You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues-all@impala.apache.org by "Quanlong Huang (JIRA)" <ji...@apache.org> on 2019/06/16 03:20:00 UTC
[jira] [Updated] (IMPALA-8669) Support giving a terminal commit in
compare_branches.py
[ https://issues.apache.org/jira/browse/IMPALA-8669?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Quanlong Huang updated IMPALA-8669:
-----------------------------------
Description:
Currently, bin/compare_branches.py (used in cherrypick-2.x-and-test Jenkins job) will cherry-pick as more commits as possible. However, a clean pick from the master branch does not mean it can always pass the tests. If we find such a problematic commit, we'd like to let the cherrypick-2.x-and-test Jenkins job cherry-pick till the commit before it. And then submit only a patch for the problematic commit for review.
For example, currently the HEAD of the 2.x branch is e6c1eb85eb31264eaf5ed92782bf181225ce9581. bin/compare_branches.py will pick the following commits of the master branch:
|200870275|IMPALA-7314: Doc generation should fail on error| |
|940d536f2|IMPALA-7252: Backport rate limiting of fadvise calls into toolchain glog| |
|c7d2c2ec7|IMPALA-7298: Stop passing IP address as hostname in Kerberos principal| |
|70e2d57fc|IMPALA-7315: fix test_update_with_clear_entries_flag race| |
|1d491d648|IMPALA-7259: Improve Impala shell performance| |
|2a40e8f2a|test_recover_partitions.py had asserts that were always true.| |
|649397e37|KUDU-2492: Make the use of SO_REUSEPORT conditional on it being defined| |
|eaea7d289|IMPALA-6677: [DOCS] Document the next_day function| |
|79eef4924|IMPALA-7256: Aggregator mem usage isn't reflected in summary| |
|42809cbd1|IMPALA-7173: [DOCS] Added check options in the load balancer examples| |
|cb0f8a0ad|IMPALA-3675: part 1: -Werror for ASAN| |
|f7efba236|IMPALA-5031: Fix undefined behavior: memset NULL| |
|514dbb79a|IMPALA-6299: Use LlvmCodeGen's internal list of white-listed CPU attributes for handcrafting IRs| |
|f9e7d9385|IMPALA-7291: [DOCS] Note about no codegen support for CHAR| |
|ac4acf1b7|IMPALA-3040: Remove cache directives if a partition is dropped externally| |
|02389d4dd|IMPALA-6174: [DOCS] Fixed the seed data type for RAND and RANDOM functions| |
|21d0c06a4|IMPALA-5826 IMPALA-7162: [DOCS] Documented the IDLE_SESSION_TIMEOUT query option| |
|b76207c59|IMPALA-7330. After LOAD DATA, only refresh affected partition| |
|cdc8b9ba7|IMPALA-5031: Fix undefined behavior: memset NULL| |
|def5c881b|IMPALA-7218: [DOCS] Support column list in ALTER VIEW| |
|c333b5526|IMPALA-7257. Support Kudu tables in LocalCatalog|<-- Clean picked but introduced compile errors.|
|ba8138694|IMPALA-7276. Support CREATE TABLE AS SELECT with LocalCatalog| |
c333b5526 is a clean pick but it will introduce a compile error in FE:
{code:java}
[ERROR] /home/ubuntu/Impala/fe/src/test/java/org/apache/impala/catalog/local/LocalCatalogTest.java:[193,68] cannot find symbol
[ERROR] symbol: method startsWith(java.lang.String)
[ERROR] location: class org.hamcrest.CoreMatchers{code}
We hope the cherrypick-2.x-and-test Jenkins job can pick till def5c881b (the previous commit) and merge these good commits. Then we can just submit one patch to fix the pick of c333b5526.
was:Currently, bin/compare_branches.py (used in cherrypick-2.x-and-test Jenkins job) will cherry-pick as more commits as possible. However, a clean pick from the master branch does not mean it can always pass the tests. If we find such a problematic commit, we'd like to let the cherrypick-2.x-and-test Jenkins job cherry-pick till the commit before it. And then submit only a patch for the problematic commit for review.
> Support giving a terminal commit in compare_branches.py
> -------------------------------------------------------
>
> Key: IMPALA-8669
> URL: https://issues.apache.org/jira/browse/IMPALA-8669
> Project: IMPALA
> Issue Type: Improvement
> Affects Versions: Impala 2.12.0
> Reporter: Quanlong Huang
> Assignee: Quanlong Huang
> Priority: Major
>
> Currently, bin/compare_branches.py (used in cherrypick-2.x-and-test Jenkins job) will cherry-pick as more commits as possible. However, a clean pick from the master branch does not mean it can always pass the tests. If we find such a problematic commit, we'd like to let the cherrypick-2.x-and-test Jenkins job cherry-pick till the commit before it. And then submit only a patch for the problematic commit for review.
> For example, currently the HEAD of the 2.x branch is e6c1eb85eb31264eaf5ed92782bf181225ce9581. bin/compare_branches.py will pick the following commits of the master branch:
> |200870275|IMPALA-7314: Doc generation should fail on error| |
> |940d536f2|IMPALA-7252: Backport rate limiting of fadvise calls into toolchain glog| |
> |c7d2c2ec7|IMPALA-7298: Stop passing IP address as hostname in Kerberos principal| |
> |70e2d57fc|IMPALA-7315: fix test_update_with_clear_entries_flag race| |
> |1d491d648|IMPALA-7259: Improve Impala shell performance| |
> |2a40e8f2a|test_recover_partitions.py had asserts that were always true.| |
> |649397e37|KUDU-2492: Make the use of SO_REUSEPORT conditional on it being defined| |
> |eaea7d289|IMPALA-6677: [DOCS] Document the next_day function| |
> |79eef4924|IMPALA-7256: Aggregator mem usage isn't reflected in summary| |
> |42809cbd1|IMPALA-7173: [DOCS] Added check options in the load balancer examples| |
> |cb0f8a0ad|IMPALA-3675: part 1: -Werror for ASAN| |
> |f7efba236|IMPALA-5031: Fix undefined behavior: memset NULL| |
> |514dbb79a|IMPALA-6299: Use LlvmCodeGen's internal list of white-listed CPU attributes for handcrafting IRs| |
> |f9e7d9385|IMPALA-7291: [DOCS] Note about no codegen support for CHAR| |
> |ac4acf1b7|IMPALA-3040: Remove cache directives if a partition is dropped externally| |
> |02389d4dd|IMPALA-6174: [DOCS] Fixed the seed data type for RAND and RANDOM functions| |
> |21d0c06a4|IMPALA-5826 IMPALA-7162: [DOCS] Documented the IDLE_SESSION_TIMEOUT query option| |
> |b76207c59|IMPALA-7330. After LOAD DATA, only refresh affected partition| |
> |cdc8b9ba7|IMPALA-5031: Fix undefined behavior: memset NULL| |
> |def5c881b|IMPALA-7218: [DOCS] Support column list in ALTER VIEW| |
> |c333b5526|IMPALA-7257. Support Kudu tables in LocalCatalog|<-- Clean picked but introduced compile errors.|
> |ba8138694|IMPALA-7276. Support CREATE TABLE AS SELECT with LocalCatalog| |
> c333b5526 is a clean pick but it will introduce a compile error in FE:
> {code:java}
> [ERROR] /home/ubuntu/Impala/fe/src/test/java/org/apache/impala/catalog/local/LocalCatalogTest.java:[193,68] cannot find symbol
> [ERROR] symbol: method startsWith(java.lang.String)
> [ERROR] location: class org.hamcrest.CoreMatchers{code}
> We hope the cherrypick-2.x-and-test Jenkins job can pick till def5c881b (the previous commit) and merge these good commits. Then we can just submit one patch to fix the pick of c333b5526.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscribe@impala.apache.org
For additional commands, e-mail: issues-all-help@impala.apache.org