From 74596b3e2b7af21cdf5a02946c07768dcc3a11e6 Mon Sep 17 00:00:00 2001 From: Jeffrey Wu Date: Thu, 23 Jan 2020 10:25:46 -0800 Subject: [PATCH] increased badWorkedFailInterval and added more failure attempts (#2936) --- src/toil/test/src/jobServiceTest.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/toil/test/src/jobServiceTest.py b/src/toil/test/src/jobServiceTest.py index 10d79397ef..7be097124a 100644 --- a/src/toil/test/src/jobServiceTest.py +++ b/src/toil/test/src/jobServiceTest.py @@ -162,7 +162,7 @@ def testServiceParallelRecursive(self, checkpoint=True): finally: list(map(os.remove, outFiles)) - def runToil(self, rootJob, retryCount=1, badWorker=0.5, badWorkedFailInterval=0.05, maxServiceJobs=sys.maxsize, deadlockWait=60): + def runToil(self, rootJob, retryCount=1, badWorker=0.5, badWorkedFailInterval=0.1, maxServiceJobs=sys.maxsize, deadlockWait=60): # Create the runner for the workflow. options = Job.Runner.getDefaultOptions(self._getTestJobStorePath()) options.logLevel = "DEBUG" @@ -182,7 +182,7 @@ def runToil(self, rootJob, retryCount=1, badWorker=0.5, badWorkedFailInterval=0. break except FailedJobsException as e: i = e.numberOfFailedJobs - if totalTrys > 40: #p(fail after this many restarts) = 0.5**32 + if totalTrys > 50: #p(fail after this many restarts) = 0.5**32 self.fail() #Exceeded a reasonable number of restarts totalTrys += 1 options.restart = True