set exec_flag in execTC_COMMITREQ so that api will get rollbackrep
ndb/include/ndbapi/NdbConnection.hpp:
Make getConnectedNode public
ndb/src/kernel/blocks/dbtc/DbtcMain.cpp:
Fix TC_COMMITREQ in abort cases
ndb/test/ndbapi/testNodeRestart.cpp:
Testcase for bug#8853
ndb/test/run-test/daily-devel-tests.txt:
Testcase for bug#8853
More explicit naming of variables
ndb/include/kernel/signaldata/AlterIndx.hpp:
Add union to indicate where master node is passed in ref
ndb/include/kernel/signaldata/BuildIndx.hpp:
Add union to indicate where master node is passed in ref
ndb/include/kernel/signaldata/CreateIndx.hpp:
Add union to indicate where master node is passed in ref
ndb/include/kernel/signaldata/CreateTrig.hpp:
Add union to indicate where master node is passed in ref
ndb/include/kernel/signaldata/DropIndx.hpp:
Add union to indicate where master node is passed in ref
ndb/src/kernel/blocks/dbdict/Dbdict.cpp:
Set error/master node also on BuildIndxRef
ndb/src/ndbapi/NdbDictionaryImpl.cpp:
Use masterNodeId instead of errornode
into mysql.com:/home/jonas/src/mysql-4.1
ndb/src/ndbapi/NdbDictionaryImpl.cpp:
Auto merged
ndb/src/ndbapi/ndb_cluster_connection.cpp:
Auto merged
sql/ha_ndbcluster.cc:
Auto merged
remove dict forwarding
add api retries on NotMaster
ndb/include/kernel/signaldata/AlterIndx.hpp:
Add NotMaster error code
ndb/include/kernel/signaldata/BuildIndx.hpp:
Add NotMaster error code
ndb/include/kernel/signaldata/CreateIndx.hpp:
Add NotMaster error code
ndb/include/kernel/signaldata/CreateTrig.hpp:
Add NotMaster error code
ndb/include/kernel/signaldata/DropIndx.hpp:
Add NotMaster error code
ndb/src/kernel/blocks/dbdict/Dbdict.cpp:
Never forward requests,
instead REF to API who will retry towards correct node
ndb/src/ndbapi/NdbDictionaryImpl.cpp:
1) Set error code for timeout
2) Handle NotMaster with retry in all DICT requests
ndb/src/ndbapi/ndb_cluster_connection.cpp:
Fix so that input values is in seconds and not 100ms's
sql/ha_ndbcluster.cc:
Wait 3 sec for all nodes to connect...
make sure transporter connections are close early in shutdown
ndb/src/common/logger/LogHandler.cpp:
removed printout "- Repeated 1 times"
ndb/src/kernel/main.cpp:
make sure transporter connections are close early in shutdown
ndb/src/kernel/vm/Emulator.cpp:
make sure transporter connections are close early in shutdown
ndb/src/kernel/vm/Emulator.hpp:
make sure transporter connections are close early in shutdown
(not to be merged as code has changed in 5.0)
ndb/src/mgmapi/LocalConfig.cpp:
Allow connecstring = "nodeid=X"
ndb/src/mgmsrv/MgmtSrvr.cpp:
remove incorrect asserts
mysql-test/r/ndb_index_ordered.result:
fix old decimal type mysql vs ndb (re-commit 3)
mysql-test/t/ndb_index_ordered.test:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/include/kernel/signaldata/DictTabInfo.hpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/include/ndbapi/NdbDictionary.hpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/include/util/NdbSqlUtil.hpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/src/common/util/NdbSqlUtil.cpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/src/ndbapi/NdbDictionary.cpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/src/ndbapi/NdbDictionaryImpl.cpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/src/ndbapi/NdbRecAttr.cpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/test/include/NdbSchemaOp.hpp:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/tools/restore/consumer.cpp:
fix old decimal type mysql vs ndb (re-commit 3)
sql/ha_ndbcluster.cc:
fix old decimal type mysql vs ndb (re-commit 3)
ndb/test/run-test/daily-basic-tests.txt:
run mysql-test-run for max 1 hour
(normally 30 minutes)
ndb/test/run-test/daily-devel-tests.txt:
Add benchmark to devel suite
ndb/test/run-test/main.cpp:
Add bench mode. always produce report
ndb/test/run-test/run-test.hpp:
Add bench mode. always produce report
Fix bug in bug fix in FastScheduler
remove possibility of endless loop
resulting in APZJobBuffer overflow due to statistics reporting
ndb/src/kernel/vm/FastScheduler.cpp:
Fix bug in bug fix in FastScheduler
remove possibility of endless loop
resulting in APZJobBuffer overflow due to statistics reporting
crash if scan is poped from queue before all attrinfo has arrived
ndb/src/kernel/blocks/dblqh/DblqhMain.cpp:
only restart scan if all attrinfo has arrived
otherwise just set remove from queue and set state
ndb/src/kernel/blocks/dbtc/DbtcMain.cpp:
new error insert - force send attrinfo
ndb/test/ndbapi/testScan.cpp:
new test bug8262 many threads - scanning small tables
ndb/test/run-test/daily-basic-tests.txt:
run new test case in autotest
corrected documentation on fragmentation
set "fragmentation medium" to mean 2 fragments per node instead of 1
set default fragmentation to small instead of medium
bug#8284 adjust fragmentation to max_rows
mysql-test/r/ndb_basic.result:
added test to trigger drifferent fragmentations in ndb
mysql-test/t/ndb_basic.test:
added test to trigger drifferent fragmentations in ndb
ndb/include/ndbapi/NdbDictionary.hpp:
corrected documentation on fragmentation
ndb/src/kernel/blocks/dbdih/DbdihMain.cpp:
set "fragmentation medium" to mean 2 fragments per node instead of 1
ndb/src/ndbapi/NdbDictionaryImpl.cpp:
set default fragmentation to small instead of medium
sql/ha_ndbcluster.cc:
bug#8284 adjust fragmentation to max_rows
more test program fixes
replace pNdb->closeTransaction with closeTransaction(pNdb)
ndb/test/src/HugoTransactions.cpp:
more test program fixes
replace pNdb->closeTransaction with closeTransaction(pNdb)
add abort to multi op test case
ndb/include/ndbapi/NdbConnection.hpp:
Add aborts to test case
ndb/test/include/HugoOperations.hpp:
Add aborts to test case
ndb/test/ndbapi/testOperations.cpp:
Add aborts to test case
ndb/test/src/HugoOperations.cpp:
Add aborts to test case
ndb/test/src/HugoTransactions.cpp:
Add aborts to test case
workaround for HPUX signal.h error, missing extern "C"
moved my_thread_end to NdbThreadExit
more checks for shared memory transporter signum setup
acinclude.m4:
backported configure flag from 5.0
include/my_global.h:
workaround for HPUX signal.h error, missing extern "C"
ndb/src/common/portlib/NdbThread.c:
moved my_thread_end to NdbThreadExit
ndb/src/mgmsrv/ConfigInfo.cpp:
more checks for shared memory transporter signum setup
1) New testcase
Check every combination of ins/upd/del of length 5
Check reading savepoint's
2) Fix 1 liner in acc wrt committing read
ndb/include/ndbapi/NdbConnection.hpp:
Make testcase friend
ndb/src/kernel/blocks/dbacc/DbaccMain.cpp:
Fix so that committing a READ can _not_ result in setting elementIsDisappeared
ndb/test/include/NDBT_Test.hpp:
Make copy of testcase name
ndb/test/ndbapi/testOperations.cpp:
New testcase
Check every combination of ins/upd/del of length 5
Check reading savepoint's
ndb/test/src/HugoOperations.cpp:
Close transaction in destructor
ndb/test/src/NDBT_Test.cpp:
Make copy of testcase name
Fixing this bugs improves performance by 40% for very small read
statements and with 12-13 % for very simple updating transactions
(flexBench -o 10000) in single threaded application. Removes a fixed
cost of around 100.000 cycles every time the ndbd process wakes up to
execute some queries.
ndb/src/kernel/vm/FastScheduler.cpp:
Integrate sendPacked into doJob
The lack of integration meant that several loops in
ipControlLoop were executed each time the ndbd process
woke up, also for reads it meant that response was divided
in two TCP/IP packets
Is necessary to integrate this with overload protection
ndb/src/kernel/vm/ThreadConfig.cpp:
Integrate sendPacked into doJob
The lack of integration meant that several loops in
ipControlLoop were executed each time the ndbd process
woke up, also for reads it meant that response was divided
in two TCP/IP packets
Is necessary to integrate this with overload protection
+ ion shutdown - don't close files that are not open
+ abort in debug - if closing a fd == -1
BitKeeper/etc/logging_ok:
Logging to logging@openlogging.org accepted
