Limit Sidekiq args logging to 10 KB of JSON

When logging arguments from Sidekiq to JSON, restrict the size of `args`
to 10 KB (when converted to JSON). This is to avoid blowing up with
excessively large job payloads.

changelogs/unreleased/56547-limit-sidekiq-logging-based-on-argument-size.yml

+---
+title: Prevent Sidekiq arguments over 10 KB in size from being logged to JSON
+merge_request: 24493
+author:
+type: changed

doc/development/sidekiq_debugging.md

@@ -14,3 +14,8 @@ gitlab_rails['env'] = {"SIDEKIQ_LOG_ARGUMENTS" => "1"}
 Please note: It is not recommend to enable this setting in production because some
 Sidekiq jobs (such as sending a password reset email) take secret arguments (for
 example the password reset token).
+
+When using [Sidekiq JSON logging](../administration/logs.md#sidekiqlog),
+arguments logs are limited to a maximum size of 10 kilobytes of text;
+any arguments after this limit will be discarded and replaced with a
+single argument containing the string `"..."`.

lib/gitlab/sidekiq_logging/structured_logger.rb

@@ -5,6 +5,7 @@ module Gitlab
     class StructuredLogger
       START_TIMESTAMP_FIELDS = %w[created_at enqueued_at].freeze
       DONE_TIMESTAMP_FIELDS = %w[started_at retried_at failed_at completed_at].freeze
+      MAXIMUM_JOB_ARGUMENTS_LENGTH = 10.kilobytes
 
       def call(job, queue)
         started_at = current_time
@@ -64,6 +65,7 @@ module Gitlab
         job['pid'] = ::Process.pid
 
         job.delete('args') unless ENV['SIDEKIQ_LOG_ARGUMENTS']
+        job['args'] = limited_job_args(job['args']) if job['args']
 
         convert_to_iso8601(job, START_TIMESTAMP_FIELDS)
 
@@ -93,6 +95,21 @@ module Gitlab
 
         Time.at(timestamp).utc.iso8601(3)
       end
+
+      def limited_job_args(args)
+        return unless args.is_a?(Array)
+
+        total_length = 0
+        limited_args = args.take_while do |arg|
+          total_length += arg.to_json.length
+
+          total_length <= MAXIMUM_JOB_ARGUMENTS_LENGTH
+        end
+
+        limited_args.push('...') if total_length > MAXIMUM_JOB_ARGUMENTS_LENGTH
+
+        limited_args
+      end
     end
   end
 end

spec/lib/gitlab/sidekiq_logging/structured_logger_spec.rb

@@ -82,15 +82,36 @@ describe Gitlab::SidekiqLogging::StructuredLogger do
           end.to raise_error(ArgumentError)
         end
       end
+
+      context 'when the job args are bigger than the maximum allowed' do
+        it 'keeps args from the front until they exceed the limit' do
+          Timecop.freeze(timestamp) do
+            job['args'] = [
+              1,
+              2,
+              'a' * (described_class::MAXIMUM_JOB_ARGUMENTS_LENGTH / 2),
+              'b' * (described_class::MAXIMUM_JOB_ARGUMENTS_LENGTH / 2),
+              3
+            ]
+
+            expected_args = job['args'].take(3) + ['...']
+
+            expect(logger).to receive(:info).with(start_payload.merge('args' => expected_args)).ordered
+            expect(logger).to receive(:info).with(end_payload.merge('args' => expected_args)).ordered
+            expect(subject).to receive(:log_job_start).and_call_original
+            expect(subject).to receive(:log_job_done).and_call_original
+
+            subject.call(job, 'test_queue') { }
+          end
+        end
+      end
     end
 
     context 'with SIDEKIQ_LOG_ARGUMENTS disabled' do
-      it 'logs start and end of job' do
+      it 'logs start and end of job without args' do
         Timecop.freeze(timestamp) do
-          start_payload.delete('args')
-
-          expect(logger).to receive(:info).with(start_payload).ordered
-          expect(logger).to receive(:info).with(end_payload).ordered
+          expect(logger).to receive(:info).with(start_payload.except('args')).ordered
+          expect(logger).to receive(:info).with(end_payload.except('args')).ordered
           expect(subject).to receive(:log_job_start).and_call_original
           expect(subject).to receive(:log_job_done).and_call_original