Example build: https://uberchromegw.corp.google.com/i/chromeos/builders/daisy-paladin/builds/23028/steps/Report/logs/stdio

"This build versus last week of daisy-paladin
Averages for 74 Builds: 887956 - 900884
 1:36:25 success  95% timeouts   1% median 1:16:18 mean 1:15:40 min 0:39:59 max 1:56:57

CleanUp:
  start:    0:13:45 median 0:02:03 mean 0:04:28 min 0:00:00 max 0:44:51
  duration: 0:00:00 median 0:00:16 mean 0:00:24 min 0:00:00 max 0:01:56
  finish:   0:13:45 median 0:02:03 mean 0:04:53 min 0:00:17 max 0:44:51
CommitQueueSync:
  start:    0:13:45 median 0:02:04 mean 0:04:53 min 0:00:18 max 0:44:51
  duration: 0:00:01 median 0:01:34 mean 0:03:58 min 0:00:00 max 0:44:14
  finish:   0:13:46 median 0:04:34 mean 0:08:52 min 0:01:53 max 0:44:51
BuildStart:
  start:    0:13:44 median 0:04:43 mean 0:08:56 min 0:02:03 max 0:44:50
  duration: 0:00:01 median 0:00:00 mean 0:00:00 min 0:00:00 max 0:00:01
  finish:   0:13:45 median 0:04:43 mean 0:08:56 min 0:02:03 max 0:44:50
BuildReexecutionFinished:
  start:    0:13:46 median 0:04:44 mean 0:08:57 min 0:02:04 max 0:44:52
  duration: 0:00:05 median 0:00:05 mean 0:00:04 min 0:00:04 max 0:00:10
  finish:   0:13:51 median 0:04:49 mean 0:09:02 min 0:02:14 max 0:44:56
ConfigDump:
  start:    0:13:51 median 0:04:49 mean 0:09:03 min 0:02:15 max 0:44:57
  duration: 0:00:00 median 0:00:00 mean 0:00:00 min 0:00:00 max 0:00:01
  finish:   0:13:51 median 0:04:49 mean 0:09:03 min 0:02:15 max 0:44:57
"
..
..
..

Claims that CommitQueueSync stage for this run took ~1 second. In actuality it took ~11 minutes.

My guess without further analysis is that there are multiple CommitQueueSyncStage entries in cidb for this build, generated by different re-executions, and that this is looking at the wrong one (i.e. one that was instantly skipped).