[GitHub] incubator-livy pull request #107: [LIVY-494] Add thriftserver to Livy server
Github user vanzin commented on a diff in the pull request: https://github.com/apache/incubator-livy/pull/107#discussion_r219612825 --- Diff: server/src/main/scala/org/apache/livy/server/LivyServer.scala --- @@ -115,6 +117,16 @@ class LivyServer extends Logging { error("Failed to run kinit, stopping the server.") sys.exit(1) } + // This is and should be the only place where a login() on the UGI is performed. --- End diff -- This should be pretty easy to check with a scalastyle rule that disallows calls to `UserGroupInformation.login*` (and disabling the check around this call). Not foolproof, but should catch inadvertent use at least. ---
[GitHub] incubator-livy issue #112: [LIVY-511][LIVY-512] Remove support for old Spark...
Github user codecov-io commented on the issue: https://github.com/apache/incubator-livy/pull/112 # [Codecov](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=h1) Report > Merging [#112](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=desc) into [master](https://codecov.io/gh/apache/incubator-livy/commit/d39ab356b4e36d8780bd44b9d5020bfeec90cb2f?src=pr=desc) will **increase** coverage by `<.01%`. > The diff coverage is `57.14%`. [![Impacted file tree graph](https://codecov.io/gh/apache/incubator-livy/pull/112/graphs/tree.svg?width=650=0MkVbiUFwE=150=pr)](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=tree) ```diff @@ Coverage Diff @@ ## master #112 +/- ## + Coverage 71.12% 71.12% +<.01% - Complexity 796 928 +132 Files98 99 +1 Lines 5500 5483 -17 Branches829 828 -1 - Hits 3912 3900 -12 - Misses 1056 1059 +3 + Partials532 524 -8 ``` | [Impacted Files](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=tree) | Coverage Π| Complexity Π| | |---|---|---|---| | [.../scala/org/apache/livy/repl/SparkInterpreter.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-cmVwbC9zY2FsYS0yLjExL3NyYy9tYWluL3NjYWxhL29yZy9hcGFjaGUvbGl2eS9yZXBsL1NwYXJrSW50ZXJwcmV0ZXIuc2NhbGE=) | `82.97% <ø> (ø)` | `10 <0> (?)` | | | [...rver/src/main/scala/org/apache/livy/LivyConf.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-c2VydmVyL3NyYy9tYWluL3NjYWxhL29yZy9hcGFjaGUvbGl2eS9MaXZ5Q29uZi5zY2FsYQ==) | `93.79% <ø> (ø)` | `21 <0> (ø)` | :arrow_down: | | [...la/org/apache/livy/server/batch/BatchSession.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-c2VydmVyL3NyYy9tYWluL3NjYWxhL29yZy9hcGFjaGUvbGl2eS9zZXJ2ZXIvYmF0Y2gvQmF0Y2hTZXNzaW9uLnNjYWxh) | `85.22% <ø> (-0.17%)` | `13 <0> (ø)` | | | [...main/java/org/apache/livy/rsc/ContextLauncher.java](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-cnNjL3NyYy9tYWluL2phdmEvb3JnL2FwYWNoZS9saXZ5L3JzYy9Db250ZXh0TGF1bmNoZXIuamF2YQ==) | `82.65% <100%> (-1.27%)` | `22 <0> (+4)` | | | [...in/scala/org/apache/livy/repl/SQLInterpreter.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-cmVwbC9zcmMvbWFpbi9zY2FsYS9vcmcvYXBhY2hlL2xpdnkvcmVwbC9TUUxJbnRlcnByZXRlci5zY2FsYQ==) | `70.37% <100%> (-2.61%)` | `7 <1> (+6)` | | | [...n/scala/org/apache/livy/utils/LivySparkUtils.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-c2VydmVyL3NyYy9tYWluL3NjYWxhL29yZy9hcGFjaGUvbGl2eS91dGlscy9MaXZ5U3BhcmtVdGlscy5zY2FsYQ==) | `67.6% <100%> (-4.02%)` | `0 <0> (ø)` | | | [...e/livy/server/interactive/InteractiveSession.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-c2VydmVyL3NyYy9tYWluL3NjYWxhL29yZy9hcGFjaGUvbGl2eS9zZXJ2ZXIvaW50ZXJhY3RpdmUvSW50ZXJhY3RpdmVTZXNzaW9uLnNjYWxh) | `69.2% <25%> (+2.02%)` | `43 <0> (ø)` | :arrow_down: | | [.../java/org/apache/livy/rsc/driver/SparkEntries.java](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-cnNjL3NyYy9tYWluL2phdmEvb3JnL2FwYWNoZS9saXZ5L3JzYy9kcml2ZXIvU3BhcmtFbnRyaWVzLmphdmE=) | `64.06% <33.33%> (-0.33%)` | `10 <0> (-1)` | | | [...ava/org/apache/livy/rsc/driver/JobContextImpl.java](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-cnNjL3NyYy9tYWluL2phdmEvb3JnL2FwYWNoZS9saXZ5L3JzYy9kcml2ZXIvSm9iQ29udGV4dEltcGwuamF2YQ==) | `91.11% <66.66%> (ø)` | `17 <0> (ø)` | :arrow_down: | | [...rg/apache/livy/repl/AbstractSparkInterpreter.scala](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree#diff-cmVwbC9zcmMvbWFpbi9zY2FsYS9vcmcvYXBhY2hlL2xpdnkvcmVwbC9BYnN0cmFjdFNwYXJrSW50ZXJwcmV0ZXIuc2NhbGE=) | `52.5% <0%> (-4.3%)` | `30% <0%> (+29%)` | | | ... and [29 more](https://codecov.io/gh/apache/incubator-livy/pull/112/diff?src=pr=tree-more) | | -- [Continue to review full report at Codecov](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=continue). > **Legend** - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta) > `Π= absolute (impact)`, `ø = not affected`, `? = missing data` > Powered by [Codecov](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=footer). Last update [d39ab35...6459626](https://codecov.io/gh/apache/incubator-livy/pull/112?src=pr=lastupdated). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments). ---
[GitHub] incubator-livy pull request #112: [LIVY-511][LIVY-512] Remove support for ol...
GitHub user vanzin opened a pull request: https://github.com/apache/incubator-livy/pull/112 [LIVY-511][LIVY-512] Remove support for old Spark, Scala versions. This change restricts Livy support to Spark 2.2+ and Scala 2.11. Both changes are made together because by supporting Spark 2.2+ only, it becomes impossible to test Scala 2.10. As part of the change, a lot of code that used reflection to support different versions of Spark could be cleaned up and directly call Spark APIs. The Scala 2.10 parts of the builds also have been removed, but the actual support for building and running with different Scala versions (and related tests) have been left as is. This will allow us to support 2.12 in the future. This change intentionally does not touch the public API (the "api/" module). There are things that could be cleaned up now that Spark 1.x is not supported, but that would mean an API breakage so I chose to leave those alone for now. The test matrix and build profiles have also been simplified a lot. There are now two profiles to choose from (for Spark 2.2 and 2.3); integration tests can be run against a different version of Spark by running just the integration test module with the desired profile. Tested with Spark 2.2 and 2.3, and also by building against 2.2 and running integration tests against 2.3. You can merge this pull request into a Git repository by running: $ git pull https://github.com/vanzin/incubator-livy LIVY-511 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/incubator-livy/pull/112.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #112 commit 6459626488a050935e25cceafe4e8b22e91e3012 Author: Marcelo Vanzin Date: 2018-09-20T18:39:23Z [LIVY-511][LIVY-512] Remove support for old Spark, Scala versions. This change restricts Livy support to Spark 2.2+ and Scala 2.11. Both changes are made together because by supporting Spark 2.2+ only, it becomes impossible to test Scala 2.10. As part of the change, a lot of code that used reflection to support different versions of Spark could be cleaned up and directly call Spark APIs. The Scala 2.10 parts of the builds also have been removed, but the actual support for building and running with different Scala versions (and related tests) have been left as is. This will allow us to support 2.12 in the future. This change intentionally does not touch the public API (the "api/" module). There are things that could be cleaned up now that Spark 1.x is not supported, but that would mean an API breakage so I chose to leave those alone for now. The test matrix and build profiles have also been simplified a lot. There are now two profiles to choose from (for Spark 2.2 and 2.3); integration tests can be run against a different version of Spark by running just the integration test module with the desired profile. Tested with Spark 2.2 and 2.3, and also by building against 2.2 and running integration tests against 2.3. ---