[kudu-CR] [java] Deflake TestKuduScanner.testDiffScan

2019-06-03 Thread Will Berkeley (Code Review) 
Will Berkeley has submitted this change and it was merged. ( 
http://gerrit.cloudera.org:8080/13495 )

Change subject: [java] Deflake TestKuduScanner.testDiffScan
..

[java] Deflake TestKuduScanner.testDiffScan

I noticed a funny failure in TestKuduScanner.testDiffScan:

21:55:41.681 [DEBUG - New I/O worker #13] (AsyncKuduScanner.java:556) Can not 
open scanner
org.apache.kudu.client.NonRecoverableException: snapshot scan start timestamp 
is earlier than the ancient history mark. Consider increasing the value of the 
configuration parameter --tablet_history_max_age_sec. Snapshot timestamp: P: 0 
usec, L: 0 Ancient History Mark: P: 1558907741677535 usec, L: 0 Physical time 
difference: -1558907741.678s

The client logs the request it sent, which indeed has a
'snap_start_timestamp' of 0. This timestamp is supposed to come from
incrementing the propagated timestamp received from the client after
some random inserts and mutations. Fortunately, this test has nice
logging: it logs the operations it did and the timestamp:

21:55:40.874 [INFO - Time-limited test] (TestKuduScanner.java:250) Before: {}
21:55:40.874 [INFO - Time-limited test] (TestKuduScanner.java:255) startHT: 0

'generateMutationOperations' could generate 0 mutations, causing this
test to fail. This requires the random number generator to generate 0 3
times in a row when sampling uniformly from {0, 1, 2, 3, 4}, so the odds
of failure due to this defect in the test were only 1/125. I fixed this
by adjusting the number of initial inserts generated so it is always at
least 1.

I ran the test 1250 times with the fix and saw zero failures. Without
the fix, I saw 9/1250 failures.

Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Reviewed-on: http://gerrit.cloudera.org:8080/13495
Tested-by: Will Berkeley 
Reviewed-by: Alexey Serbin 
---
M java/kudu-client/src/test/java/org/apache/kudu/client/TestKuduScanner.java
M java/kudu-test-utils/src/main/java/org/apache/kudu/test/RandomUtils.java
2 files changed, 32 insertions(+), 14 deletions(-)

Approvals:
  Will Berkeley: Verified
  Alexey Serbin: Looks good to me, approved

--
To view, visit http://gerrit.cloudera.org:8080/13495
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: merged
Gerrit-Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Gerrit-Change-Number: 13495
Gerrit-PatchSet: 3
Gerrit-Owner: Will Berkeley 
Gerrit-Reviewer: Adar Dembo 
Gerrit-Reviewer: Alexey Serbin 
Gerrit-Reviewer: Grant Henke 
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Will Berkeley

[kudu-CR] [java] Deflake TestKuduScanner.testDiffScan

2019-06-03 Thread Alexey Serbin (Code Review) 
Alexey Serbin has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/13495 )

Change subject: [java] Deflake TestKuduScanner.testDiffScan
..


Patch Set 2: Code-Review+2


--
To view, visit http://gerrit.cloudera.org:8080/13495
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Gerrit-Change-Number: 13495
Gerrit-PatchSet: 2
Gerrit-Owner: Will Berkeley 
Gerrit-Reviewer: Adar Dembo 
Gerrit-Reviewer: Alexey Serbin 
Gerrit-Reviewer: Grant Henke 
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Will Berkeley 
Gerrit-Comment-Date: Mon, 03 Jun 2019 06:54:52 +
Gerrit-HasComments: No

[kudu-CR] [java] Deflake TestKuduScanner.testDiffScan

2019-06-03 Thread Will Berkeley (Code Review) 
Will Berkeley has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/13495 )

Change subject: [java] Deflake TestKuduScanner.testDiffScan
..


Patch Set 2: Verified+1

When will the flakes end :'(


--
To view, visit http://gerrit.cloudera.org:8080/13495
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Gerrit-Change-Number: 13495
Gerrit-PatchSet: 2
Gerrit-Owner: Will Berkeley 
Gerrit-Reviewer: Adar Dembo 
Gerrit-Reviewer: Alexey Serbin 
Gerrit-Reviewer: Grant Henke 
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Will Berkeley 
Gerrit-Comment-Date: Mon, 03 Jun 2019 06:02:43 +
Gerrit-HasComments: No

[kudu-CR] [java] Deflake TestKuduScanner.testDiffScan

2019-06-03 Thread Will Berkeley (Code Review) 
Will Berkeley has removed a vote on this change.

Change subject: [java] Deflake TestKuduScanner.testDiffScan
..


Removed Verified-1 by Kudu Jenkins (120)
--
To view, visit http://gerrit.cloudera.org:8080/13495
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: deleteVote
Gerrit-Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Gerrit-Change-Number: 13495
Gerrit-PatchSet: 2
Gerrit-Owner: Will Berkeley 
Gerrit-Reviewer: Adar Dembo 
Gerrit-Reviewer: Alexey Serbin 
Gerrit-Reviewer: Grant Henke 
Gerrit-Reviewer: Kudu Jenkins (120)

[kudu-CR] [java] Deflake TestKuduScanner.testDiffScan

2019-06-02 Thread Will Berkeley (Code Review) 
Hello Alexey Serbin, Kudu Jenkins, Adar Dembo, Grant Henke,

I'd like you to reexamine a change. Please visit

http://gerrit.cloudera.org:8080/13495

to look at the new patch set (#2).

Change subject: [java] Deflake TestKuduScanner.testDiffScan
..

[java] Deflake TestKuduScanner.testDiffScan

I noticed a funny failure in TestKuduScanner.testDiffScan:

21:55:41.681 [DEBUG - New I/O worker #13] (AsyncKuduScanner.java:556) Can not 
open scanner
org.apache.kudu.client.NonRecoverableException: snapshot scan start timestamp 
is earlier than the ancient history mark. Consider increasing the value of the 
configuration parameter --tablet_history_max_age_sec. Snapshot timestamp: P: 0 
usec, L: 0 Ancient History Mark: P: 1558907741677535 usec, L: 0 Physical time 
difference: -1558907741.678s

The client logs the request it sent, which indeed has a
'snap_start_timestamp' of 0. This timestamp is supposed to come from
incrementing the propagated timestamp received from the client after
some random inserts and mutations. Fortunately, this test has nice
logging: it logs the operations it did and the timestamp:

21:55:40.874 [INFO - Time-limited test] (TestKuduScanner.java:250) Before: {}
21:55:40.874 [INFO - Time-limited test] (TestKuduScanner.java:255) startHT: 0

'generateMutationOperations' could generate 0 mutations, causing this
test to fail. This requires the random number generator to generate 0 3
times in a row when sampling uniformly from {0, 1, 2, 3, 4}, so the odds
of failure due to this defect in the test were only 1/125. I fixed this
by adjusting the number of initial inserts generated so it is always at
least 1.

I ran the test 1250 times with the fix and saw zero failures. Without
the fix, I saw 9/1250 failures.

Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
---
M java/kudu-client/src/test/java/org/apache/kudu/client/TestKuduScanner.java
M java/kudu-test-utils/src/main/java/org/apache/kudu/test/RandomUtils.java
2 files changed, 32 insertions(+), 14 deletions(-)


  git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/95/13495/2
--
To view, visit http://gerrit.cloudera.org:8080/13495
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Gerrit-Change-Number: 13495
Gerrit-PatchSet: 2
Gerrit-Owner: Will Berkeley 
Gerrit-Reviewer: Adar Dembo 
Gerrit-Reviewer: Alexey Serbin 
Gerrit-Reviewer: Grant Henke 
Gerrit-Reviewer: Kudu Jenkins (120)

[kudu-CR] [java] Deflake TestKuduScanner.testDiffScan

2019-06-02 Thread Will Berkeley (Code Review) 
Will Berkeley has uploaded this change for review. ( 
http://gerrit.cloudera.org:8080/13495


Change subject: [java] Deflake TestKuduScanner.testDiffScan
..

[java] Deflake TestKuduScanner.testDiffScan

I noticed a funny failure in TestKuduScanner.testDiffScan:

21:55:41.681 [DEBUG - New I/O worker #13] (AsyncKuduScanner.java:556) Can not 
open scanner
org.apache.kudu.client.NonRecoverableException: snapshot scan start timestamp 
is earlier than the ancient history mark. Consider increasing the value of the 
configuration parameter --tablet_history_max_age_sec. Snapshot timestamp: P: 0 
usec, L: 0 Ancient History Mark: P: 1558907741677535 usec, L: 0 Physical time 
difference: -1558907741.678s

The client logs the request it sent, which indeed has a 'snap_start_timestamp'
of 0. This timestamp is supposed to come from incrementing the propagated
timestamp received from the client after some random inserts and mutations.
Fortunately, this test has nice logging: it logs the operations it did and the
timestamp:

21:55:40.874 [INFO - Time-limited test] (TestKuduScanner.java:250) Before: {}
21:55:40.874 [INFO - Time-limited test] (TestKuduScanner.java:255) startHT: 0

'generateMutationOperations' could generate 0 mutations, causing this test to
fail. This requires the random number generate to generate 0 3 times in
a row when sample uniformly from {0, 1, 2, 3, 4}, so the odds of failure
due to this defect in the tests were only 1/125. I fixed this by
adjusting the number of inserts generated so it is always at least 1.

I ran the test 1250 times with the fix and saw zero failures. Without
the fix, I saw 9/1250 failures.

Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
---
M java/kudu-client/src/test/java/org/apache/kudu/client/TestKuduScanner.java
M java/kudu-test-utils/src/main/java/org/apache/kudu/test/RandomUtils.java
2 files changed, 32 insertions(+), 14 deletions(-)



  git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/95/13495/1
--
To view, visit http://gerrit.cloudera.org:8080/13495
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newchange
Gerrit-Change-Id: Iadb46e5dae71724aa8ff88d04c40ef4eaf1ddd2a
Gerrit-Change-Number: 13495
Gerrit-PatchSet: 1
Gerrit-Owner: Will Berkeley