # raw is a single array of [.., len_stack, *stack_frames(len_stack), num_stack_seen , ..] while (len = results[:raw][idx]) idx += 1 # our call graph is reversed stack = results[:raw].slice(idx, len).map { |id| frame_map[id] }.compact.reverse stacks << stack num_seen << results[:raw][idx + len] idx += len + 1 log('Unknown frame in stack') if stack.size != len end idx = 0 elapsed_since_start_ns = 0 samples = [] num_seen.each_with_index do |n, i| n.times do # stackprof deltas are in microseconds delta = results[:raw_timestamp_deltas][idx] elapsed_since_start_ns += (delta * MICRO_TO_NANO_SECONDS).to_i idx += 1 # Not sure why but some deltas are very small like 0/1 values, # they pollute our flamegraph so just ignore them for now. # Open issue at https://github.com/tmm1/stackprof/issues/201 next if delta < 10 samples << { stack_id: i, # TODO-neel-profiler we need to patch rb_profile_frames and write our own C extension to enable threading info. # Till then, on multi-threaded servers like puma, we will get frames from other active threads when the one # we're profiling is idle/sleeping/waiting for IO etc. # https://bugs.ruby-lang.org/issues/10602 thread_id: '0', elapsed_since_start_ns: elapsed_since_start_ns.to_s } end end log('Some samples thrown away') if samples.size != results[:samples] if samples.size <= 2 log('Not enough samples, discarding profiler') return {} end profile = { frames: frames, stacks: stacks, samples: samples } { event_id: @event_id, platform: PLATFORM, version: VERSION, profile: profile } end