Frank Knot

Forum Replies Created

Viewing 15 posts - 1 through 15 (of 356 total)
  • Author
    Posts
  • in reply to: General #8660
    Frank Knot
    Participant

    Welcome to the HVR Forum

    in reply to: Welcome to the HVR Forum #8661
    Frank Knot
    Participant

    We have setup this new forum.

    This forum is intended to allow us (HVR Software) to share detailed technical info with our experienced users.

    This will include hints, tips & best practices for using HVR and also background info about common problems or new features.

    Users are also welcome to post technical questions and/or answers to this maillist.

    This could be useful if the problem [appears] generic or to seek input from the wider HVR community.

    We will respond to such posts, but it’s not a replacement for HVR Technical Support.

    Most problems are best logged via tickets to support.hvr-software.com

    in reply to: Community Discussions #8884
    Frank Knot
    Participant

    Can SQL Server log-based capture be used when source database is in the Simple recovery model?

    SHORT ANSWER:

    It should be possible when using the old transaction log access method (“SP”). By default, however, HVR uses a newer t-log access method (“FN”) which requires that the source database is in the Full recovery model.

    FULL ANSWER:

    Currently, HVR supports two method of reading data from the SQL Server transaction log. Internally we call the new one “FN”, while the old one is called “SP”. HVR uses FN method by default.

    When using “FN” t-log access method, it is mandatory that source database is in the Full recovery model and a proper log archiving (log backup) procedure is setup. This is required to control transaction log retention- switching database to Full recovery model and making a full database backup ensures that transaction log will never be truncated off the data that has not been backed up (archived) yet. We then read the log backups if the data we need is no longer available in the current (active) transaction log file.

    According to some online sources, SQL Server native transactional replication works with database that’s in either replication model (including the Simple model), so I guess our old t-log access method (“SP”) should work with database in Simple recovery model as well (though we have never tested this).

    It may also be possible to implement an alternative log-retention (or, rather, log-release) logic for the “FN” by calling sp_repldone manually, rather then relying on the SQL Server Agent job. It may allow using “FN” method to capture changes from database in Simple recovery model. This approach has its own drawbacks, though.

    in reply to: Community Discussions #8886
    Frank Knot
    Participant

    HVR log-based capture from DB2 LUW tables can give a NULL values for out-of-row LOB datatypes.

    Root cause

    HVR uses IBM’s DB2ReadLog API for log based capture (LBC). This API can be used in unfiltered or filtered mode.

    DB2LOB

    In unfiltered mode, DB2ReadLog API returns all log records, many of them internal to DB2 and of no interest to HVR. A major limitation of this mode is that column data is not uncompressed.

    In filtered mode, DB2ReadLog API returns only the log records marked as propagatable, by which IBM means log records that would be of interest to replication solutions such as HVR. Filtered mode also decompresses column data.

    Starting with versions 4.7.1/29 and 4.7.3/3, HVR defaults to filtered mode (HVR-785).

    This change was motivated by the realization that DB2 compression is commonly used, yet HVR would only support compression if the environment variable ZIZ_READ_FILTER was set to 1.

    What are my possible recourses?

    Upgrade to DB2 10.1 or higher

    Do not use DB2 compression in DB2 versions prior to 10.1, and add an environment action with /Name=ZIZ_READ_FILTER and /Value=0 to your channel.

    Move all tables containing out-of-row data to a new channel, disable compression for these tables, and add an environment action with /Name=ZIZ_READ_FILTER and /Value=0 to this channel. This is not an option if on the integration side there exist referential constraints between tables captured in separate channels.

    If none of the above options is suitable, please contact HVR Technical Support.

    in reply to: Community Discussions #8888
    Steve Sheil
    Participant

    Background (for all DBMS)

    HVR typically ‘streams’ data into its target databases; the data goes straight over the network pipe into the DBMS’s bulk loading interface (e.g. Oracle direct-path-load) without touching disk.

    We dislike putting data into a temporary file before loading data (this is called ‘staging”), but this is necessary for some DBMS’s that lack a high speed streaming interface.

    HVR bulk refresh streams data straight over the network into the target table via its bulk-load interface. If bulk-load streaming is not available then a staging file is used.

    HVR row-wise compare and refresh streams data straight over the network into row-wise ‘diff’ logic (no table bulk-load). The fact that HVR does this diff without any staging is quite elegant.

    HVR integrate with /Burst streams data straight over the network into the ‘burst’ table via bulk-load interface. If bulk-load streaming is not available then a staging file is used. Afterwards burst uses a ‘set-wise’ insert, update or delete to apply these changes to the target table.

    HVR integrate without /Burst (we call this ‘trickle’ internally) streams data straight over the network into HVR’s row-wise ‘integrate’ logic (no table bulk-load)

    ODBC, LIBPQ and Bulk-load into PostgreSQL

    In the past HVR used ODBC to talk to PostresSQL, but in 4.7.3 we switched to using LIBPQ interface.

    One reason that we switched to LIBPQ is because we wanted to use (abuse!) PostreSQL’s “copy (c1, c2, c3) from stdin” statement.
    This allows HVR to stream data into PostgreSQL at maximum speed, without having to write data to a temporary file (staging ).
    Another reason to switch from ODBC to LIBPQ is that we are building log-based capture from PostgreSQL and that requires LIBPQ too.

    Work on building log-based capture from PostgreSQL is proceeding now.

    Bulk-load into Greenplum

    HVR uses ODBC to talk to Greenplum.

    Greenplum does not have a suitable (e.g. sufficiently fast) ‘streaming’ bulk-load interface, so does HVR needs to do staging (load from a temp file) instead.

    In the past (early HVR 4.7.1 versions) HVR did this staging-bulk-load by writing into a file and then calling ‘copy (c1, c2) from ‘, but this had disadvantageous;

    (a) HVR had to be installed on the Greenplum head node machine.

    (b) HVR needed special privileges for this ‘copy’ interface (a stored procedure could be used to avoid this privilege).

    (c) It was slower.

    So HVR has now dropped this interface.

    HVR for Greenplum now uses the ‘gpfdist’ interface.

    This requires action LocationProperties with parameters /StagingDirectoryHvr and /StagingDirectoryDb (see https://www.hvr-software.com/wiki/locationproperties).

    HVR does bulk-load by writing the data into a temporary file in a staging directory on the machine where the HVR process which is connected to GPDB is running. This directory does not have to be on the GPDB machine; typically it’s not.

    This temp-file is written in ‘TEXT’ format and is compressed.

    Then HVR tells GPDB to bulk-load the data in this file by doing SQL ‘insert into as select * from ‘. This requires that a special Greenplum ‘external table’ exists for each target table which HVR loads data into. HVR will create these when it needs them and it should reuse them afterwards. These external tables are named ‘__x’ or ‘__bx’. They don’t really contain data; instead they are created with a URL which tells them to load data from a URL which points to the staging directory which HVR just filled with new data. The URL used (supplied with /StagingDirectoryDb) should start with ‘gpfdist://’. This requires that the user is running a Greenplum ‘gpfdist’ daemon on the machine that HVR is using; the GPDB nodes then connect to this gpfdist daemon and use it as a file-server to pull the compressed data into the target table.

    Bulk-load into Redshift

    Redshift is a postgres-based distributed DBMS that is hosted in Amazon’s cloud.

    HVR uses ODBC to talk to Redshift.

    Redshift does not have a suitable (e.g. sufficiently fast) ‘streaming’ bulk-load interface, so does HVR need to do staging (loading from temp files).

    For Redshift HVR these staging files are located in an ‘S3’ directory (S3 is Amazon’s big-data file system).
    This requires action LocationProperties with parameters /StagingDirectoryHvr, /StagingDirectoryDb and /StagingDirectoryCredentials (see https://www.hvr-software.com/wiki/locationproperties).

    HVR does bulk-load by first writing the data into temporary S3 staging files.

    This S3 temp-file is written in a form of CSV format (currently it is not compressed).

    Then HVR uses Redshift SQL ‘copy from s3://’. The Redshift data nodes contact S3 to load the data from those temp files.
    No ‘external tables’ are necessary.

    We recommend that the HVR process which is talking to Redshift and S3 is running inside the Amazon AWS, so that the network performance is optimal.

    Regards, Steve

    in reply to: Announcements #8687
    Steve Sheil
    Participant

    We have now release HVR 4.7.3/7 as GA (General Availability).
    Everyone with a validated account on http://www.hvr-software.com can now dowload this.

    The last GA release was HVR 4.7.1, so there are lots of new features.

    Below are some, extracted from the release notes.

    Regards,
    Steve Sheil

    ———————————————–
    NEW FEATURES IN HVR 4.7.3
    ———————————————–

    [HVR-882] S3 SUPPORT AS STAGING LOCATION FOR REDSHIFT BULK LOAD:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-848] SUPPORT XTREMEDATA
    To use this feature, upgrade HVR on the integrate machine(s)..

    [HVR-837] POSTGRESQL: USE LIBPQ INSTEAD OF ODBC CONNECTION:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-333] SUPPORT FOR DB2 ON WINDOWS:
    To use this feature, upgrade HVR on the hub, capture and integrate machine(s).

    [HVR-794] ADD SUPPORT FOR DB2 10.5 FOR LINUX 64 BIT:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-752] ADDED HVR COMPARE & REFRESH MEASUREMENTS TO HVR STATISTICS:
    To use this feature, upgrade HVR on the hub, capture and integrate machine(s).

    [HVR-744] ADDED SUPPORT FOR SQL SERVER 2005 LOG-BASED CAPTURE:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-759] SUPPORT FOR: /TIMEKEY, /INTEGRATEEXPRESSION={HVR_OP}, /RENAMEEXPRESION/CONTEXT:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-657] SUPPORT POSTGRESQL HUB:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-572] DB2 (LUW): ADDED SUPPORT FOR DATATYPES LONG VARCHAR AND LONG VARGRAPHIC:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-625] DB2 FOR I: SUPPORT FOR ENCODINGS EBCDIC 273 AND EBCDIC 1141:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-630] FASTER BURST BY BUNDLING UNCHANGED INTO SINGLE CHANGE:
    To use this feature, upgrade HVR on the integrate and hub machine(s).

    [HVR-618] REDSHIFT SUPPORTS STAGING BULK LOAD:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-592] DB2 FOR I: SUPPORT PHYSICAL FILES (DDS):
    To use this feature, upgrade HVR on all machine(s).

    [HVR-574] GREENPLUM BULK LOAD USES GPFDIST PROTOCOL:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-587] POSTGRESQL 8 SUPPORT:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-527] TERADATA LOB SUPPORT:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-504] ADDED SUPPORT FOR BULK LOAD INTO DB2 LUW:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-221] DB2 FOR LUW: LOG BASED CAPTURE OF LOBS:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-534] DATABASE TO FILE REFRESH USES TEMPORARY FILE FOR WRITING:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-270] MADE ALL CATALOG DATA UNICODE BASED:
    To use this feature, upgrade HVR on the gui and hub machine(s).

    [HVR-519] SUPPORT DB2 FOR I NAMED DATABASES ON OTHER AUXILIARY STORAGE POOLS:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-467] ADD SUPPORT FOR {HVR_SCHEMA} IN FILEINTEGRATE /RENAMEEXPRESSION:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-300] ADDED TRANSFORM /ROWS2CSV FOR MORE EFFICIENT DB->FILE CSV REPLICATION:
    To use this feature, upgrade HVR on the hub, integrate and gui machine(s).

    [HVR-466] SEPARATE FILEINTEGRATE /RENAMEEXPRESSION USING /CONTEXT:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-337] SUPPORT LOG BASED CAPTURE FOR DB2 FOR I:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-384] ADD SUPPORT FOR POSTGRESQL AS TARGET:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-334] ADDED DB2 FOR I FOR LINUX:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-303] IMPROVED WRITE SPEED FOR NON-LOCAL (E.G. FTP) FILE LOCATIONS:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-398] INCREASED INTEGRATE PERFORMANCE WITH NOW() EXPRESSIONS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-391] MSSQL LBC: ALLOW USE OF CDC TABLE TO ENABLE SUPPLEMENTAL LOGGING:
    To use this feature, upgrade HVR on the hub and capture machine(s).

    in reply to: Announcements #8685
    Simon Faas
    Keymaster

    HVR 4.7.3/25 has been released as GA on Jan 27th 2016.

    Below is an extract from the release notes containing a list of new features and problems fixed since the last GA release (4.7.3/7)

    ROBLEMS FIXED IN HVR 4.7.3/25 (2016-01-22)
    ——————————————-

    [HVR-1163] F_JD0A1F DURING CAPTURE FROM COMPRESSED TABLES ON SQL SERVER:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1156] FIXED F_JD21AD IN VERBOSE COMPARE WITH SOFTDELETE:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1154] FIXED F_JR0206 WHEN DBINTEGRATE/COALESCE IS DEFINED:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1155] FIXED F_JT1477 WITH /BURST AND /DUPLICATEROWS:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1137] FIXED STRING-BASED INTEGRATED EXPRESSIONS BEING APPLIED TWICE DURING ROW-WISE REFRESH:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1168] JD0A41 DURING CAPTURE OF NCHAR OR NVARCHAR DATA FROM COMPRESSED TABLE ON SQL SERVER:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1160] ORACLE CAPTURE FAILED WITH F_JZ193E: PARSER ENCOUNTERED UNKNOWN OP FLAG WHILE PARSING RECORD:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/24 (2016-01-11)
    —————————————–

    [HVR-1139] CLIENT-SIDE SSL CERTIFICATES FOR FTPS AND WEBDAV:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/24 (2016-01-11)
    ——————————————-

    [HVR-1126] FIXED F_JD2146 DURING VERBOSE COMPARE WITH EXPRESSIONS:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1135] FIXED F_JR0349: ILLEGAL OPTIONS IN REFRESH ON AIX:
    To fix this bug, upgrade HVR on the hub machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/23 (2016-01-07)
    ——————————————-

    [HVR-1129] FIXED DUPLICATE KEY VIOLATES UNIQUE CONSTRAINT FOR GREENPLUM BURST TABLES:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1078] FIXED F_JD20F2 ‘CONNECTION BUSY’ ON SQL SERVER VERBOSE COMPARE:
    To fix this bug, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 4.7.3/22 (2015-12-31)
    —————————————–

    [HVR-1109] ADD HVR_TABLE_CREATE_CHAR_MULTIPLIER TO OVERRIDE LENGTH IN TABLE CLONE:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1069] ADD DURATION AND SPEED MEASUREMENTS TO HVRSTATISTICS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1098] IMPROVED DB2 FOR I CAPTURE SPEED:
    To use this feature, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/22 (2015-12-31)
    ——————————————-

    [HVR-1066] ELIMINATE UNNECESSARY AUGMENTING DURING CAPTURE FROM COMPRESSED TABLES ON SQL SERVER:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1114] FIXED F_JT148B: PIPE ROWAUGMENT ENCOUNTERED ILLEGALLY LONG VALUE FOR ‘HVR_TX_SEQ’:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1086] FIXED F_JD22AF PRIMARY KEY AND DISTRIBUTED BY DEFINITIONS INCOMPATIBLE FOR GREENPLUM:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1108] FIXED F_JT0527 WITH DB2 FOR I LOG-BASED CAPTURE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1103] FIXED INCORRECTLY IRREGULAR OUTPUT IN HVRSTATISTICS:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1091] FIXED MULTI-DELETE DURING COALESCE RESULTING IN DUPLICATES:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1090] FIXED HVRREFRESH FAILS WITH F_JG200A IN SELECT ON HVR CATALOGS:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1089] OCCATIONAL F_JT0406 DURING SQL SERVER LOG-BASED CAPTURE WHEN READING FROM SEVERAL BACKUP FILES IN A ROW:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1044] ORACLE CAPTURE FAILED WITH F_JZ1917: PARTIAL CHAINED ROW SHARES THE SAME HRID WITH NEWLY ARRIVED CHANGE BUT THEIR ZOP TYPES ARE DIFFERENT:
    To fix this bug, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/21 (2015-12-17)
    ——————————————-

    [HVR-1084] FIXED F_JR0290 TIMESTAMP PRECISION 12 OUT OF RANGE WHEN ADDING TABLE TO CATALOG:
    To fix this bug, upgrade HVR on the hub and gui machine(s).

    [HVR-1083] FIXED F_JR0337: DBINTEGRATE /NOTRIGGERFIRING IS DEFINED:
    To fix this bug, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 4.7.3/20 (2015-12-14)
    —————————————–

    [HVR-967] ADD OPTION -DISABLE AND -CLEAR_PAST_ERRORS TO HVRMAINT:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1074] FILE INTEGRATE XML SERIALIZATION WITH COMPACT XML:
    To use this feature, upgrade HVR on the hub and gui machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/20 (2015-12-14)
    ——————————————-

    [HVR-1073] FIXED HVRMAINT REPORTING 0 INTEGRATE CHANGES IF THERE WERE CHANGES:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1068] FIXED MISSING LATENCY INFORMATION PER TABLE WITH BURST INTEGRATE:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1052] ORACLE CAPTURE FAILED WITH F_JZ1107: COLUMN HAS BEEN MAPPED TWICE INTO COLUMNS RECORDED IN ORACLE QUICK MULTIPLE INSERT/DELETE (11.11) REDO RECORD:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/19 (2015-12-02)
    —————————————–

    [HVR-1062] ALLOW MULTIPLE DBSEQUENCE /SCHEMA ACTIONS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1049] CHANGE ISOLATION LEVEL FOR DB2I AND SQL SERVER TO READ COMMITTED:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-1048] ELIMITATE THE NEED IN THE STATE TABLE ON CAPTURE SIDE FOR SQL SERVER:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1057] HVR_ISOLATION_LEVEL TO OVERRIDE DEFAULT ISOLATION LEVEL:
    To use this feature, upgrade HVR on all machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/19 (2015-12-02)
    ——————————————-

    [HVR-1059] FIXED TABLE SELECT DISPLAYS GREENPLUM EXTERNAL TABLES:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1036] ORACLE CAPTURE INCORRECTLY CAPTURES NULL VALUE FOR CLOB COLUMN:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1047] SQL SERVER CAPTURE MAY FAIL WITH F_JG200A OR F_JD0A1F IF COMPRESSED TABLE HAS LOB OR XML COLUMNS:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/18 (2015-11-26)
    —————————————–

    [HVR-1040] VARIOUS SFTP COMPATIBILITY ENHANCEMENTS:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/18 (2015-11-26)
    ——————————————-

    [HVR-1041] FIXED F_JD2313 REDSHIFT BULK LOAD FAILS WHEN THERE ARE OTHER TABLES PREFIXED WITH THE TBL NAME:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1045] FIXED HVRLOAD -OC (CHANGE TABLE) OPTION DISABLED IN HVRGUI FOR /BURST:
    To fix this bug, upgrade HVR on the gui machine(s).

    [HVR-1043] FIXED REDSHIFT STAGING FILES ARE NOT CLEANED UP WHEN STAGINGEXEC ERROR OCCURS:
    To fix this bug, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/17 (2015-11-19)
    —————————————–

    [HVR-643] ORACLE ADD HVR_ENROLL_VLOGFILE_RESTRICT TO SPECIFY REDO DISKGROUP:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1025] ADDED FILEINTEGRATE /ORDERBYTABLE TO IMPROVE {HVR_TBL_NAME} RENAME EXPRESSION FILE GENERATION:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/17 (2015-11-19)
    ——————————————-

    [HVR-1022] /AGENT INVOCATION GIVES F_JG2142 PROTOTYPE ERROR:
    To fix this bug, upgrade HVR on the capture, hub and integrate machine(s).

    [HVR-1016] FIXED GREENPLUM NUMERIC/DECIMAL ZERO PRECISION ERROR WHEN DATATYPEMATCH IS SET:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1004] FIXED HVR FAILS WITH F_JG290D DURING BULK LOAD OF BIG TABLES INTO REDSHIFT:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1026] FIXED JT152D WITH FILEINTEGRATE /BURST:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-991] FIXED MISSING BEFORE UPDATES (HVR_OP=3 OR 4) FOR INTEGRATE WITH /TIMEKEY:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1027] ORACLE LBC F_JZ196A – NULL BITMAP LENGTH ERROR IN UPDATE:
    To fix this bug, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/16 (2015-11-12)
    ——————————————-

    [HVR-1019] FIXED BURST INTEGRATE INTO GREENPLUM ALWAYS RECREATE THE EXTERNAL TABLE:
    To fix this bug, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/15 (2015-11-10)
    —————————————–

    [HVR-1009] IMPROVED ORACLE RAC CLUSTER REDIRECTS:
    To use this feature, upgrade HVR on the hub machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/15 (2015-11-10)
    ——————————————-

    [HVR-995] HVR LOADS INTO BURST TABLES IN REDSHIFT DISABLES AUTOMATIC COMPRESSION:
    To fix this bug, upgrade HVR on the integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/14 (2015-11-03)
    ——————————————-

    [HVR-993] FIXED GREENPLUM SQLPREPARE HANGS AFTER DATABASE RESTART:
    To fix this bug, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/13 (2015-10-30)
    —————————————–

    [HVR-977] ADD HVR_ENROLL_VLOG_RESTRICT AS WORK-AROUND FOR ORACLE F_JD2822 [HVR-976]:
    To use this feature, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/13 (2015-10-30)
    ——————————————-

    [HVR-974] FIXED UNBOUNDED HEAP GROWTH ON SOLARIS:
    To fix this bug, upgrade HVR on all machine(s).

    [HVR-985] ORACLE CAPTURE FAILED WITH F_JZ1547: READ OVERFLOW IN EXTRACTING COLUMN DATA IN CHANGE TYPE 11.11 REDO QMI (UNPACKED CASE):
    To fix this bug, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/12 (2015-10-23)
    ——————————————-

    [HVR-963] F_JZ1976. PARSER ENCOUNTERS EMPTY RECORD PIECE DURING RECORD CHAINING:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/11 (2015-10-20)
    —————————————–

    [HVR-964] RE-IMPLEMENTATION OF SFTP DIRECTLY ON TOP OF LIBSSH2:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/11 (2015-10-20)
    ——————————————-

    [HVR-929] ORACLE F_JZ1919 CHAINER INSERT AND ROW BOTH IN START STATE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-966] FIXED F_JT0472 DURING CAPTURE FROM HEAP TABLE ON SQL SERVER:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-926] FIXED ORA-00972: IDENTIFIER IS TOO LONG DURING HVR COMPARE EVEN WHEN THE COLUMN IS RENAMED:
    To fix this bug, upgrade HVR on the integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/10 (2015-10-15)
    ——————————————-

    [HVR-929] ORACLE F_JZ1919 CHAINER INSERT AND ROW BOTH IN START STATE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-959] FIXED HVR DOESN’T DETECT FEW INVALID UNICODE CODEPOINTS:
    To fix this bug, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/9 (2015-10-13)
    —————————————-

    [HVR-914] ADD OPTION -D TO HVRADAPT TO REMOVE TABLES FROM CHANNEL:
    To use this feature, upgrade HVR on the hub machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/9 (2015-10-13)
    ——————————————

    [HVR-942] DB2 LBC MAY FAIL WITH F_JD0701 IF VALUE COMPRESSION IS ENABLED:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-945] FIXED DB2 (LUW) RECAPTURING OF TRUNCATE TABLE STATEMENTS:
    To fix this bug, upgrade HVR on the capture and integrate machine(s).

    [HVR-953] FIXED F_JG200A WITH SQLSERVER LOGBASED CAPTURE WITH PARTITIONED TABLES WITH COMPUTED COLUMNS:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-931] FIXED HVR REFRESH FAILS WITH F_JT029F ERROR WHEN RESTARTING JOBS IN BI-DIRECTIONAL CHANNEL:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-958] FIXED S3 UPLOAD FAILS IF IT TAKE TOO LONG TIME:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-957] FIXED GUI-HUB COMPATABILITY ISSUE INTRODUCED WITH HVR-921:
    To fix this bug, upgrade HVR on the gui machine(s).

    [HVR-916] HVR CAPTURES 2 CHARACTERS WHEN THERE IS ONLY ONE CHARACTER FOR ORACLE (CHAINING):
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-955] IMPROVED SCANNED TRANSACTION MESSAGE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-934] ORACLE LOG-BASED CAPTURE FAILED WITH F_JZ1547: READ OVERFLOW IN EXTRACTING COLUMN DATA IN CHANGE TYPE 11.11 REDO QMI:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/8 (2015-10-01)
    —————————————-

    [HVR-912] ADDED $HVR_SF_TIMEOUT_SECS FOR SALEFORCE DATALOADER TUNING:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-915] ADDED /INTEGRATEEXPRESSION={HVR_INTEG_KEY} AS CHAR(16) FOR UNIQUE KEY GENERATION:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-917] ADDED FILECAPTURE /IGNORESIZECHANGES:
    To use this feature, upgrade HVR on the capture and hub machine(s).

    [HVR-923] ADDED HVR_IDENTITY_RESEED_ENABLE TO RESEED SQL SERVER IDENTITY COLUMNS:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-896] ORACLE REDO BACKWARD COMPATIBILE VERSION HANDLING:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-921] SUPPORT ADVANCING OF SQL SERVER TRANSACTION LOG TRUNCATION POINT:
    To use this feature, upgrade HVR on the hub and capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/8 (2015-10-01)
    ——————————————

    [HVR-918] CHANGED DEFAULT ISOLATION TO READ COMMITTED INSTEAD OF SERIALIZABLE FOR GREENPLUM, POSTGRES AND DB2:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-922] FIXED F_JD0A11 DURING CAPTURE FROM HEAP TABLE ON SQL SERVER:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-924] FIXED JG352F WHEN THE REDSHIFT STAGINGDIRECTORYHVR IS A LOCAL DIRECTORY:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-920] FIXED SQL SERVER LBC UNCONDITIONALLY PRODUCING HUGE DIAGNOSTIC FILES ON ERROR:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-919] FIXED HANDLING OF PARTIAL ROLLBACKS IN SQL SERVER LBC:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-901] FIXED VARIOUS ERRORS WITH /EXTRA/INTEGRATEEXPRESSION ON POSTGRESQL:
    To fix this bug, upgrade HVR on the integrate machine(s).

    in reply to: Knowledge Base #8705
    Simon Faas
    Keymaster

    The Ingres database has had a change in its logging system with patch level 14983: introducing the new REPLACE type (for Ingres Bug #130681).

    This causes compatibility issues with older HVR versions. Versions before HVR 4.7.3/33 encounter error:
    F_JT0400: Log scanning error F_JZ2204. Unexpected Ingres record type 179 encountered while processing record with LSN [..]

    Whereas HVR version 4.7.3/34 & /35 would silently ignore updates to source tables.

    The is bug report HVR-1458 which was fixed in HVR 4.7.3/36

    Please contact customer support if you encounter problems.

    in reply to: Knowledge Base #8710
    Simon Faas
    Keymaster

    Due to a problem with compatibility between HUB and certain GUI versions, several problems can occur:

    Wrong matching of parameters to actions, causing, e.g. the gui to display: Environment /Name /NoBeforeUpdate
    Crashed on actions using very new parameters.
    GUI not showing some actions (e.g TableProperties)

    This can occur when an HVR GUI version older than 4.7.3/8 connecting to a hub version newer than version 4.7.3/8

    If that is the case in your environment then please upgrade the HVR GUI to a version greater than 4.7.3/8 (e.g. 4.7.3/25).

    This is bug-report HVR-957 which was fixed in versions 4.7.1/40 & 4.7.3/9

    Please contact customer support if you encounter problems

    in reply to: Knowledge Base #8762
    admin
    Member

    When trying to connect to a location with correct ORACLE_HOME set, the following error “F_JG202F: Attempt to load shared library failed” can happen even though environment variable LIBPATH (AIX) is set for HVR Operating System user.

    The workaround is different depending on whether the remote connection uses an hvrremotelistener or an inet-daemon.

    For the hvrremotelistener make sure that LIBPATH is set for the HVR Operating System user BEFORE the hvrremotelistener is started. The environment of the running process can be checked using command: ps eww {pid of hvrremotelistener}
    If LIBPATH is not returned from previous command, set the correct LIBPATH and restart the hvrremotelistener.

    For the inet-daemon it can happen that the environment variable LIBPATH is not passed to the HVR process and then the workaround is to add the following in /etc/inetd.conf on the line of the HVR service: -ELIBPATH={path to oracle_home/lib}

    (example -ELIBPATH=/u01/app/oracle/product/12.1.0/dbhome_1/lib)

    After this the inet-daemon needs to be restarted as root by the following;

    stopsrc -s inetd

    startsrc -s inetd

    Re-test connection and now it should work.

    in reply to: Knowledge Base #8768
    Frank Knot
    Participant

    Question: What are these files?

    Answer: These are dumps of the contents of the Hvr catalogs, created by Hvr Load so that capture and integ jobs can inspect actions without having to make an expensive connection to the hvr hub database.

    Question: Where are they?

    Answer: Hvr Load creates these in $HVR_CONFIG/router///catalog

    Question: How their contents be inspected?

    Answer: To see the contents of these files, use this command;

    $ hvrrouterview hub chn 53ba7f7e_6a5a4.cache > cache.xml

    Question: Why are there sometimes multiple files?

    Answer: Hvr Load does not want to delete an old file and create a new one because it’s worried that a running job may be reading that file.

    Instead it creates a new file with a bigger timestamp in its basename (“53db3914” means “Aug 1 2014” in hex), leaves the file that was ‘current’ and deletes other cache files.

    At the start of each cycle the running jobs should check if a new cache file has been created and then reload it.

    in reply to: Announcements #8683
    Steve Sheil
    Participant

    We have just released HVR 5.0 as GA (General Availability).
    Previous versions of HVR 5 were only Beta.

    This release is a major milestone; with general availability of its flagship feature (DDL replication) HVR has now jumped past our competitors in one of the few areas where it was weaker.

    HVR 5’s other big features include stuff for SQL Server (lower/faster interface), Postgres (we can capture!) and Teradata (huge speed boost).

    For this release we also took the opportunity for some specific renaming and reorganization of HVR commands and actions which had were not intuitive or had aged badly.

    See the wiki post [url=https://www.hvr-software.com/wiki/Migrating_from_HVR_Version_4_and_5]Migrating from HVR 4 to 5 [/url] for that.

    The last GA release before HVR 5.0 and 4.7 diverged was 4.7.3/7.
    Below is a list of new features since then.

    Regards,
    Steve

    NEW FEATURES IN HVR 5.0.1/0 (2016-04-19) GA
    ——————————————-

    [HVR-1474] ADD SUPPORT FOR HVR INITIALIZE CAPTURE REWIND FOR POSTGRESQL:
    To use this feature, upgrade HVR on the hub and capture machine(s).

    [HVR-1334] FORCED AUGMENT FOR ORACLE SECUREFILES BASED LOB COLUMNS:
    To use this feature, upgrade HVR on the capture and hub machine(s).

    [HVR-1439] HVR 5 ACTION RENAMES:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-1236] IMPLEMENT -TASK_GROUP PARAMETER FOR HVRMAINT TO SHARE STATE BETWEEN TASKS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1445] IMPROVED SUPPORT FOR SQL SERVER GEOMETRY DATATYPE:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-1449] PERFORMANCE IMPROVEMENTS ON FP_PROJECT:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-1467] TERADATA SUPPORTS BULK LOAD ON WINDOWS PLATFORMS:
    To use this feature, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 5.0.0/6 (2016-04-11)
    —————————————-

    [HVR-737] SET HVR_HUB_CHARSET DEFAULT TO UTF-8:
    To use this feature, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 5.0.0/5 (2016-04-03)
    —————————————-

    [HVR-1101] ADDED SUPPORT FOR DBINTEGRATE /NOTRIGGERFIRING OPTION FOR MS SQLSERVER:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-1342] ADDED SUPPORT FOR FILECAPTURE COLUMNPROPERTIES /EXTRA, /ABSENT, /CAPTUREEXPRESSION:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1391] ADDED SUPPORT FOR POSTGRESQL LOG-BASED CAPTURE OF TOASTED VALUES:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1417] ALLOW ANY DATATYPE FOR /ADDRESSTO={COL}, NOT JUST INTEGER:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1300] HVR SWITCHES TO MULTILOAD IF FASTLOAD FAILS DURING TERADATA BULK LOAD:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-1197] IMPROVE E-MAIL LAYOUT OF HVRMAINT:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1326] IMPROVED DB2 FOR I LOGBASED CAPTURE PERFORMANCE BY FILTERING UNUSED RECORDS:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1424] IMPROVED LSN TO SCAN SEQUENCE NUMBER TRANSLATION RULES FOR SQL SERVER:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1372] PROFILE FPROC CAN RETURN MIN AND MAX VALUE OF HVR OR DBMS VERSIONS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1338] RESTRUCTURE ADAPTDDL REFRESH OPTIONS, ADD CREATEUNICODEDATATYPES TABLEPROPERTY:
    To use this feature, upgrade HVR on the hub and gui machine(s).

    [HVR-1243] S3 INTERNAL SERVER ERRORS (STATUS CODE 500) ARE NOW RETRIED AS PER S3 RECOMMENDATIONS:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-1362] SHOW UNITS IN HVRSTATISTICS GRAPHS:
    To use this feature, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 5.0.0/4 (2016-02-29)
    —————————————-

    [HVR-981] POSTGRESQL LOG-BASED CAPTURE
    To use this feature, upgrade HVR on the hub and capture machine(s).

    [HVR-1238] INCREASE {HVR_CAP_USER} METACOLUMN LENGTH FROM 14 TO 24 CHARACTERS:
    To use this feature, upgrade HVR on the hub, capture and integrate machine(s).

    [HVR-1258] WORKAROUND FOR SYNCHRONY GATEWAY FTP SERVERS DROPPING UPLOADED FILES:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-107] HVRSCHEDULER NOW USES OS ASSIGNED PORTS BY DEFAULT. FIXES F_JG210E (ADDRESS ALREADY IN USE):
    To use this feature, upgrade HVR on the hub and gui machine(s).

    NEW FEATURES IN HVR 5.0.0/2 (2016-02-16)
    —————————————-

    [HVR-1219] ADD COMPARECONDITION TO RESTRICT ACTION:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-643] ADD HVR_ENROLL_VLOGFILE_RESTRICT TO SPECIFY REDO DISKGROUP:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1134] ADD TERADATA DISTRIBUTION KEY SUPPORT:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-738] ADD DISTRIBUTION KEY, CHARACTER LENGTH AND ENCODING TO HVR CATALOG:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1069] ADD DURATION AND SPEED MEASUREMENTS TO HVRSTATISTICS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-967] ADD OPTION -DISABLE AND -CLEAR_PAST_ERRORS TO HVRMAINT:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-1190] ADD SUPPORT FOR TERADATA BULK LOAD VIA TPT:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-1025] ADDED FILEINTEGRATE /ORDERBYTABLE TO IMPROVE {HVR_TBL_NAME} RENAME EXPRESSION FILE GENERATION:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-982] ALLOW DIFFERENT CAPTURE REWIND TIMES FOR SCAN START AND EMIT, AND ALLOW HVR SEQUENCE NUMBER OR ORACLE SCN FOR EMIT:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-1062] ALLOW MULTIPLE DBSEQUENCE /SCHEMA ACTIONS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-550] AVOID USING ROW AUGMENT DURING SQL SERVER LBC WHEN POSSIBLE:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1049] CHANGE ISOLATION LEVEL FOR DB2I AND SQL SERVER TO READ COMMITTED:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-1139] CLIENT-SIDE SSL CERTIFICATES FOR FTPS AND WEBDAV:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-1217] CONSTANT SQL EXPRESSION MATCHING SHOULD IGNORE WHITESPACES FOR ROBUSTNESS:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-1218] CUSTOMIZE FTP CONTROL SOCKET TIMEOUT VALUES VIA ENV VAR:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-1048] ELIMITATE THE NEED IN THE STATE TABLE ON CAPTURE SIDE FOR SQL SERVER:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1029] EXPOSE HVRMAINT OPTION -OUTPUT_VERBOSE:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-738] EXTENDED TABLE COLUMN VIEW WITH BYTELENGTH, CHARACTER LENGTH AND ENCODING:
    To use this feature, upgrade HVR on the gui machine(s).

    [HVR-1074] FILE INTEGRATE XML SERIALIZATION WITH COMPACT XML:
    To use this feature, upgrade HVR on the hub and gui machine(s).

    [HVR-761] HVR STATISTICS SUPPORT FOR HVR WARNINGS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-721] HVR PURGES THE UNUSED FILES IN HVR_HOME AFTER UPGRADE:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-1057] HVR_ISOLATION_LEVEL TO OVERRIDE DEFAULT ISOLATION LEVEL:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-1123] HVRREFRESH: ADD (DISTRIBUTION) KEY SENSITIVITY, CHANGE OPTION -CA TO -CE:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-1195] IMPLEMENT CAPTURE OF ORIGINAL USER NAME FOR SQL SERVER LBC:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-965] IMPLEMENTED FILE DIRECT (“FD”) TRANSACTION LOG ACCESS METHOD FOR SQL SERVER:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1098] IMPROVED DB2 FOR I CAPTURE SPEED:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1009] IMPROVED ORACLE RAC CLUSTER REDIRECTS:
    To use this feature, upgrade HVR on the hub machine(s).

    [HVR-545] IMPROVEMENTS TO CBUFLOCALIZE, ADDED CBUF2SQL AND CBUFDIFF FUNCTIONALITY:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-665] REFRESH TABLE RECREATE NOW SENSITIVE TO DATATYPE CHANGES:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-943] STANDARDIZING WARNING MESSAGES TO NEW FORMAT THAT INCLUDES MSG-ID W_JX0000:
    To use this feature, upgrade HVR on all machine(s).

    [HVR-807] SUPPORT FOR INDEX REBUILD DDL IN SQL SERVER:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-1040] VARIOUS SFTP COMPATIBILITY ENHANCEMENTS:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    NEW FEATURES IN HVR 5.0.0/1 (2015-10-30)
    —————————————-

    [HVR-977] ADD HVR_ENROLL_VLOG_RESTRICT AS WORK-AROUND FOR F_JD2822 [HVR-976]:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-263] DDL REPLICATION, USING ADAPTDDL ACTION:
    To use this feature, upgrade HVR on the capture and hub machine(s).

    NEW FEATURES IN HVR 5.0.0/1 (2015-10-30)
    —————————————-

    [HVR-977] ADD HVR_ENROLL_VLOG_RESTRICT AS WORK-AROUND FOR ORACLE F_JD2822 [HVR-976]:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-263] DDL REPLICATION, USING ADAPTDDL ACTION:
    To use this feature, upgrade HVR on the capture and hub machine(s).

    NEW FEATURES IN HVR 4.7.3/11 (2015-10-20)
    —————————————–

    [HVR-964] RE-IMPLEMENTATION OF SFTP DIRECTLY ON TOP OF LIBSSH2:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/9 (2015-10-13)
    —————————————-

    [HVR-914] ADD OPTION -D TO HVRADAPT TO REMOVE TABLES FROM CHANNEL:
    To use this feature, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 4.7.3/8 (2015-10-01)
    —————————————-

    [HVR-912] ADDED $HVR_SF_TIMEOUT_SECS FOR SALEFORCE DATALOADER TUNING:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-915] ADDED /INTEGRATEEXPRESSION={HVR_INTEG_KEY} AS CHAR(16) FOR UNIQUE KEY GENERATION:
    To use this feature, upgrade HVR on the hub and integrate machine(s).

    [HVR-917] ADDED FILECAPTURE /IGNORESIZECHANGES:
    To use this feature, upgrade HVR on the capture and hub machine(s).

    [HVR-923] ADDED HVR_IDENTITY_RESEED_ENABLE TO RESEED SQL SERVER IDENTITY COLUMNS:
    To use this feature, upgrade HVR on the integrate machine(s).

    [HVR-896] ORACLE REDO BACKWARD COMPATIBILE VERSION HANDLING:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-921] SUPPORT ADVANCING OF SQL SERVER TRANSACTION LOG TRUNCATION POINT:
    To use this feature, upgrade HVR on the hub and capture machine(s).

    in reply to: Knowledge Base #8758
    Frank Knot
    Participant

    HVR is able to capture (log based) from an Oracle Materialized View.
    This type of view is actually just a table which is updated periodically by Oracle using the view definition.

    To capture from such a view, its definition has to be entered manually (as a table) in the HVR catalogs. Table select / table explore currently only selects regular tables.

    On the target, the contents of the materialized view will be replicated to a regular table.

    in reply to: Knowledge Base #8760
    Simon Faas
    Keymaster

    For certain datatypes (like SQL Server geometry types) HVR uses “extended datatype support” instead of native datatype support.

    Here is how to setup replication (including capture and refresh) of such datatypes.

    Note that extra configuration is needed to replicate these columns.

    In this example have a table t1:

    CREATE TABLE [dbo].[t1](
    [id] [int] IDENTITY(1,1) NOT NULL,
    [geom1] [geometry] NULL
    ) ON [PRIMARY] TEXTIMAGE_ON [PRIMARY]

    GO

    The table is added the channel. After adding this table to the channel you will see that the column is not present. You will need to add it manually in the Table Properties dialog. It needs to be added with the same name but with datatype clob.

    The following actions are needed for HVR 4;

    { group=SRC table=* DbCapture /LogBased }
    { group=SRC table=t1 ColumnProperties /Name=geom1 /Absent /CaptureExpression="geom1.ToString()" /Context=refresh }
    { group=SRC table=t1 ColumnProperties /Name=geom1 /Absent /CaptureExpression="select t.geom1.ToString() from t1 t where t.id={id}" /Context=!refresh }
    { group=TGT table=* DbIntegrate }
    { group=TGT table=t1 DbIntegrate /DbProc /DbProcDuringRefresh }
    { group=TGT table=t1 ColumnProperties /Name=geom1 /CaptureExpression="geom1.ToString()" /IntegrateExpression="geometry::STGeomFromText({geom1},0)" }

    The actions for HVR 5 are nealy the same;

    { group=SRC table=* Capture }
    { group=SRC table=t1 ColumnProperties /Name=geom1 /Absent /CaptureExpression="geom1.ToString()" /Context=refresh }
    { group=SRC table=t1 ColumnProperties /Name=geom1 /Absent /CaptureExpression="select t.geom1.ToString() from t1 t where t.id={id}" /Context=!refresh }
    { group=TGT table=* Integrate }
    { group=TGT table=t1 Integrate /DbProc }
    { group=TGT table=t1 ColumnProperties /Name=geom1 /CaptureExpression="geom1.ToString()" /IntegrateExpression="geometry::STGeomFromText({geom1},0)" }

    As you see, I have 2 capture expressions on the source; with different context. One will be used during refresh (remember to activate it in the Context tab); the other one is used for replication.

    On the target you need to use integration and refresh with database procedures for table t1

    On the integrate we have an integrate expression and also a capture expression. The latter is needed to make HVR compare work.

    in reply to: Knowledge Base #8756
    admin
    Member

    This article describes how to setup parallel refreshes for large tables in HVR using a refresh condition with user variables to split the refresh load. For each refresh the content of the user variables will be changed, so the result will be multiple scheduled refreshes, each with it’s own range of rows to be refreshed for a specific table.

    View PDF here

Viewing 15 posts - 1 through 15 (of 356 total)
Test drive
Contact us