Clone Tools
Constraints: committers
Constraints: files
Constraints: dates
Merge "Possible fix for RS crash"

Fix for bug 1401012

The first CLI call initializes the cli_globals global structure.

There was a loop hole that can cause this structure to be accessed

from another thread while it is being initialized and this was

causing core files in multi-threaded testing using T2 driver.

Change-Id: I13651ed241d9a4dd5973d8ab260af5de65757efc

Merge "Reverting a change causing issues on a cluster"

Reverting a change causing issues on a cluster

Change-Id: Ib8bd62b8e95fdc9ffa7c05b4aed8b764bb9480dd

Improved error handling during transaction hasConflict checking

Change-Id: I990460d321bfd89c71cdfd76783b2b6c0768dbf0

Misc fixes related to LOB code

Fix to generator that accidentally removed a whole method earlier.

Fix to executor/TEST130 to cleanup and execute a delete

Fixing version info for mxlobsrvr process

Change-Id: Ifb724543299172c4f8d01a7040a74144f8e9e397

    • -0
    • +48
    • -27
    • +19
Merge "Add support for DCS and phoenix_tests to local_hadoop"

Possible fix for RS crash

Turning off block cache for regular full table scan and the coprocessor

count operation. This avoid these operations to add new block cache entries

in RS and evict the useful entries.

Ed atomic Testing has shown improvments in java memory usage in RS and in

throughput for large table full scans.

Change-Id: Ief1b9bb60c500100d3e63650b13cebfd99c2d041

Remove duplicates in PATH and CLASSPATH to fix bug 1401216

In sqf/ :

1. Removed duplicates from environment variables $CLASSPATH and $PATH,

so can be invoked repeatedly without them growing.

2. Removed test for JAVA_HOME containing exactly "1.7.0_67", which should

help fix bug 1387238.

3. To reduce the places the recommended jdk version "1.7.0_67" appears,

set this variable earlier:


This is used only in setting JAVA_HOME if it does not exist.

4. Removed jdk 1.6 lines since we stopped using that when directory

sql/hbaseExtensions was removed about the end of September.

5. Made indentation more consistent.

6. Made formatting of "if" and "do" more consistent.

7. Used double-brackets in conditionals, which is safer as

documented in

8. Use $(..) for subshells instead of less readable backticks.

In sqf/tools/, added function remove_duplicates_in_path which

removes second and later appearances of the same value.

Change-Id: Iec48874b14d08ca8155006ab6a6a3f41c1714ca8

Merge "Fixes and removal of obsolete code."

    • -126
    • +76
Miscellaneous minor changes

1) Revert a part of previous change to ComTdbHbaseAccess::displayRowId() for

SHOWPLAN. Each valid row id should have 2 bytes of length followed by the

value of id, if it is not missing or null. We need to treat the id as

binary value so we'll print each id byte as 2-digit hex value up to 20


2) Remove the part of comments in sql/sqlcomp/DefineConstants.h that

requires not to delete and reuse existing constants. This is no longer

mandatory in Trafodion.

3) Enable NA memory overflow checking when running executor and qat

suites in debug flavor. This would increase the run time for these two

suites by about 10%.

Rework according to review comments. Explanations above changed too.

Change-Id: I854ec776fa5fb815c39677d469f8f46f18d51402

    • -20
    • +11
    • -0
    • +5
Merge "Lookup object UID for Sequence generator"

Merge "Changes to first review comments from Ruo-Yu"

Lookup object UID for Sequence generator

The NATable object representing a Sequence generator sometimes has

an object UID of 0 at the time privileges are checked, resulting

in failure to locate the object and denial of access. This change

looks up the UID if the corresponding NATable member variable is 0.

Change-Id: I889d2d823b8c0faca5318572250b45e07134b6cf

Merge "Fix for bug 1400963"

Merge "Changes to delay region split if txn present"

Merge "Give warning, not error, when Hadoop not found"

Updated Phoenix tests to cope with ANSI schemas

Updated Phoenix tests to do 'create schema' and 'drop schema' before and after

each test to cope with the new behavior with ANSI schemas. Up to now, schemas

in Trafodion do not need to be created or dropped, but the ANSI schema support

will change that. This update for Phoenix tests need to go in first before

the implement of ANSI schemas can go in, to ensure a smooth transition for

Jenkins builds. Since 'create schema' and 'drop schema' are just no-op right

now, Phoenix tests will continue to work for the existing builds even with this


Change-Id: Ifa6902a122c78bdb899516cf8121295092cad0a9

Enable Phoenix T2 to work in local_hadoop environment

This will enable Phoenix T2 to work in developer's local_hadoop

environment. It makes the assumption that HBase/Hive lib

directory is at the same level as the HBase/Hive directory.

Change-Id: If2a2c84dca1d7cd8fc17da0ad56485c8bfdb9e74

Merge "TMUDF C++ compiler interface, part of log-reading TMUDF"

    • -0
    • +331
Fixes and removal of obsolete code.

-- LP 1400556 'get tables in schema' is not supported on external

hbase tables. An error is now returned.

-- LP 1400553 Insert into external hbase tables in _ROW_ format must use

column_create function and VALUES clause to create rows.

An error is returned otherwise.

-- a bug with that prevented a boundary case when sequence increment value

was one less than largeint max has been fixed.

-- error message to indicate what options can be used during alter sequence

has been updated

-- create table as select stmt now returns an error if running within a user

transaction. This is the same behavior as other DDL operations.

This will be

removed once we have transaction support for DDL stmts.

-- create table as select now uses non-transactional 'upsert using load' to

populate target table instead of transactional '' stmt.

-- hive/test020 has been enabled. This tests for access to ORC files.

-- obsolete sidetree insert and NVT user load code has been removed.

Change-Id: I14d321deaa52321777acd1d8ca55420f1e973367

    • -2800
    • +367
    • -369
    • +16
    • -126
    • +76
  1. … 17 more files in changeset.
Changes to delay region split if txn present

Region splits will be delayed until no more transactions

are in the active or pending list of the region. Split work

is being checked-in separately from the balance changes.

Change-Id: I800b06711d14f019d7f1a23d1c3d123e7aa805ff

Give warning, not error, when Hadoop not found

When Trafodion has just been untarred from a product file obtained from, and no Hadoop is found, give a warning instead of

an error and suggest that install_local_hadoop may be needed.

In that situation, do not put 4 colons at front of CLASSPATH.

Change-Id: If207b5999df37d3139015ff88bc4bde4686f7919

Fix node down processing and stale branch

Change-Id: Idb4e8df05c99a349118942725fefac70a8e1a318

Removing confusing typo in trafodion_mods

trafodion_mods stated it was only for cloudera.

Change-Id: Ic78e8f2022b0b96c883e20b1cde596ec2f386cfb

Fix for bug 1400963

OEW test with T2 multithread cored at NAList<CollationDB*>::removeCounted

User collation code had static member variable that was cauing this issue.

User collation feature is not supported and hence removed part of this featu

and reworked the rest to make it thread safe.

Change-Id: I4aec79ace3e331692549809cee81502a9102993d

Add support for DCS and phoenix_tests to local_hadoop

By default it gets the latest DCS from github, but env variables

can be used to over-ride and get DCS from another URL or local


Includes a convenience script to run phoenix tests. Phoenix tests

only works with T4 driver in local_hadoop environment so far.

Change-Id: I2574ab637bd7db23b42e4d7eb67afc9ad27c7bb6

    • -2
    • +169
SHOWPLAN related changes

Reworked the fix for bug 1392522 - mxosrvr core dumped doing showplan

(with N.E. enabled), plus other related code changes.

1) The native code (also known as native expression), if generated, is

stored in the expression's constant area. SHOWPLAN will dump the native

code in the assembly format. The display can be disabled by CQD

PCODE_NE_IN_SHOWPLAN to "OFF". It is "ON" by default. This part had been

reviewed by Jim Capps and Mike Hanlon.

2) Add several SHOWPLAN statements in core/TEST019 without logging the

output. This is to ensure no core generated when getting executor

operator (TDBs) info via SHOWPLAN.

3) A temporary fix to ComTdbHbaseAccess::displayRowId(). The current way

of retrieving begin or end row IDs from the HbaseScanRows for SHOWPLAN

does not match with the way those row IDs are generated (see

HbaseAccess::genListsOfRows), causing core dumps in some cases.

Change-Id: Id33e8c854452c03e15c4bd21f668f123180d9471

    • -26
    • +31
Merge "Avoiding Thread.sleep() when thread is exiting"

Merge "Indentation fix - recovery node down message"