Sidekiq queue latency not correct (huge negative value)

[geoffharcourt]

Issue Description

We just turned on the Queues feature in our Sentry dashboard. Despite our queue latencies being mostly zero and or at most a few seconds, the latencies in the dashboard are showing as values of negative thousands of years.

Reproduction Steps

Compare the values in Sidekiq dashboard to Sentry's Backend -> Queues tab.

Expected Behavior

I'd expect our queue latencies to look like our admin dashboard in Sidekiq.

Actual Behavior

Sentry data:

Sidekiq admin dashboard:

Ruby Version

3.4.3

SDK Version

sentry-ruby 5.23.0 and sentry-sidekiq 5.23.0

Integration and Its Version

Sidekiq

Sentry Config

Sentry.init do |config|
  config.release = ENV.fetch("IMAGE_TAG", "development")

  config.propagate_traces = false

  if ENV["SENTRY_TRACE_RATE"].present?
    config.traces_sample_rate = ENV["SENTRY_TRACE_RATE"].to_f
  end

  # get query params
  config.send_default_pii = true

  if ENV["SENTRY_INCLUDE_ALL_EXCEPTIONS"]
    config.excluded_exceptions = []
  end

  config.environment = ENV["SENTRY_ENVIRONMENT"] ||
    ENV["COMMONLIT_NAMESPACE"] ||
    ENV["RAILS_ENV"]

  # NOTE: Only print Sentry logger info if we have a transport issue. This
  # avoids printing (and sending to APM) millions of lines of:
  # I, [2022-10-24T01:26:33.227290 #45]  INFO -- sentry: [Transport] Sending envelope with items [sessions]  to Sentry
  #
  # Note that the logger level has to be set after init, hence the tap
  config.logger = Sentry::Logger.new($stdout).tap do |sentry_logger|
    sentry_logger.level = ::Logger::WARN
  end
end

[solnic]

Thanks for reporting the issue! Which version of Sidekiq do you use?

[geoffharcourt]

@solnic 8.0.1. We use Sidekiq Enterprise if that might be relevant

[solnic]

@geoffharcourt ah thanks, I believe this will address it then #2570