Use monotonic time in Redis instrumentation

Redis calls are very fast so clock skew is unlikely to be an issue, but
we should still use the better method of measuring time.

lib/gitlab/instrumentation/redis_interceptor.rb

@@ -6,7 +6,7 @@ module Gitlab
       APDEX_EXCLUDE = %w[brpop blpop brpoplpush bzpopmin bzpopmax xread xreadgroup].freeze
 
       def call(*args, &block)
-        start = Time.now # must come first so that 'start' is always defined
+        start = Gitlab::Metrics::System.monotonic_time # must come first so that 'start' is always defined
         instrumentation_class.instance_count_request
         instrumentation_class.redis_cluster_validate!(args.first)
 
@@ -15,7 +15,7 @@ module Gitlab
         instrumentation_class.instance_count_exception(ex)
         raise ex
       ensure
-        duration = Time.now - start
+        duration = Gitlab::Metrics::System.monotonic_time - start
 
         unless APDEX_EXCLUDE.include?(command_from_args(args))
           instrumentation_class.instance_observe_duration(duration)

lib/gitlab/sidekiq_queue.rb

@@ -14,7 +14,7 @@ module Gitlab
     end
 
     def drop_jobs!(search_metadata, timeout:)
-      start_time = Gitlab::Metrics::System.monotonic_time
+      start_time = monotonic_time
       completed = true
       deleted_jobs = 0
 
@@ -62,7 +62,11 @@ module Gitlab
     end
 
     def timeout_exceeded?(start_time, timeout)
-      (Gitlab::Metrics::System.monotonic_time - start_time) > timeout
+      (monotonic_time - start_time) > timeout
+    end
+
+    def monotonic_time
+      Gitlab::Metrics::System.monotonic_time
     end
   end
 end

spec/lib/gitlab/sidekiq_queue_spec.rb

@@ -31,7 +31,7 @@ RSpec.describe Gitlab::SidekiqQueue, :clean_gitlab_redis_queues do
 
       context 'when the queue is not processed in time' do
         before do
-          allow(Gitlab::Metrics::System).to receive(:monotonic_time).and_return(1, 2, 12)
+          allow(sidekiq_queue).to receive(:monotonic_time).and_return(1, 2, 12)
         end
 
         it 'returns a non-completion flag, the number of jobs deleted, and the remaining queue size' do