Follow

Kinetica 7.0.9.0 Release Notes

Build version: 7.0.9.0

  • 164 Resolved Issues

FIXES: 

KDEVOPS-744 Added yum update and yum cleanup to make-gaia-docker.sh
KECO-1517 Fix test broken due to API now attempting to failover in offline mode
KGEO-582 Added a test for st_simplify/st_simplifypreservetopology
KGEO-665 Added an option parameter for the cutoff maximum number combinations to the /match/graph endpoint
KGEO-679 Changes to receive messages -especially errors- from the graph server
KGEO-682 Fix heatmap issue when the minimum value in the heatmap is >17M (the offset calculation suffered from floating-point problems and was zeroing out that data
KGEO-683 /solve/graph endpoint primary parameters are updated to eliminate the obscure way of assigning source nodes See the new schema
KGEO-685 st_buffer with spherical distance metric uses a local UTM transformation with much less skew than the previously used mercator projection
KGEO-686 Added a raster_cb wms option to enable/disable alpha-blending
KGEO-687 Fixed ordered class break crash with Intel build and a single class
KGEO-694 ShapeProcessor clear() -> clearMem(), use swap to clear std containers, make ShapeAtttributeBase virtual to fix clear()
KGEO-701 Grid functions (st_hexgrid/st_pointgrid/st_trianglegrid/st_squaregrid) can output up to 20 million cells (up from 1 million). Plus much faster grid generation since rank-0 doesn't need to convert the geometry to WKT strings
KINFRA-1085 gaiadb-template.spec check that /etc/init.d/gpudb_host... and /opt/gpudb/core/bin exist before checking if db is running before install
KINFRA-1085 GlobalManager log ERROR_NOBT, we don't need a bt in this code
KINFRA-1085 INFO log 'clearing all tables' instead of DEBUG
KINFRA-1095 all_test.sh -t #testers now runs serial tests too, use --parallel-only to have tests pass, used for crash testing
KINFRA-1095 Serialize don't access empty vect[0], skip writing values
KINFRA-1095 WorkerGetRecordsUtils ensure that num_records == memblock size, else throw instead of crashing (unknown what causes this)
KINFRA-1160 Add GPUDB_PROCESS_ID env var to bin/gpudb script to track child pids and 'kill' them, even when orphaned
KINFRA-1160 bin/gpudb also kill gpudb_proc user processes with GPUDB_PROCESS_ID env var
KINFRA-1431 Add 'system-limited-fatal' error message to all file related errors in SetIndexedDb
KINFRA-1431 Fix unit tests to work now that we don't always create non-existing directories on the fly
KINFRA-1431 Throw actual 'system-limited-fatal' error code vice generic Gaia error code Found a number of other places related to file i/o errors that needed the error code
KINFRA-1451 Use move iterator for evict_columns
KINFRA-1490 Add fsync parent dir in FileSync::deleteFile()
KINFRA-1490 Add fsync parent dir on delete files in indexedDB
KINFRA-1523 Change the 'long tom bulk add time' threshold from 5 to 60 when fsync_indexdb_immediate is true
KINFRA-1525 Add tracers to (v)ram tier mgr calls
KINFRA-1536 Change error message from alter table, when table has a numeric column name
KINFRA-1540 Add missing code for conf on_startup_script to actually run script
KINFRA-1544 Add override option for max number of table objects returned from /show/resource/statistics KINFRA-1497 Make /show/resource/statistics fully cancellable Add method to AbortableConditionSet to throw if aborted Make OrderedRWLock throw if AbortableConditionSet is aborted even if not waiting for lock
KINFRA-1544 Limit number of table tiered objects returned from /show/resource/statistics to 100
KINFRA-1568 Update executor dtor poll to start with 1ms
KINFRA-1573 Update TierObjRamPool to keep RamPool in sync during alterSize
KINFRA-1576 Fix segfault in union w/DEBUG enabled
KINFRA-1582 Fix auditing for /insert/records and /update/records to properly handle record data
KOLAP-1108 Shadow cubes implementation for filters
KOLAP-1582 Remove the word ERROR from Integer divide-by-zero log string
KOLAP-1600 Add support for * in /show/sql/proc
KOLAP-1600 Add support for periodic executable stored procedures
KOLAP-1600

Add support for SQL Stored Procedures -- Initial Commit

KOLAP-1600 Add support for stored procedures security
KOLAP-1600 Add support to /show/sql/proc to return all sql procedures info
KOLAP-1600 Bypass security checks only for endpoints which allow it
KOLAP-1600 Fix tests to drop procedure until procedures are blocked in show table response
KOLAP-1600 Return multiple procedure names in response
KOLAP-1600 Revert username changes
KOLAP-1607 Compilation fix gcc 4.8.5, map::insert(make_move_it(begin(obj)), make_move(end(obj))) the moved obj must be non-const, newer versions allow const
KOLAP-1607 Remove deprecated variables for flushing RocksDb
KOLAP-1607 Remove extra switch-break and fix indentation
KOLAP-1609 Add tests to verify chunk invalidation for multi-chunk outer-joins and primary-key-joins is done correclty
KOLAP-1633 Improve chunk skipping filter-plan reporting for unary and between filters 2nd suffix parameter now are the non-zero count chunks out of the plan
KOLAP-1676 Fix the ki_rank() and ki_tom() simplification bug in simplify_rpn_for_join function
KOLAP-1676 GaiaSet::get_tom returns tom_index rather than tom_id
KOLAP-1678 Implement filter rollback the same way create_join_table rollback is implemented that is don't clear the table while it is being refreshed
KOLAP-1687 Fixed a wrong result with a filter having IN clause
KOLAP-1712 Set source table ttl to -1 to make sure the append/records input does not expire
KOLAP-1720 Add tracers to catch WorkItemThreadPoolExecutor destructor taking 1 second to complete. nosetests -s benchmark_kolap1720.py:benchmark_concurrent_filter create tracer file that illustrates the problem
KOLAP-1720 Benchmark improvements to many chunk joins and get_records_by_column
KOLAP-1720 Eliminate unnecessary pinning of all-true/all-false stencils
KOLAP-1720 JoinedSet.cpp: Use subtask concurrency limit for all parallelization Avoid tier management when creating all true stencils Worker_between_filter Use subtask concurrency limit avoid tier management when creating all true stencils Use IndexedExecutor to gather up to 20 chunks into one work item
KOLAP-1720 not-on-shard chunk skipping improved to avoid parallel-skip-analysis when it is determmined the chunk is not on the shard JoinedSet::requires_full_refresh and requires_incremental_refresh always return false if chunk is not on the shard. benchmark_kolap1720.py:benchmark_concurrent_sharded_filter added
KOLAP-1721 Enhance show table output to print table properties using new syntax
KOLAP-1732 Changed chunk_cache_enabled to shadow_cube_enabled split up GaiaSetChunkCache resident set int agg and filter resident sets
KOLAP-1732 Fix testing error in test_chunk_skipping.py:test_chunk_skipping_on_unary_filters2
KOLAP-1733 Fix problem with restart on manual list partition On restart show_table is called to get table info The partition_keys vector was not being correctly set from the JSON until the partitions were actually cached The partition_keys_external were loaded from disk but partition_keys were not set Changed the mode of operation to always set partition_keys when loading from disk so that we can't have any problems Do some defensive programming to make sure we don't divide or mod by 0 even if the partition keys are empty
KOLAP-1736 Experimental time based partitioning of chunks for between filter tested but reverted to non-experimental method
KOLAP-1741 Only reserve the current chunks size for pk lookup table vectors
KOLAP-1742 FilterPlanner changes: avoid bitmask1d plan by not doing primary-key filter if an invalid bitmask 1d plan would be needed. Fix bug in left/right equi-join where sense of left/right was flipped
KOLAP-1743 Add RWLocks around all accesses to PKLookupTable::get{Indices,ChunkIndices}{,Mutable}
KOLAP-1743 Disable 'asof' join (temporarily) until it is working correctly
KOLAP-1743 Fix a deadlock with the PKLookupTable RW lock
KOLAP-1743 Fix RW access to PKLookupTable indices/chunk indices. Rename the functions to getPKIndices/getPKChunkIndices to enable easier searching. Use local 'pinning' objects that go out of scope before the RWLock to avoid pin errors
KOLAP-1743 Re-enable asof join no locking needed
KOLAP-1751 Support /update/records on bytes columns
KOLAP-1754 Add 'order_by' option on /get/records/bycolumn to make sure the output order uniqueness
KOLAP-1755 Do not do shard based chunk skipping if plans covering shard key have different sequence numbers (plans are part of or separated by different outer joins
KOLAP-1755 Fix minor error in Type::get_contained_shard_key_list_of_joined_type
KOLAP-1755 Fix test_chunk_skipping.py:test_chunk_skipping_sharded to skip if there is only 1 shard
KOLAP-1755 Implement shard based chunk-skipping
KOLAP-1761 Add test_join_kolap1761.py that iterates test 20 times to insure repeatability. Only works with multiple shards
KOLAP-1761 Worker_unary_join_filter.cpp uses BitmaskStencil allocated function rather than all_true and all_false functions to determine if input stencils need to be pinned. uses new FilterPlan null information variables to determine if null filtering can be done generate attribute-index rpn differently than unary filter rpn as skip_info.m_rpn not compatible with the evaluate_fast_lookup function JoinedSet::skip_chunk_for_comparision_operators now returns NULL_TREE if chunk has nulls. Will not do null filtering as a result future remove this change and have between filter also use the has_null_operator FilterPlanner.cpp, Parser.cpp now records in unary FilterPlan null information variables: m_null_forcing_ variables, m_hasa_null_operator, m_hasa_null_constant BitmaskStencil.cpp, GaiaBaseStencil.cpp allocated function GaiaBaseStencil.cpp only check pinning if stencil is allocated
KOLAP-1762 Fix for default parameter for is_rolemember()
KOLAP-1765 Add string type to types checked by test_chunk_skipping.py:test_chunk_skipping_sharded
KOLAP-1765 Fix shard key chunk skipping by making raw strings (DATA_TYPE_STRING) 'comparable' so they can be used in between filters. Note that we still reject expressions of the form: <raw_string> > 'value' since we don't have a HASH_GT_OP
KOLAP-1768 Guard against arg_min/arg_max on raw strings (was previously crashing). Also guard against it for geometry columns
KOLAP-1772 Add accelerated parallel-chunk-skip-index-analysis path that processes just the chunks returned by the index
KOLAP-1772 Chunk skip index file is deleted when the table is cleared
KOLAP-1772 Chunk skip index supports date/time/datetime/timestamp datatypes
KOLAP-1772 Fixed Missing endpoint def
KOLAP-1772 Initial implementation of a 'chunk skip' index. -Index is created via /alter/table with the option 'index_type' set to 'chunk_skip' (the option set to 'column' is the default and that is for our standard attribute indexes. -The index will be populated via column metadata when possible (i.e. which the column min/max values are identical so in that case the column data itself won't need to be read. -The index maintains a map at the GaiaSetData level of unique values to a list of chunk ids that contain that value (i.e. the value 5 is in chunks 0, 3 and 7). -At query time the filter planner will use the chunk skip index to satisfy queries of the form 'column = <value>'. -The index is only helpful when there are tons of chunks and when the speed of our standard chunk skipping becomes costly
KOLAP-1773 Sort get/records/by/coulmn resulsts
KOLAP-1778 Do not cache aggregate chunk results when the request refers to ki_partition()
KOLAP-1779 Fix /insert/records when an empty string is in a record along with an init-with-now column (and the record needs to be re-encoded)
KOLAP-1781 Fix vector overrun issue prior to saving predicate join stencil to cache
KOLAP-1784 RefreshViewProc::doRun now exits when a periodically refreshable view no longer has source tables (has become a base table)
KOLAP-1787 Add option for retaining partition information on create projection set the option to false by default so that we don't retain source table partitioning information by default
KOLAP-1801 LIKE operator accepts a third argument, with default escape char of '\'
KOLAP-1824 Add SQL support for chunk skip index
KOLAP-1825 Wrap synchronous worker execution into a utility function which throws exception if the worker fails
KOLAP-1827 GaiaSetBase::hasNonPartitionedFactSet
KOLAP-1827 JoinedSets and aggregate_group_by result sets are now not incrementally refreshed if input fact set is partitioned
KOLAP-1831 Fix bug that made non-partitioned GaiaSets not incrementally refreshable
KOLAP-1834 Fix showing shadow filter/aggregate sizes
KOLAP-1835 Make a combo cache entry for pk lookup table and bitmask stencil
KOLAP-684 Fixed returning info along with actual insertions and added some tests
KOLAP-684 Support new options in decode_objects
KUI-1002 Case Sensitivity with External Authentication
KUI-1002 Strip @ symbol from external users so they can be used to authenticate proxy
KUI-1006 Add "Member Of Roles" Selection Control to Create/Edit Role Page
KUI-1012 Add new UI for Query graph
KUI-1012 Fix input padding for Firefox
KUI-1012 Fix per record WMS preview modal
KUI-1012 Improve confirmation messages for create and query graph when no visualization
KUI-1013 Add new UI for Match graph
KUI-1013 Add support to visualize only certain solve methods
KUI-1013 Fix incorrect value being set
KUI-1013 Revert match visual restrictions
KUI-1014 Add missing showStartEnd to other places MapTableViewCtrl is opened
KUI-1014 Add multi-point picking and fix globe input overlap
KUI-1014 Add new UI for Solve graph
KUI-1014 Add point picking for solve graph
KUI-1014 Fix incorrect value being set
KUI-1014 Show start and end markers for solutions and do a better check if solution can be rendered
KUI-1015 Fix bug with identifying when WKT is present
KUI-1015 Reveal Map Slice to identify selected records
KUI-1016 Add support for WKT columns
KUI-1016 Reveal to Support Map Zoom/Pan on Data Browser Selection
KUI-1020 Remove and clean up embedded KAgent app
KUI-1022 Update KiSQL autocomplete SQL functions
KUI-1024 Allow user to select cluster when using KiSQL in HA environment
KUI-1024 Filter out empty entries for head nodes
KUI-1025 [GADMIN] Need a check box to skip header if using existing schema in "Advanced CSV Import"
KUI-1026 [GADMIN] Getting an error when trying to import a empty value into a decimal|nullable column
KUI-1030 Fix issue with refreshing not getting line widths
KUI-1030 Improve range text for clarity and fix legends overlapping layers
KUI-1030/1031 Add changes for both tickets
KUI-1032 Enhance table/graph WMS viewer to support record (edge/node) browsing and identification
KUI-1033 Add support for multiple classbreak values per row
KUI-1034 Data Browser slice not able to provide image thumbnail/preview for nullable bytes columns
KUI-1037 Add custom layer support in Map slice using external WMS (and possibly other sources)
KUI-1038 Add support for custom class break legend label
KUI-1038 Fix bug with labels not re-populating on slice designer view
KUI-1040 Demo data should import without skipping error to show users if error has occurred
KUI-1042 Fix truncation warning appearing when changing column name to number
KUI-1045 Improve handling when GPUdb is down
KUI-1045 Update login error-handling to check different exception thrown by Java API
KUI-1046 Update SQL parser in KiSQL to support stored procedure syntax
KUI-1048 Move logging filtering to client-side to mitigate command injection risk
KUI-1050 Block stats server proxy from revealing server version in response header (WSGIServer/0.1 Python/2.7.15)
KUI-1051 Block API endpoint proxy from returning * for Access-Control-Allow-Origin
KUI-1052 Add ability to "cancel" running query or queries in KiSQL
KUI-1059 Create embeddable/consumable WMS viewer in GAdmin to use in JupyterLab
KUI-1059 Removed token list endpoint used for debugging
KUI-1060 Add point picking to query endpoint
KUI-1063 Add support for session management
KUI-1063 Roll back commits for support for session management
KUI-1067 Halt on error option not working for SQL tool
KUI-1071 Update Table-level tier strategy validation in Create/alter table UIs
KUI-1072 Fix string formatting when column is string nullable
KUI-1072 Scatterplot incorrectly detects numeric columns if they are nullable
KUI-1075 Update SQL tool parser to fix bug for handling case statements in sql proc queries
KUI-1077 Support --s-batch-iterator-column-name flag for KIO
KUI-869 WKT class break should use CB_ATTR/CB_VALS as multi-attr params fallback may go away at some point


KNOWN ISSUES:

None currently

  


Should you have any questions or concerns, please visit our support page, official documentation page or email us at support@kinetica.com

 
Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

0 Comments

Article is closed for comments.