[kudu-CR] [logging] fix logging init if linking in kudu client library
Alexey Serbin has uploaded this change for review. ( http://gerrit.cloudera.org:8080/10946 Change subject: [logging] fix logging init if linking in kudu client library .. [logging] fix logging init if linking in kudu client library Updated InitGoogleLoggingSafe() and InitGoogleLoggingSafeBasic() to get desired effects from calling InitGoogleLoggingSafe() even if the binary linked with kudu client library. Prior to this fix, the kudu CLI tool could exit on SIGPIPE when master/tserver abruptly closed connection. That's because the call to InitGoogleLoggingSafe() in the main() function of the CLI tool didn't do anything once InitGoogleLoggingSafeBasic() has been inadvertently called by the linked-in kudu_client library. Change-Id: I9438248e7f9699cf14d648160d9d516ab27d6ca2 --- M src/kudu/util/logging.cc M src/kudu/util/logging.h 2 files changed, 65 insertions(+), 13 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/46/10946/1 -- To view, visit http://gerrit.cloudera.org:8080/10946 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newchange Gerrit-Change-Id: I9438248e7f9699cf14d648160d9d516ab27d6ca2 Gerrit-Change-Number: 10946 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Alexey Serbin has abandoned this change. ( http://gerrit.cloudera.org:8080/10938 ) Change subject: [tools] ignore SIGPIPE in kudu CLI .. Abandoned Obsoleted by https://gerrit.cloudera.org/#/c/10946/ -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: abandon Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin Gerrit-Reviewer: Adar Dembo Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins
[kudu-CR] KUDU-2437 Split a tablet into primary key ranges by size
Yao Xu has posted comments on this change. ( http://gerrit.cloudera.org:8080/10406 ) Change subject: KUDU-2437 Split a tablet into primary key ranges by size .. Patch Set 21: (8 comments) http://gerrit.cloudera.org:8080/#/c/10406/14/src/kudu/common/key_range.h File src/kudu/common/key_range.h: http://gerrit.cloudera.org:8080/#/c/10406/14/src/kudu/common/key_range.h@30 PS14, Line 30: std::string start_key, :std::string stop_key, > start/stop_key should be 'std::string' not 'const std::string&' here, other Done http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tablet/rowset_info.h File src/kudu/tablet/rowset_info.h: http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tablet/rowset_info.h@53 PS19, Line 53: size > add period Done http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tablet/rowset_info.cc File src/kudu/tablet/rowset_info.cc: http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tablet/rowset_info.cc@282 PS19, Line 282: > Probably cleaner to put both conditions in the CHECK, e.g. Done http://gerrit.cloudera.org:8080/#/c/10406/14/src/kudu/tablet/tablet-test.cc File src/kudu/tablet/tablet-test.cc: http://gerrit.cloudera.org:8080/#/c/10406/14/src/kudu/tablet/tablet-test.cc@1187 PS14, Line 1187: } > That's true of every interval, though? Is there a reason to expect that th I modified the algorithm to combine these 0 sized chunks with adjacent chunk. http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tablet/tablet-test.cc File src/kudu/tablet/tablet-test.cc: http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tablet/tablet-test.cc@1150 PS19, Line 1150: AssertChunk > 'Assert' Done http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tserver/tablet_service.cc File src/kudu/tserver/tablet_service.cc: http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tserver/tablet_service.cc@1450 PS19, Line 1450: s = ColumnPBsToSchema(req->columns(), ); > Sorry I missed this before, but user requests should never include column I Done http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tserver/tablet_service.cc@1466 PS19, Line 1466: r column_ids; > The first clause here can be dropped, since protobuf will return 0 if it's Done http://gerrit.cloudera.org:8080/#/c/10406/19/src/kudu/tserver/tablet_service.cc@1468 PS19, Line 1468: > target Done -- To view, visit http://gerrit.cloudera.org:8080/10406 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I9ec4395919f4b54102e458ef5154334c08412e8a Gerrit-Change-Number: 10406 Gerrit-PatchSet: 21 Gerrit-Owner: Yao Xu Gerrit-Reviewer: Dan Burkert Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon Gerrit-Reviewer: Yao Xu Gerrit-Comment-Date: Sat, 14 Jul 2018 03:19:10 + Gerrit-HasComments: Yes
[kudu-CR] KUDU-2437 Split a tablet into primary key ranges by size
Hello Dan Burkert, Kudu Jenkins, Todd Lipcon, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/10406 to look at the new patch set (#20). Change subject: KUDU-2437 Split a tablet into primary key ranges by size .. KUDU-2437 Split a tablet into primary key ranges by size When reading data in a kudu table using spark, if there is a large amount of data in the tablet, reading the data takes a long time. The reason is that KuduRDD uses a tablet to generate the scanToken, so a spark task needs to process all the data in a tablet. TabletServer should provide an RPC interface, which can be split tablet into multiple primary key ranges by size. The kudu-client can choose whether to perform parallel scan according to the case. Change-Id: I9ec4395919f4b54102e458ef5154334c08412e8a --- M src/kudu/common/CMakeLists.txt M src/kudu/common/common.proto A src/kudu/common/key_range.cc A src/kudu/common/key_range.h M src/kudu/tablet/cfile_set.cc M src/kudu/tablet/cfile_set.h M src/kudu/tablet/diskrowset.cc M src/kudu/tablet/diskrowset.h M src/kudu/tablet/memrowset.h M src/kudu/tablet/mock-rowsets.h M src/kudu/tablet/rowset.cc M src/kudu/tablet/rowset.h M src/kudu/tablet/rowset_info.cc M src/kudu/tablet/rowset_info.h M src/kudu/tablet/tablet-harness.h M src/kudu/tablet/tablet-test-util.h M src/kudu/tablet/tablet-test.cc M src/kudu/tablet/tablet.cc M src/kudu/tablet/tablet.h M src/kudu/tserver/tablet_service.cc M src/kudu/tserver/tablet_service.h M src/kudu/tserver/tserver.proto M src/kudu/tserver/tserver_service.proto 23 files changed, 721 insertions(+), 6 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/06/10406/20 -- To view, visit http://gerrit.cloudera.org:8080/10406 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I9ec4395919f4b54102e458ef5154334c08412e8a Gerrit-Change-Number: 10406 Gerrit-PatchSet: 20 Gerrit-Owner: Yao Xu Gerrit-Reviewer: Dan Burkert Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon Gerrit-Reviewer: Yao Xu
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/10938 ) Change subject: [tools] ignore SIGPIPE in kudu CLI .. Patch Set 1: > (1 comment) Yep, sure -- I'll do. But so far I'm sure the SIGPIPE is not ignored since the kudu CLI tool binary exits upon SIGPIPE, and you can see that not only while running via kudu::tools::RunTool(). * vc1320:~$ ps axw | grep kudu 14458 pts/1S+ 0:00 sudo -u kudu ./kudu cluster rebalance 127.0.0.1 14459 pts/1Sl+0:00 ./kudu cluster rebalance 127.0.0.1 14514 pts/2S+ 0:00 grep kudu vc1320:~$ sudo kill -SIGPIPE 14459 * vc1320:~$ sudo -u kudu ./kudu cluster rebalance 127.0.0.1 > /tmp/kudu-rebalance.24.log 2>&1 vc1320:~$ echo $? 141 * -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin Gerrit-Reviewer: Adar Dembo Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins Gerrit-Comment-Date: Fri, 13 Jul 2018 20:55:37 + Gerrit-HasComments: No
[kudu-CR] spark: Add support for scanRequestTimeout to DefaultSource
Mike Percy has submitted this change and it was merged. ( http://gerrit.cloudera.org:8080/10725 ) Change subject: spark: Add support for scanRequestTimeout to DefaultSource .. spark: Add support for scanRequestTimeout to DefaultSource The timeout was manually tested via spark-shell and when set to a very low value (1ms) a combination or error messages and timeout warnings were printed due to Spark task retries. Added a simple verification test to ensure that parameter parsing works. Change-Id: I62b01af9d0532448fc3a6cf9328b3912a7865ebb Reviewed-on: http://gerrit.cloudera.org:8080/10725 Tested-by: Kudu Jenkins Reviewed-by: Dan Burkert --- M java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/DefaultSource.scala M java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduContext.scala M java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduRDD.scala M java/kudu-spark/src/test/scala/org/apache/kudu/spark/kudu/DefaultSourceTest.scala 4 files changed, 50 insertions(+), 7 deletions(-) Approvals: Kudu Jenkins: Verified Dan Burkert: Looks good to me, approved -- To view, visit http://gerrit.cloudera.org:8080/10725 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: merged Gerrit-Change-Id: I62b01af9d0532448fc3a6cf9328b3912a7865ebb Gerrit-Change-Number: 10725 Gerrit-PatchSet: 10 Gerrit-Owner: Mike Percy Gerrit-Reviewer: Dan Burkert Gerrit-Reviewer: Grant Henke Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy
[kudu-CR] spark: Expose socketReadTimeoutMs to Spark connector
Mike Percy has posted comments on this change. ( http://gerrit.cloudera.org:8080/10839 ) Change subject: spark: Expose socketReadTimeoutMs to Spark connector .. Patch Set 4: (1 comment) http://gerrit.cloudera.org:8080/#/c/10839/4/java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduContext.scala File java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduContext.scala: http://gerrit.cloudera.org:8080/#/c/10839/4/java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduContext.scala@22 PS4, Line 22: import com.google.common.annotations.VisibleForTesting > Remove this for same reason as in other review. Thanks again for the catch! -- To view, visit http://gerrit.cloudera.org:8080/10839 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I0ab0ff0b242790caffb7e2848958148ffe547c4d Gerrit-Change-Number: 10839 Gerrit-PatchSet: 4 Gerrit-Owner: Mike Percy Gerrit-Reviewer: Dan Burkert Gerrit-Reviewer: Grant Henke Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy Gerrit-Comment-Date: Fri, 13 Jul 2018 21:59:49 + Gerrit-HasComments: Yes
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Alexey Serbin has uploaded this change for review. ( http://gerrit.cloudera.org:8080/10938 Change subject: [tools] ignore SIGPIPE in kudu CLI .. [tools] ignore SIGPIPE in kudu CLI Since OpenSSL does not provide much control over underlying socket's options for SSL_write() and there is no analogous option for SO_NOSIGPIPE on Linux, let's ignore SIGPIPE in the kudu CLI. That helps to avoid random crashes of the tool when the server abruptly closes the connection. The issue with the kudu CLI tool exiting on SIGPIPE surfaced while running the TserverGoesDownDuringRebalancingTest scenario from kudu-admin-test. Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 --- M src/kudu/tools/tool_main.cc 1 file changed, 5 insertions(+), 0 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/38/10938/1 -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newchange Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin
[kudu-CR] WIP Add error handling to KuduBackup DONT BUILD KuduBackup will attempt to backup every table. If any one table fails the entire job will fail. All exceptions are logged when they happen.
Tony Foerster has posted comments on this change. ( http://gerrit.cloudera.org:8080/10941 ) Change subject: WIP Add error handling to KuduBackup DONT_BUILD KuduBackup will attempt to backup every table. If any one table fails the entire job will fail. All exceptions are logged when they happen. .. Patch Set 1: Ignore the 'KuduRestore' changes, that's even more half-baked than the KuduBackup (and not tested) -- To view, visit http://gerrit.cloudera.org:8080/10941 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I104fd604e12b70fcff9beace71ed4ef96b54d691 Gerrit-Change-Number: 10941 Gerrit-PatchSet: 1 Gerrit-Owner: Tony Foerster Gerrit-Reviewer: Grant Henke Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Tony Foerster Gerrit-Comment-Date: Fri, 13 Jul 2018 14:43:19 + Gerrit-HasComments: No
[kudu-CR] WIP Add error handling to KuduBackup DONT BUILD KuduBackup will attempt to backup every table. If any one table fails the entire job will fail. All exceptions are logged when they happen.
Tony Foerster has uploaded this change for review. ( http://gerrit.cloudera.org:8080/10941 Change subject: WIP Add error handling to KuduBackup DONT_BUILD KuduBackup will attempt to backup every table. If any one table fails the entire job will fail. All exceptions are logged when they happen. .. WIP Add error handling to KuduBackup DONT_BUILD KuduBackup will attempt to backup every table. If any one table fails the entire job will fail. All exceptions are logged when they happen. A backup attempt can have 3 results: - Success - table was backed up - Exists - table backup 'metadata' directory already existed. Assuming this was a retry. - Failure - table failed to backup. All data At the end of the job lists of failures, successes, and existing backups are printed out. A lot of this code is still assuming the 'batch' backup mode. I think a lot of this will change when we start doing the incremental backups. TODO - - if 'overwrite' flag is specified, overwrite the 'metadata' folder. Not sure yet how to handle this on retries or if it's even a good idea. Change-Id: I104fd604e12b70fcff9beace71ed4ef96b54d691 --- M java/kudu-backup/src/main/scala/org/apache/kudu/backup/KuduBackup.scala M java/kudu-backup/src/main/scala/org/apache/kudu/backup/KuduBackupOptions.scala M java/kudu-backup/src/main/scala/org/apache/kudu/backup/KuduRestore.scala M java/kudu-backup/src/test/scala/org/apache/kudu/backup/TestKuduBackup.scala 4 files changed, 208 insertions(+), 46 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/41/10941/1 -- To view, visit http://gerrit.cloudera.org:8080/10941 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newchange Gerrit-Change-Id: I104fd604e12b70fcff9beace71ed4ef96b54d691 Gerrit-Change-Number: 10941 Gerrit-PatchSet: 1 Gerrit-Owner: Tony Foerster
[kudu-CR] Change yetus dependency to compile in spark modules
Tony Foerster has posted comments on this change. ( http://gerrit.cloudera.org:8080/10927 ) Change subject: Change yetus dependency to compile in spark modules .. Patch Set 2: Code-Review+1 -- To view, visit http://gerrit.cloudera.org:8080/10927 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I16867a75b7f29f57c557449f94e52439cb9642c9 Gerrit-Change-Number: 10927 Gerrit-PatchSet: 2 Gerrit-Owner: Grant Henke Gerrit-Reviewer: Grant Henke Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy Gerrit-Reviewer: Tony Foerster Gerrit-Comment-Date: Fri, 13 Jul 2018 14:42:13 + Gerrit-HasComments: No
[kudu-CR] Change yetus dependency to compile in spark modules
Grant Henke has submitted this change and it was merged. ( http://gerrit.cloudera.org:8080/10927 ) Change subject: Change yetus dependency to compile in spark modules .. Change yetus dependency to compile in spark modules Due to Spark’s usage of reflection the Yetus annotation is required at runtime. Because it is getting in the way of other development and Yetus is a small dependency, I changed the scope to compile in the Spark modules to ensure the class is available. See KUDU-2500 for details and to track the final solution. Change-Id: I16867a75b7f29f57c557449f94e52439cb9642c9 Reviewed-on: http://gerrit.cloudera.org:8080/10927 Tested-by: Kudu Jenkins Reviewed-by: Mike Percy Reviewed-by: Tony Foerster --- M java/kudu-backup/build.gradle M java/kudu-backup/pom.xml M java/kudu-spark-tools/build.gradle M java/kudu-spark-tools/pom.xml M java/kudu-spark/build.gradle M java/kudu-spark/pom.xml 6 files changed, 24 insertions(+), 27 deletions(-) Approvals: Kudu Jenkins: Verified Mike Percy: Looks good to me, approved Tony Foerster: Looks good to me, but someone else must approve -- To view, visit http://gerrit.cloudera.org:8080/10927 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: merged Gerrit-Change-Id: I16867a75b7f29f57c557449f94e52439cb9642c9 Gerrit-Change-Number: 10927 Gerrit-PatchSet: 3 Gerrit-Owner: Grant Henke Gerrit-Reviewer: Grant Henke Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy Gerrit-Reviewer: Tony Foerster
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/10938 ) Change subject: [tools] ignore SIGPIPE in kudu CLI .. Patch Set 1: (1 comment) http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc File src/kudu/tools/tool_main.cc: http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc@268 PS1, Line 268: kudu::InitGoogleLoggingSafe(argv[0]); > I don't think so since the kudu CLI tool was exiting on SIGPIPE if during t However, I think it requires some clarification. I'll take a closer look. Thank you for pointing at IgnoreSigPipe() in InitGoogleLoggingSafe() -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin Gerrit-Reviewer: Adar Dembo Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins Gerrit-Comment-Date: Fri, 13 Jul 2018 20:26:30 + Gerrit-HasComments: Yes
[kudu-CR](branch-1.6.x) KUDU-2312: Scan predicate application ordering is non-deterministic
Dan Burkert has posted comments on this change. ( http://gerrit.cloudera.org:8080/10945 ) Change subject: KUDU-2312: Scan predicate application ordering is non-deterministic .. Patch Set 1: This applied cleanly except for the release_notes.adoc change, which I stripped. -- To view, visit http://gerrit.cloudera.org:8080/10945 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: branch-1.6.x Gerrit-MessageType: comment Gerrit-Change-Id: I99b2cabecd8626cad7e11fbdd492af7276e08348 Gerrit-Change-Number: 10945 Gerrit-PatchSet: 1 Gerrit-Owner: Dan Burkert Gerrit-Reviewer: Attila Bukor Gerrit-Reviewer: Dan Burkert Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon Gerrit-Comment-Date: Fri, 13 Jul 2018 20:47:31 + Gerrit-HasComments: No
[kudu-CR] spark: Expose socketReadTimeoutMs to Spark connector
Hello Dan Burkert, Kudu Jenkins, Grant Henke, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/10839 to look at the new patch set (#5). Change subject: spark: Expose socketReadTimeoutMs to Spark connector .. spark: Expose socketReadTimeoutMs to Spark connector This patch exposes socketReadTimeoutMs in the KuduContext and the DefaultSource. This patch also performs a bit of cleanup by renaming the KuduConnection object to KuduClientCache, which seems like a more appropriate name. Because socketReadTimeout is a KuduClient configuration parameter related to connection handling, socketReadTimeout was incorporated into the client cache key. Manually tested in spark-shell using spark-on-yarn. Added a basic test to ensure that the parameter is properly parsed by the DefaultSource and configured in the KuduRelation instance. Change-Id: I0ab0ff0b242790caffb7e2848958148ffe547c4d --- M java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/DefaultSource.scala M java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduContext.scala M java/kudu-spark/src/main/scala/org/apache/kudu/spark/kudu/KuduRDD.scala M java/kudu-spark/src/test/scala/org/apache/kudu/spark/kudu/DefaultSourceTest.scala M java/kudu-spark/src/test/scala/org/apache/kudu/spark/kudu/KuduContextTest.scala 5 files changed, 61 insertions(+), 27 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/39/10839/5 -- To view, visit http://gerrit.cloudera.org:8080/10839 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I0ab0ff0b242790caffb7e2848958148ffe547c4d Gerrit-Change-Number: 10839 Gerrit-PatchSet: 5 Gerrit-Owner: Mike Percy Gerrit-Reviewer: Dan Burkert Gerrit-Reviewer: Grant Henke Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/10938 ) Change subject: [tools] ignore SIGPIPE in kudu CLI .. Patch Set 1: (1 comment) http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc File src/kudu/tools/tool_main.cc: http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc@268 PS1, Line 268: kudu::InitGoogleLoggingSafe(argv[0]); Doesn't this already lead to an IgnoreSigPipe call? -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin Gerrit-Reviewer: Adar Dembo Gerrit-Reviewer: Kudu Jenkins Gerrit-Comment-Date: Fri, 13 Jul 2018 20:08:39 + Gerrit-HasComments: Yes
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/10938 ) Change subject: [tools] ignore SIGPIPE in kudu CLI .. Patch Set 1: (1 comment) http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc File src/kudu/tools/tool_main.cc: http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc@268 PS1, Line 268: kudu::InitGoogleLoggingSafe(argv[0]); > I don't think so since the kudu CLI tool was exiting on SIGPIPE if during t I phrased it as a question, but I'm sure it does: https://github.com/apache/kudu/blob/master/src/kudu/util/logging.cc#L272 Maybe something else is going on. Perhaps you can instrument the test to run under strace and see what sort of signal handling changes are being made? -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin Gerrit-Reviewer: Adar Dembo Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins Gerrit-Comment-Date: Fri, 13 Jul 2018 20:28:11 + Gerrit-HasComments: Yes
[kudu-CR] [tools] ignore SIGPIPE in kudu CLI
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/10938 ) Change subject: [tools] ignore SIGPIPE in kudu CLI .. Patch Set 1: (1 comment) http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc File src/kudu/tools/tool_main.cc: http://gerrit.cloudera.org:8080/#/c/10938/1/src/kudu/tools/tool_main.cc@268 PS1, Line 268: kudu::InitGoogleLoggingSafe(argv[0]); > Doesn't this already lead to an IgnoreSigPipe call? I don't think so since the kudu CLI tool was exiting on SIGPIPE if during that TserverGoesDownDuringRebalancingTest. Probably, logging was already initialized at that point? At least, after adding that IgnoreSigPipe() the issue with SIGPIPE during TserverGoesDownDuringRebalancingTest went away. -- To view, visit http://gerrit.cloudera.org:8080/10938 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I379376e7856a167b590f90c20f82402f2b825db9 Gerrit-Change-Number: 10938 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin Gerrit-Reviewer: Adar Dembo Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins Gerrit-Comment-Date: Fri, 13 Jul 2018 20:23:27 + Gerrit-HasComments: Yes
[kudu-CR](branch-1.6.x) KUDU-2312: Scan predicate application ordering is non-deterministic
Hello Attila Bukor, Kudu Jenkins, Todd Lipcon, I'd like you to do a code review. Please visit http://gerrit.cloudera.org:8080/10945 to review the following change. Change subject: KUDU-2312: Scan predicate application ordering is non-deterministic .. KUDU-2312: Scan predicate application ordering is non-deterministic This changes the scan predicate evaluation ordering so that it primarily orders by selectivity (as before), but breaks ties by column index. Change-Id: I99b2cabecd8626cad7e11fbdd492af7276e08348 Reviewed-on: http://gerrit.cloudera.org:8080/9440 Tested-by: Kudu Jenkins Reviewed-by: Todd Lipcon --- M src/kudu/common/generic_iterators-test.cc M src/kudu/common/generic_iterators.cc M src/kudu/common/generic_iterators.h M src/kudu/tablet/cfile_set.h 4 files changed, 111 insertions(+), 20 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/45/10945/1 -- To view, visit http://gerrit.cloudera.org:8080/10945 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: branch-1.6.x Gerrit-MessageType: newchange Gerrit-Change-Id: I99b2cabecd8626cad7e11fbdd492af7276e08348 Gerrit-Change-Number: 10945 Gerrit-PatchSet: 1 Gerrit-Owner: Dan Burkert Gerrit-Reviewer: Attila Bukor Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon
