> On Dec. 12, 2017, 11:15 p.m., Stephan Erb wrote: > > The code looks fine and reasonable to me. I would still recommend proper > > scale testing though. > > > > At my company, we operate at a small scale and the Mesos replicated log > > works still well for us. Operating a HA MySQL or PostgreSQL cluster would > > come with a higher operational burden. If possible, I would therefore like > > to see that Aurora continues to support its simple replicated log > > deployment mode for now. I believe that other memobers of the community > > might feel similar. I can totally see though that this does not work for > > Twitter > > Mohit Jaggi wrote: > +1 to this. Our scale is high (not due to cluster size but due to a lot > of update/rollback/failure activity) and I like this option. But I don't want > to depend on external entities. The dependency on Zookeeper and Mesos is > already a problem. > > Bill Farner wrote: > Noted. I had a weak inclination to replace the replicated log with a > `ZooKeeperPersistence` implementation for OSS use. Would that be of interest > to either of you as an alternative?
I have not looked very deeply into ZK persistence yet. From what I have heared, Marathon has been struggerling with its ZK storage backend for years. On the upside, removing the replicated log from Aurora would be one further steps towards removing the libmesos dependency from the scheduler. - Stephan ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/64288/#review193234 ----------------------------------------------------------- On Jan. 5, 2018, 1:30 a.m., Bill Farner wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/64288/ > ----------------------------------------------------------- > > (Updated Jan. 5, 2018, 1:30 a.m.) > > > Review request for Aurora, Daniel Knightly, Jordan Ly, and Stephan Erb. > > > Repository: aurora > > > Description > ------- > > Introduces a `Persistence` implementation that uses a SQL database via JDBC. > I've opted to lean towards MySQL SQL dialect, as unfortunately there are > vendor differences for even the very simple SQL used here. > > I chose [HikariCP](https://github.com/brettwooldridge/HikariCP) to serve as a > `DataSource` (connection pool) implementation. We don't really need much out > of a connection pool aside from general connection lifecycle management (i.e. > not for concurrency). I chose this library based on recent development > activity and several positive comparisons to other pools. > > Note that the implementation is not yet wired into the scheduler application. > That will come in a follow-up. > > > Diffs > ----- > > build.gradle af119910e84c48f75f2573ababcfa287c3b986fc > config/spotbugs/excludeFilter.xml 51790cce8d9047e40741f05ee55af15dbdc3065e > examples/vagrant/upstart/aurora-scheduler.conf > dbbe1d1689ed3e455a95f529f914dc6823427d37 > src/dist/etc/h2-database.properties PRE-CREATION > src/main/java/org/apache/aurora/scheduler/app/SchedulerMain.java > 7ffcf4f471b97e32426c82972472115c5c5c4d02 > src/main/java/org/apache/aurora/scheduler/config/CliOptions.java > b7f43e0d6efbddcac640c3d39c7bc56400e68e68 > > src/main/java/org/apache/aurora/scheduler/log/mesos/MesosLogStreamModule.java > a4984a95f938396c82244f91e4a3d592df1c1539 > src/main/java/org/apache/aurora/scheduler/storage/backup/BackupReader.java > PRE-CREATION > > src/main/java/org/apache/aurora/scheduler/storage/durability/DurableStorage.java > 85b2113631586f43d854c4d2812f43b7b864d452 > > src/main/java/org/apache/aurora/scheduler/storage/durability/Persistence.java > 9eb862c01bf451252bfbcc7a2eac60d2c965c9f0 > src/main/java/org/apache/aurora/scheduler/storage/log/LogPersistence.java > e70e6051582ca90ae72014626b983bbf4b8d5b48 > > src/main/java/org/apache/aurora/scheduler/storage/sql/DisabledDistributedSnapshotStore.java > PRE-CREATION > src/main/java/org/apache/aurora/scheduler/storage/sql/Mode.java > PRE-CREATION > src/main/java/org/apache/aurora/scheduler/storage/sql/SqlPersistence.java > PRE-CREATION > > src/main/java/org/apache/aurora/scheduler/storage/sql/SqlPersistenceModule.java > PRE-CREATION > src/main/resources/org/apache/aurora/scheduler/storage/sql/schema.sql > PRE-CREATION > src/test/java/org/apache/aurora/scheduler/config/CommandLineTest.java > 5cb5310ed096ca1fb47b980401e3712948271ac4 > > src/test/java/org/apache/aurora/scheduler/storage/durability/DataCompatibilityTest.java > PRE-CREATION > > src/test/java/org/apache/aurora/scheduler/storage/durability/RecoveryTest.java > PRE-CREATION > > src/test/java/org/apache/aurora/scheduler/storage/log/LogPersistenceTest.java > PRE-CREATION > > src/test/java/org/apache/aurora/scheduler/storage/sql/SqlPersistenceTest.java > PRE-CREATION > src/test/sh/org/apache/aurora/e2e/test_end_to_end.sh > 1500bda8844316e0fcff0534a9f9bb9b5cdaea5f > > > Diff: https://reviews.apache.org/r/64288/diff/6/ > > > Testing > ------- > > > Thanks, > > Bill Farner > >
