KYLIN-4432 duplicated queries with sytax error take unexpect long time when lazy query enabled

[xiacongling]

Proposed changes

This PR fix the bug described in KYLIN-4432 that duplicated queries with sytax error take unexpect long time when lazy query enabled.

Types of changes

What types of changes does your code introduce to Kylin?
Put an x in the boxes that apply

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation Update (if none of the other choices apply)

Checklist

Put an x in the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your code.

• I have create an issue on Kylin's jira, and have described the bug/feature there in detail
• Commit messages in my PR start with the related jira ID, like "KYLIN-0000 Make Kylin project open-source"
• Compiling and unit tests pass locally with my changes
• I have added tests that prove my fix is effective or that my feature works
• If this change need a document change, I will prepare another pr against the document branch
• Any dependent changes have been merged

Further comments

N/A

[coveralls]

Pull Request Test Coverage Report for Build 5807

• 0 of 6 (0.0%) changed or added relevant lines in 1 file are covered.
• 387 unchanged lines in 13 files lost coverage.
• Overall coverage decreased (-0.05%) to 27.475%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
server-base/src/main/java/org/apache/kylin/rest/service/QueryService.java	0	6	0.0%

Files with Coverage Reduction	New Missed Lines	%
engine-mr/src/main/java/org/apache/kylin/engine/mr/steps/CreateDictionaryJob.java	1	0%
engine-mr/src/main/java/org/apache/kylin/engine/mr/common/MapReduceExecutable.java	3	14.76%
core-cube/src/main/java/org/apache/kylin/cube/inmemcubing/MemDiskStore.java	5	77.81%
server-base/src/main/java/org/apache/kylin/rest/service/DiagnosisService.java	5	32.14%
core-cube/src/main/java/org/apache/kylin/cube/common/SegmentPruner.java	9	69.23%
core-metrics/src/main/java/org/apache/kylin/metrics/MetricsManager.java	14	0%
core-common/src/main/java/org/apache/kylin/common/util/ToolUtil.java	18	21.28%
core-common/src/main/java/org/apache/kylin/common/util/HadoopUtil.java	40	18.45%
engine-mr/src/main/java/org/apache/kylin/engine/mr/CubingJob.java	40	1.53%
core-cube/src/main/java/org/apache/kylin/cube/CubeManager.java	41	42.16%

Totals
Change from base Build 5776:	-0.05%
Covered Lines:	24316
Relevant Lines:	88502

---

💛 - Coveralls

[zhangayqian]

Verified.
In kylin 3.0.1，KYLIN-4432 could be reproduce.

Add this PR will fix.

[kyotoYaho]

LGTM. How about extract evicting the dummy response logic to the finally block?

[xiacongling]

@kyotoYaho Thanks for the suggestion. I updated the PR with a finally block.

[codecov-io]

Codecov Report

Merging #1167 into master will decrease coverage by 0.05%.
The diff coverage is 0.00%.

@@             Coverage Diff              @@
##             master    #1167      +/-   ##
============================================
- Coverage     25.06%   25.01%   -0.06%     
- Complexity     6250     6253       +3     
============================================
  Files          1446     1449       +3     
  Lines         88309    88502     +193     
  Branches      12358    12374      +16     
============================================
  Hits          22136    22136              
- Misses        64003    64189     +186     
- Partials       2170     2177       +7     

Impacted Files	Coverage Δ	Complexity Δ
...va/org/apache/kylin/rest/service/QueryService.java	0.00% <0.00%> (ø)	0.00 <0.00> (ø)
...org/apache/kylin/rest/util/QueryRequestLimits.java	35.71% <0.00%> (-4.77%)	5.00% <0.00%> (-1.00%)
...va/org/apache/kylin/cube/common/SegmentPruner.java	58.46% <0.00%> (-4.04%)	26.00% <0.00%> (+1.00%)	⬇️
...rg/apache/kylin/cube/inmemcubing/MemDiskStore.java	69.30% <0.00%> (-1.83%)	7.00% <0.00%> (ø%)
...he/kylin/engine/mr/common/MapReduceExecutable.java	12.91% <0.00%> (-0.42%)	6.00% <0.00%> (ø%)
...ain/java/org/apache/kylin/engine/mr/CubingJob.java	1.53% <0.00%> (-0.04%)	2.00% <0.00%> (ø%)
...n/java/org/apache/kylin/tool/DiagnosisInfoCLI.java	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
...ava/org/apache/kylin/tool/JobDiagnosisInfoCLI.java	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
.../java/org/apache/kylin/common/util/HadoopUtil.java	14.56% <0.00%> (ø)	10.00% <0.00%> (ø%)
.../java/org/apache/kylin/metrics/MetricsManager.java	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
... and 26 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 734a997...7034a04. Read the comment docs.

[kyotoYaho]

Hi xiacongling, the previous one without finally block is better. I'm sorry I mislead you.

[xiacongling]

Never mind, @kyotoYaho ! I think the problem is that conditions are lost in finally block. We can either rebuild the conditions or use the cached value.

• The former solution will add complicated condition checks in the final block. These checks are difficult to understand and test, and may be buggy in the future when features are addd or improvements are made.
• The latter will visit the cache one more time and will reduce the query performance.

Using "try .. finally" is more elegant, but some code refactoring is needed to keep it clean, I think. To get out of the dilemma, we can treat this PR as a quick bug-fix, and discuss the refactoring in another issue.

I will revert the commit. Thanks for the review!

[shaofengshi]

+1 LGTM