This patch:
1. Properly processes situation when pm join result count is exceeded.
2. Adds session variable 'columnstore_max_pm_join_result_count` to control the limit.
Internal memory representation of MEDIUMINT datatype uses 24 bits. This is
true for both MariaDB server as well as ColumnStore. MCS plugin code uses
TypeHandlerSInt24 and TypeHandlerUInt24 classes to respectively convert the
binary representation of the signed and unsigned MEDIUMINT values passed by
the server to the plugin. The plugin then outputs the text representation
of these values into an open file descriptor which is piped to cpimport
for the final load into the MCS db files.
The TypeHandlerXInt24 classes were earlier incorrectly using
WriteBatchField::ColWriteBatchXInt32() functions which operate on a 4 byte
buffer. This resulted in incorrect parsing of MEDIUMINT values. As a fix,
we implement WriteBatchField::ColWriteBatchXInt24() functions which
correctly handle the 24 bit input buffer used for MEDIUMINT datatype.
For the following query:
select item from (
select item from (select a as item from t1) tt
union all
select item from (select a as item from t1) tt
) ttt;
There is an if predicate in buildSimpleColFromDerivedTable() that compares
the outermost query field name (ttt.item) to the returned column list of
the inner query (tt.item) when building the returned column list of the
outer most query. In the above query example, the inner query field name
is an alias set in the inner most query and is set to "`tt`.`item`",
while the outermost query field name is set to "item". The use of
backticks "`" in the inner query alias is causing the execution to
not enter the if block which creates the SimpleColumn for the outermost
query field name. As a fix, we strip off the backticks from the inner
query alias.
1. Input and output RowGroup's used in GROUP_CONCAT classes
are currently allocating a raw memory buffer of size equal
to the actual width of the string datatype. As an example,
for the following query:
SELECT col1, GROUP_CONCAT(col2) FROM t GROUP BY col1;
If col2 is a TEXT field with default width, the input
RowGroup containing the target rows to be concatenated will
assign 64kb of memory for every input row in the RowGroup.
This is wasteful as actual field values in real workloads
would be much smaller. We fix this by enabling the
RowGroup to use the StringStore when the RowGroup contains
long strings.
2. RowAggregation::initialize() allocates a memory buffer
for a NULL row. The size of this buffer is equal to the
row size for the output RowGroup. For the above scenario,
using the default group_concat_max_len (which is a server
variable that sets the maximum length of the GROUP_CONCAT string)
value of 1mb, the buffer size would be
(1mb + 64kb + some additional metadata). If the user sets
group_concat_max_len to a higher value, say 3gb, this buffer
size would be ~3gb. Now if the runtime initiates several
instances of RowAggregation, total memory consumption by
PrimProc could exceed the hardware memory limits causing the
OS OOM to kill the process. We fix this problem by again
enabling the StringStore for the NULL row allocation.
3. In the plugin code in buildAggregateColumn(), there is
an integer overflow when the server group_concat_max_len
variable (which is an uint32_t) is set to a value > INT32_MAX
(such as 3gb) and is assigned to
CalpontSystemCatalog::ColType::colWidth (which is an int32_t).
As a short term fix, we saturate the assigned value to colWidth
to INT32_MAX. Proper fix would be to upgrade
CalpontSystemCatalog::ColType::colWidth to an uint32_t.
Added logical transformation of the execplan::ParseTrees with the taking out the common factor in expression of the form "(A and B) or (A and C)" for the purposes of passing a TPCH 19 query.
Co-authored-by: Leonid Fedorov <leonid.fedorov@mariadb.com>
1. In TupleUnion::writeNull(), add the missing switch case for
wide decimal with 16bytes column width.
2. MCOL-5432 Disable complete/partial pushdown of UNION operation
if the query involves an ORDER BY or a LIMIT clause, until
MCOL-5222 is fixed. Also add MTR test cases for this.
When a UNION operation involving DECIMAL datatypes with scale and digits
before the decimal exceeds the currently supported maximum precision
of 38, we throw an error to the user:
"MCS-2060: Union operation exceeds maximum DECIMAL precision of 38".
This is until MCOL-5417 is implemented where ColumnStore will have
full parity with MariaDB server in terms of maximum supported DECIMAL
precision and scale of 65 and 38 digits respectively.
CSC default ctor was private b/c it must not allow to use CSC outside thread cache.
However there are some places in the plugin code that need a standalone syscat that
is cleaned up leaving the scope. The decision is to make the restriction mentioned
organizational rather than syntactical.
These seem to have all fallen out of a recent Boost update to 1.81 which
dropped some internal includes. All of these uses within columnstore
relied on these transitive includes, so explicitly include what we need
to fix build.
Signed-off-by: Sam James <sam@gentoo.org>
Analyze needs to be completed differently than a normal query. In server, when an ANALYZE is seen, it calls init_scan() immediatly followed by end_scan(). This leaves the sqlfrontendsession (ExeMgr) in a state where it expects to return rows. This patch fixes end_scan to clean this up via reads and writes to get everything back in synch.
ANALYZE should display the number of rows to be displayed if the query were run normally. We have that information available, but no way to return it. A modification to server side to ask for that in the handler is required.
This patch also includes a beautification of sqlfrontsessionthread.cpp since it looked bad. The important change is at line 774
if (!swallowRows)
which short circuits the actual return of data
* fix C API includes
ColumnStore used to include server's mysql.h
but link all tools with libmariadb.so
There's no guarantee that this would work, even with workarounds
it had in dbcon/mysql/sm.cpp
Fix:
* tools (linked with libmariadb.so) *must* include libmariadb's mysql.h
* as a hack prevent service_thd_timezone.h from being loaded into tools,
as it conflicts with libmariadb's mysql.h
* server plugin *must* include server's mysql.h
* also don't link every tool with libmariadb.so, link the helper library
(liblibmysqlclient.so) that actually needs it, tools use this
helper library, not libmariadb.so directly
* do *not* link ha_columnstore.so with libmariadb.so
this means some libraries have to be compiled twice -
for tools with libmariadb.so and for plugin, without.
* use system boost, if possible
boost 1.71.0 is what ubuntu focal has, so let's start with that version.
boost 1.77.0 is the first that supports c++20
* add dependency for generated header files errorids.h messageids.h
see 3edd51610
* bump the version
* MCOL-5322 This patch replaces boost::mutex with std::mutex b/c IMHO std::unique_lock::lock is
less troublesome comparing with the boost alternative
* MCOL-5310 This patch replaces move-assignment with copy-assignment to avoid memory corruption (#2661)
* Bump VERSION to 22.08.7-1
* MCOL-5306 Re-read the config (Columnstore.xml) file if it was updated.
The existing implementation of Config::makeConfig() factory method
was returning a possibly stale config to the caller, without checking
if the config file was updated since the last read. This bug triggered
a scenario as described in MCOL-5306 where after a failover in an MCS
cluster, the controllernode coordinates changed in the config file
after failover and the existing mariadbd process was still using the
old controllernode coordinates. This lead to failed network connection
between mariadbd and the new controllernode.
The change in this fix, however, is more generic and not just limited
to this above scenario.
* MCOL-5264 This patch replaces boost mutex locks with std analogs
boost::uniqie_lock dtor calls a fancy unlock logic that throws twice.
First if the mutex is 0 and second lock doesn't own the mutex.
The first condition failure causes unhandled exception for one of the clients
in DEC::writeToClient(). I was unable to find out why Linux can have a 0
mutex and replaced boost::mutex with std::mutex b/c stdlibc++ should
be more stable comparing with boost.
* MCOL-5311 Add timezone to jobList in subquerytransformer
TimeZone was uninitialized in this scenario and led to undefined behavior.
* patch_out_of_band
Some changes made to 10.6-enterprise make a build using the out-of-band method of compiling columnstore not work. Out-of band means the source for the engine is not in the storage subdir of server, but rather in a stand alone directory. This is used by developers for easier develop work. In the case of out-of-band, INSTALL_LAYOUT is false in CMakeLists.txt
* MCOL-5346 This patch forces TreeNode::getIntValue to use conversion for dict-based CHAR/VARCHAR and TEXT columns (#2657)
Co-authored-by: Roman Nozdrin <rnozdrin@mariadb.com>
* MCOL-5263 Add support to ROLLBACK when PP were restarted.
DMLProc starts ROLLBACK when SELECT part of UPDATE fails b/c EM facility in PP were restarted.
Unfortunately this ROLLBACK stuck if EM/PP are not yet available.
DMLProc must have a t/o with re-try doing ROLLBACK.
* MCOL-3561 This patch updates Connector code after MDEV-29988
* This commit applies the code style format
Co-authored-by: Sergei Golubchik <serg@mariadb.com>
Co-authored-by: Roman Nozdrin <rnozdrin@mariadb.com>
Co-authored-by: David.Hall <david.hall@mariadb.com>
Co-authored-by: Gagan Goel <gagan.nith@gmail.com>
Co-authored-by: Denis Khalikov <dennis.khalikov@gmail.com>
Disable check for correlated subqueries, basically those types of queries transforms
to join (aggr(table2), table1), table2) and post join scalar filter.
ColumnStore used to include server's mysql.h
but link all tools with libmariadb.so
There's no guarantee that this would work, even with workarounds
it had in dbcon/mysql/sm.cpp
Fix:
* tools (linked with libmariadb.so) *must* include libmariadb's mysql.h
* as a hack prevent service_thd_timezone.h from being loaded into tools,
as it conflicts with libmariadb's mysql.h
* server plugin *must* include server's mysql.h
* also don't link every tool with libmariadb.so, link the helper library
(liblibmysqlclient.so) that actually needs it, tools use this
helper library, not libmariadb.so directly
Exit early from the plugin execution of ALTER TABLE statements
on the replica nodes. This is to prevent re-execution of syscat
table population from the replica nodes which should only be
executed once by the primary node in a CS cluster setup.
The main CmakeLists.txt was using MY_CHECK_AND_SET_COMPILER_FLAG before the include. This works in-band with server because it was already included in server's CmakeLists.txt.
dbcon/mysql included curl as a build dependency. We don't build curl. It's a lib dependency. Not sure why it works in-band. One wouldn't think it should.
The following functions are created:
Create function JSON_VALID and test cases
Create function JSON_DEPTH and test cases
Create function JSON_LENGTH and test cases
Create function JSON_EQUALS and test cases
Create function JSON_NORMALIZE and test cases
Create function JSON_TYPE and test cases
Create function JSON_OBJECT and test cases
Create function JSON_ARRAY and test cases
Create function JSON_KEYS and test cases
Create function JSON_EXISTS and test cases
Create function JSON_QUOTE/JSON_UNQUOTE and test cases
Create function JSON_COMPACT/DETAILED/LOOSE and test cases
Create function JSON_MERGE and test cases
Create function JSON_MERGE_PATCH and test cases
Create function JSON_VALUE and test cases
Create function JSON_QUERY and test cases
Create function JSON_CONTAINS and test cases
Create function JSON_ARRAY_APPEND and test cases
Create function JSON_ARRAY_INSERT and test cases
Create function JSON_INSERT/REPLACE/SET and test cases
Create function JSON_REMOVE and test cases
Create function JSON_CONTAINS_PATH and test cases
Create function JSON_OVERLAPS and test cases
Create function JSON_EXTRACT and test cases
Create function JSON_SEARCH and test cases
Note:
Some functions output differs from MDB because session variables that affects functions output,e.g JSON_QUOTE/JSON_UNQUOTE
This depends on MCOL-5212
in aggregation code
The patch disables padding that forces hasher to calculate over the whole 2k buffer. This patch also moves hashing code
into the common place where it belongs.
* MCOL-5092 Ensure column width is correct for datatype
Change MODA return type to STRING
Modify MODA to handle every numeric type
* MCOL-5162 MODA to support char and varchar with collation support
Fixes to the aggregate bit functions
When we fixed the storage sign issue for MCOL-5092, it uncovered a problem in the bit aggregates (bit_and, bit_or and bit_xor). These aggregates should always return UBIGINT, but they relied on the type of the argument column, which gave bad results.
* MCOL-5074 CASE with IN and aggregate asserts
gwip-scsp wasn't set and buildPredicateItem() was called which assumes it is set. Added code to set properly in this case
Introduced UDF and stored prodecure.
usage:
set columnstore_s3_key='<s3_key>';
set columnstore_s3_secret='<s3_secret>';
set columnstore_s3_region='region';
and then use UDF
select columnstore_dataload("<tablename>", "<filename>", "<bucket>", "<db_name>");
for UDF db_name can be ommited, then current connection db will be used
or stored function
call calpontsys.columnstore_load_from_s3("<tablename>", "<filename>", "<bucket>", "<db_name>");
