> On May 1, 2014, 7:44 p.m., Mark Chu-Carroll wrote: > > src/main/python/apache/aurora/client/api/job_monitor.py, line 81 > > <https://reviews.apache.org/r/20950/diff/1/?file=572657#file572657line81> > > > > shouldn't this be >=?
Line 84 guarantees it never exceeds the max value: poll_interval = min(self.MAX_POLL_INTERVAL, 2 * poll_interval) > On May 1, 2014, 7:44 p.m., Mark Chu-Carroll wrote: > > src/test/python/apache/aurora/client/api/test_job_monitor.py, line 103 > > <https://reviews.apache.org/r/20950/diff/1/?file=572663#file572663line103> > > > > You can just "assert monitor.wait_until...". > > Sure, changed. - Maxim ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/20950/#review41938 ----------------------------------------------------------- On May 1, 2014, 5:01 p.m., Maxim Khutornenko wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/20950/ > ----------------------------------------------------------- > > (Updated May 1, 2014, 5:01 p.m.) > > > Review request for Aurora, Mark Chu-Carroll and Brian Wickman. > > > Bugs: AURORA-370 > https://issues.apache.org/jira/browse/AURORA-370 > > > Repository: aurora > > > Description > ------- > > Migration plan: > 1. Prepare client to poll for the status of tasks being killed. This is > mostly a noop at this point while the scheduler wait with a shorter timeout > is still in place. > 2. Cut off scheduler killTasks wait thus relying on the client poll (coming > next). > > This RB addresses client changes: > - Repurposing JobMonitor to poll for all job tasks regardless of their > arrival time. > - Introducing a timeout cap into JobMonitor to interrupt wait upon reaching > the max_poll_interval. > - Setting the JobMonitor max_poll_interval to be higher than the scheduler > killTasks backoff to ensure smooth migration. > - Added wait_util calls into kill/killall/updater calls. > > Once this ticket is closed, the next step would be to make kill/killall truly > async: AURORA-371 > > > Diffs > ----- > > src/main/python/apache/aurora/client/api/BUILD > 32097d220a6f8a58555260ce081ace0c58b9fa99 > src/main/python/apache/aurora/client/api/job_monitor.py > b694ef62ba708c4b67c5cc931b0edfd93702027f > src/main/python/apache/aurora/client/api/updater.py > 0acf45034db124ed369d0bcb8095fe5545605eed > src/main/python/apache/aurora/client/cli/jobs.py > 782b34800def55c54b67ca5f13da66c4997a0777 > src/main/python/apache/aurora/client/cli/task.py > 62747ed3e3f5f6ecd9b9c46f6f175c634c068358 > src/main/python/apache/aurora/client/commands/core.py > 39190e06f3eac1ab4dabf48055418bc383be14ed > src/test/python/apache/aurora/client/api/BUILD > dd9b79764db0bbc3a11c51051e43b93a9b5d370a > src/test/python/apache/aurora/client/api/test_job_monitor.py > 32609e4e97afc7b8b164f3db438c60efd0ebbe11 > src/test/python/apache/aurora/client/api/test_updater.py > e7eb1e783baaabe9437e55ecb848f6a42de13dd7 > src/test/python/apache/aurora/client/cli/test_command_hooks.py > 7c6f70c7ef7534e9dd4986364331c67f6f7c36b1 > src/test/python/apache/aurora/client/cli/test_create.py > 875573e2ee534ea50da71fac2174ecc8877a714d > src/test/python/apache/aurora/client/cli/test_kill.py > cf5df648206e0cc234d603c1fd56dc7ab1311fa9 > src/test/python/apache/aurora/client/cli/test_plugins.py > 2dab749d1dcba99a1e3c792c8f26ee86441673d1 > src/test/python/apache/aurora/client/cli/test_update.py > cf077e821ebe32104b3e1345014f5c7b07c44c34 > src/test/python/apache/aurora/client/cli/util.py > 2225ab071dff72be547c43f66ffdc890171c3288 > src/test/python/apache/aurora/client/commands/test_create.py > e0ecb523db4ce3b1901adcc96a2fd5ce8184f621 > src/test/python/apache/aurora/client/commands/test_kill.py > db820a543b0ea11e5a40cc2a00209aa8fa6186c9 > src/test/python/apache/aurora/client/commands/test_update.py > 6e145db9ff8213c8f098b53b9a9668532d664249 > > Diff: https://reviews.apache.org/r/20950/diff/ > > > Testing > ------- > > ./pants src/test/python:all > > > Thanks, > > Maxim Khutornenko > >
