Frank Knot

Forum Replies Created

Viewing 15 posts - 16 through 30 (of 356 total)
  • Author
    Posts
  • in reply to: Community Discussions #8875
    Mark
    Keymaster

    What does the error mean, and how can I get around it?

    F_JG2901: UNKNOWN string ” containing invalid sequence encountered while encoding table ” column ”

    in reply to: F_JG2901: invalid sequence encountered #8877
    Mark
    Keymaster

    This error is a likely indicator of data corruption in the database.

    Internally HVR uses UTF8 to exchange data and hence there should be no data that cannot be transferred. As a result HVR will translate every string it finds in the database into a UTF8 string. HVR identifies character encoding in the database and will take encoding into consideration when the translation happens. The error “F_JG2901: UNKNOWN string ” containing invalid sequence encountered while encoding table ” column ”” (with values for the indicators in angle brackets) in indicates that somehow a string could not be translated into a valid UTF8 string.

    To get around this error you can use the action TableProperties /IgnoreCoerceError. This action will ignore the error and simply put in a valid character (like an upside-down question mark) for whichever character did not translate into a UTF8 string. This means that the actual data changes, but given it is invalid data to begin with this may be much better than having the replication fail.

    You may however want to do some root cause analysis when you run into a problem like this. I have seen this a few times on an Oracle Database. A couple of ways I have used in the past to find offending/corrupted data are:

    • Export (e.g. using expdp) the data and import it again. This has in some cases exposed the corrupted data.
    • Perform computations on the corrupted data when the data was numeric. This gave very weird results (e.g. 10x a positive number resulted in a negative amount); note that the error for this corruption was invalid target numeric size i.e. a different error than in this question.

    Of course when the corrupted data has been identified you can go back to the application/users to fix the data.

    in reply to: Community Discussions #8882
    Mark
    Keymaster

    Can I use HVR to replicate case sensitive table or column names?

    Yes, the action to use if at least one of the tables in the replication setup (channel) has a case sensitive name or column name is LocationProperties /CaseSensitiveNames. This action is particularly important against Oracle Databases when by default Oracle will assume table and column names are case insensitive unless quoted. The action LocationProperties /CaseSensitiveNames will ensure that table and column names will be quoted.

    Please note that the action LocationProperties /CaseSensitiveNames:

    1. Always applies to all tables in the channel.
    2. Is required before tables are added to the channel, since by default HVR will assume that table and column names are case insensitive. If tables were already added (through the Table Explore/Table Select function in the GUI) then tables will have to be removed and re-added to the channel.

    You may not realize that a table uses a case sensitive name or column name until you see errors like:

    • F_JD0006: A table named [.] does not appear in the DBMS catalogs during Table Explore/Table Select.
    • ORA-00942: table or view does not exist during a Refresh.
    • ORA-00904 “”: invalid identifier, e.g. during a Refresh. This may be an indication that the column reported in uppercase may in fact be in mixed case in the database.
    in reply to: Knowledge Base #8703
    Simon Faas
    Keymaster

    Due to a problem with compatibility between old HVR 4 hub installations and new HVR 5.0.1/0 GUI version action parameter names will be displayed wrong.

    This is bug report [HVR-1495]: ” HVR 5 GUI on top of HVR 4 hub gives incorrect parameter names”

    This problem will be fixed in upcoming GA release HVR 5.0.1/1
    Because the backward compatibility of HVR 5.0.1/0 is broken, we will deprecate that version.

    in reply to: Knowledge Base #8753
    Frank Knot
    Participant

    HVR tries to ‘stream’ data, but there are some moments when it has to create temp files.

    HVR”s temp files are created in $HVR_TMP, but if that is not defined then they are located in $HVR_CONFIG/tmp/

    HVR uses temp files at various moments. The following is an [incomplete] list,

    During log-based capture, if a transaction exceeds a limit (all limits are detailed below) then the capture job will spill it to disk on the capture machine.

    Row-wise compare and refresh sometimes need to sort data outside the DBMS. If the data in a table exceeds a limit then this data will be spill into multiple sort-merge files on the target machine.

    During row-wise refresh, if the number of changes detected exceeds a limit then these changes will be spilt to disk.

    During row-wise verbose compare or refresh from the GUI, if the number of changes detected exceeds a limit then these changes will be spilt to disk.

    If a single LOB value (e.g. long varchar or CLOB) exceeds a limit then it’s contents will be spilt to disk.

    Action Transform/ with parameter /Command directs the output of a command transform to a temp file and then reads it back.

    When bulk-loading data into Greenplum or Redshift, HVR puts the data into a ‘staging’ file first.

    The size of these files depends on the data that HVR is processing (obviously).

    Several of the above cases say that HVR spills to disk if it ‘exceeds a limit’.

    For cases 1, 3, 4 & 5 this limit is controlled by env-variable $HVR_TEMP_SPILL. This controls the number of bytes for the temporary ‘blobs’ in HVR’s memory; these will spill to disk if their size exceeds (X-T)/N, where X= $HVR_TEMP_SPILL, T=total_size_of_currently_cached_temp_objects and N=num_open_cached_objects. The default of $HVR_TEMP_STORE is 64M (bytes). This means that, by default, a single BLOB will spill to disk if it’s bigger than 32M..

    For case 2 (sorting), the ‘spill limit’ is controlled by env-variable $HVR_SORT_BYTES_LIMIT (default 512Mb) and $HVR_SORT_ROW_LIMIT (10M). If data exceeds one of these amounts then HVR’s sort will spill to disk. But it will never spill if the number of rows is below $HVR_SORT_ROW_MIN (default 128).

    HVR’s native [uncompressed] format was originally quite inflated, because a varchar(1000) containing “hello world” is held as 1000 bytes not as 11 bytes. So, for older HVR versions, for cases 1, 2 and 3 above (log-based capture transactions, sorting and row-wise changes), the temp file were sometimes much bigger than the original table data.

    To combat this ‘inflation’, since 4.6.1 HVR compresses most of its temp files. In HVR 4.6.1, temp files for cases 1, 3, 4 & 5 are compressed unless $HVR_TMP_NO_COMPRESS=1 is set.. Since HVR 4.6.2, temp files for cases 1, 3, 4 & 5 are compressed unless $HVR_TMP_COMPRESS_LEVEL=0 is set. Before HVR 4.6.2 files for case 2 (sorting during row-wise) are only compressed if HVR_SORT_COMPRESS=1 is set. Since 4.6.2 files for case 2 are compressed unless $HVR_SORT_COMPRESS_LEVEL=0 is set.

    in reply to: Announcements #8678
    Simon Faas
    Keymaster

    We have just released HVR 5.0.1/1 as GA (General Availability)

    This release fixes a major compatibility issue between old HVR 4 hub installations and new HVR 5.0.1/0 GUI versions: [HVR-1495]: ” HVR 5 GUI on top of HVR 4 hub gives incorrect parameter names”
    Due to the compatibility issue in 5.0.1/0 we will deprecate that version.

    Below is a list of new features and problems fixed.

    NEW FEATURES IN HVR 5.0.1/1 (2016-05-11)
    —————————————-

    [HVR-1501] ADD SUPPORT FOR POSGRESQL VERSION 9.5:
    To use this feature, upgrade HVR on the capture machine(s).

    [HVR-413] ADDED -P OPTION TO HVRMAINT.PL FOR TASK HIGH PRIVILEGES:
    To use this feature, upgrade HVR on the gui and hub machine(s).

    [HVR-1493] ENDIANNESS CONVERSION PERFORMANCE IMPROVEMENTS:
    To use this feature, upgrade HVR on the capture and integrate machine(s).

    [HVR-1491] WIDE ROW CAPTURE PERFORMANCE IMPROVEMENTS:
    To use this feature, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 5.0.1/1 (2016-05-11)
    ——————————————

    [HVR-1513] FIXED POSTGRESQL LOG-BASED CAPTURE SLOW INITIAL START OF CAPTURE JOB:
    To fix this bug, upgrade HVR on the hub and capture machine(s).

    [HVR-1508] BURST TABLES ARE RECREATED WHILE BEING EQUAL TO THE TARGET TABLE:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1405] F_JZ1537. ‘APPEND HINT’ CONTAINS DATA FOR TOO MANY COLUMNS:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1487] FIXED F_JT1482: PIPE BURSTPIPE ENCOUNTERED COLUMN HVR_CAP_LOC WITH ILLEGAL LEN FOR CESU8 DATABASES:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1500] FIXED SQL SERVER CONSUMES CPU ON A QUIET SYSTEM WITH /NOTRIGGERFIRING OPTION:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1319] FIXED CONFUSING ERROR MESSAGE AFTER F_JZ6208 FOR POSTGRESQL LOG-BASED CAPTURE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1495] FIXED INCORRECT PARAMETERS BEING DISPLAYED WHEN EDITING ACTIONS:
    To fix this bug, upgrade HVR on the gui and hub machine(s).

    [HVR-1482] ORACLE CAPTURE FAILED WITH F_JZ1952: PASTE ENCOUNTERS ENDING TYPE 0X0005 WITH LOB BUT DATA STORE IS NOT READY YET FOR UPDATE_AFTER:
    To fix this bug, upgrade HVR on the capture machine(s).

    in reply to: Community Discussions #8871
    Frank Knot
    Participant

    When a running job with a capture / integrate agent (action agent / agentplugin) is suspended in the HVR scheduler, HVR seems to kill the agent. Is this correct behaviour and why is HVR doing this.

    Frank Knot
    Participant

    The Capture / Integrate agent as defined with the agent / agentplugin action is running as an integral part of the replication stream. It is part of the replication job (hence the term plug in), HVR is capturing its output & status and acts upon it.

    When a running replication job is suspended, HVR will clean up all parts of this job including a running agent.

    This clean up is done gracefully, by sending a termination signal to the agent process . The agent can therefore be programmed to catch the termination signal and perform a proper rollback when needed.

    in reply to: Announcements #8676
    Simon Faas
    Keymaster

    HVR 4.7.3/41 has been released as GA on May 18th 2015.

    Below is an extract from the release notes containing a list of new features and problems fixed since the last GA release (4.7.3/25)

    PROBLEMS FIXED IN HVR 4.7.3/41 (2016-05-18)
    ——————————————-

    [HVR-1554] FIXED F_JD291D: TERADATA ERROR 6705 AN ILLEGALLY FORMED CHARACTER DURING REFRESH:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1514] FIXED S3 ERROR RETRY MIGHT TRIGGER 100% CPU USAGE AND HANGING WITH REDSHIFT:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1011] ORACLE CAPTURE FAILED WITH F_JD2428: PARTIAL LOB OP FOR NCLOB CONTAINED 16 NOT 64 BYTES FOR LOBS:
    To fix this bug, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/40 (2016-05-12)
    ——————————————-

    [HVR-867] FIXED F_JD20F0: UNEXPECTED ODBC STATUS ‘SQL_NO_DATA’ DURING BURST INTEGRATE INTO DB2LUW:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1536] FIXED F_JD2103 WHILE SELECTING FROM INGRES 10.2 IIGET_TABLE_LOCATIONS:
    To fix this bug, upgrade HVR on all machine(s).

    [HVR-1512] FIXED F_JD2146 DURING ROW-WISE REFRESH WITH /COLUMNPROPERTIES /BASENAME:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1510] FIXED F_JG358E OVERWRITING OTHER ERRORS DURING PARALLEL REFRESH:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1511] FIXED POSSIBLE ISSUES IN JOB PROCESS CLEANUP WHEN SCHEDULER IS FORCEFULLY TERMINATED:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1440] FORCED AUGMENT FOR ORACLE SECUREFILES – PREVENT GENERATION OF F_JT1101 FOR OLD HUBS:
    To fix this bug, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/39 (2016-04-25)
    ——————————————-

    [HVR-1405] F_JZ1537. ‘APPEND HINT’ CONTAINS DATA FOR TOO MANY COLUMNS:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1482] ORACLE CAPTURE FAILED WITH F_JZ1952: PASTE ENCOUNTERS ENDING TYPE 0X0005 WITH LOB BUT DATA STORE IS NOT READY YET FOR UPDATE_AFTER:
    To fix this bug, upgrade HVR on the capture machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/38 (2016-04-19)
    ——————————————-

    [HVR-583] F_JD0700: DATA TYPE MISMATCH FOR _CLOB DATATYPE WHEN USE REDSHIFT AS TARGET LOCATION:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1384] FIXED HVR HANG DURING INTEGRATION INTO TERADATA:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1477] FIXED CONCURRENT TERADATA BULK LOADS MAY HANG:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1106] HVRGUI STATISTICS OUTPUT SOMETIMES SHOW WRONG LABELS (E.G. ERROR CODES INSTEAD OF JOBNAME):
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1146] ORACLE CAPTURE FAILS WITH F_JG0E07: UNHANDLED ICU ERROR OCCURED (U_TRUNCATED_CHAR_FOUND):
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/37 (2016-04-15)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/37 (2016-04-15)
    ——————————————-

    [HVR-1465] FIXED F_JG2142 WHEN COLUMNPROPERTIES /TIMEKEY WITHOUT NON-/EXTRA REPLICATION KEY:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1303] MISSING COLUMN QUOTING IN FAIL TABLE INSERT WHEN INTEGRATE /BURST IS DEFINED:
    To fix this bug, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/36 (2016-04-13)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/36 (2016-04-13)
    ——————————————-

    [HVR-1458] FIXED MISSING CAPTURE OF NEW UPDATE RECORDS INTRODUCED IN INGRES 10.2 BY INGRES BUG FIX 130681:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/35 (2016-04-12)
    —————————————–

    [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).

    PROBLEMS FIXED IN HVR 4.7.3/35 (2016-04-12)
    ——————————————-

    [HVR-1017] DB2 FOR I: [DELETE FROM TBL] LOGGED AS [ALTER TABLE] WITH NON-TRIVIAL TABLE SIZES:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1222] FIXED F_JD211F: UNALLOCATED DA ENCOUNTERED FOR TABLE WHILE CONVERTING DATA FROM COPYISH TO DB NATIVE FORMAT:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1255] FIXED F_JD2259: NO COLUMNS IN SET-CLAUSE AFTER ONLINE REFRESH:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1430] FIXED F_JR0336 WITH DBOBJECTGENERATION WITH JUST /BURSTTABLECREATECLAUSE:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1426] FIXED GUI MAY CRASH WHEN TERADATA TPT LIBRARY PATH IS LEFT EMPTY:
    To fix this bug, upgrade HVR on the gui machine(s).

    [HVR-1442] FIXED BATCH LOAD PERFORMANCE DROPS TO SINGLETON INSERT:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1434] FIXED HVRSTATISTICS IRREGULAR OUTPUT IF COMPRESSION INFO IS MISSING BY REFRESH:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1094] FIXED INCORRECT LAYOUT IN THE ‘SHOW GRAPH’ BOX ON THE STATISTICS-GRAPH-ALL TAB:
    To fix this bug, upgrade HVR on the gui machine(s).

    [HVR-1444] ORACLE INTEGRATE /BURST USES BATCH LOAD INSTEAD BULK:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1419] ORACLE CAPTURE FAILED WITH F_JD2279: UNEXPECTED ERROR WHILE CONVERTING UTF-16 DATA FOR COLUMN (DATATYPE CLOB):
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1387] ORACLE CAPTURE FAILED WITH F_JG200C: INTERNAL ERROR TRIGGERED BUS ACCESS TRAP SIGBUS:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1408] W_JT044C MISSING BEGIN TRANSACTIONS IN SAP SYSTEMS:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1399] CREATING FAIL TABLE ON REDSHIFT FAILS ON HVR_ERR_MSG COLUMN WITH F_JD22CA:
    To fix this bug, upgrade HVR on the integrate machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/34 (2016-03-25)
    ——————————————-

    [HVR-1402] FIXED F_JG2901 INVALID CHAR SEQUENCE DURING TABLESELECT OF TABLE WITH CHINESE CHARS:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1406] FIXED HVR CREATES WRONG TBL-NAMES FOR TABLES WITH CHINESE NAME:
    To fix this bug, upgrade HVR on the gui and hub machine(s).

    [HVR-1138] ORACLE LBC FAILS WITH F_JT1456: LOG PARSING ERROR F_JZ1948. PASTE ENCOUNTERS LOB COLUMN AT TABLE BUT ITS LENGTH IS SMALLER THAN EXPECTED:
    To fix this bug, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 4.7.3/33 (2016-03-23)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/33 (2016-03-23)
    ——————————————-

    [HVR-1380] FIXED $ZIZ_FILE_OFF(_LIVE) WITHOUT ZIZ_FILE_READ=READDIRECT:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1381] FIXED ACL FORCED FOR SOME S3 FILE UPLOADS:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-910] FIXED F_JD22E1: LOG RECORD HAS UNEXPECTED LENGTH ON DB2 FOR I:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1257] FIXED TBLWRITE DOES NOT USE THE SCHEMA NAME DURING TABLE CREATION:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1389] HVRVALIDPW DOES NOT WORK CORRECTLY ON SOLARIS:
    To fix this bug, upgrade HVR on the hub machine(s).

    PROBLEMS FIXED IN HVR 4.7.3/32 (2016-03-18)
    ——————————————-

    [HVR-1368] DO NOT KEEP GUARD TRANSACTION OPEN WHEN READING FROM SQL SERVER TRANSACTION LOG BACKUP:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1341] F_JZ1803 FIXED COLUMN NUMBER IN ORACLE 12 APPEND HINT:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1363] FIXED F_JC007C ERROR STARTING WINDOWS SERVICE ‘HVRSCHEDULER_XXX’ FOR POSTGRESQL HUB:
    To fix this bug, upgrade HVR on the gui machine(s).

    [HVR-1344] FIXED F_JD22CA: [CURRENT TRANSACTION IS ABORTED] ON REDSHIFT WITH AMAZON DRIVER:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1351] FIXED F_JG052E WITH SFTP PROTOCOL ERROR:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1369] FIXED SIGSEGV WHEN FPROC RETURNS RESULT DIFFERENT THAN THE CONTEXT:
    To fix this bug, upgrade HVR on all machine(s).

    [HVR-1358] FIXED LEFTOVER FILES IN S3 AFTER REDSHIFT INTEGRATE:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1345] FIXED NOT WORKING JOB RETRY WHEN POSTGRESQL HUB:
    To fix this bug, upgrade HVR on all machine(s).

    [HVR-1355] ORACLE CAPTURE FAILED WITH F_JG2901: WIN-1252 STRING CONTAINING INVALID SEQUENCE ENCOUNTERED WHILE ENCODING TAB:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1360] HVRROUTERVIEW -J CANNOT FIND JOURNAL FILES FROM DBINTEGRATE/JOURNAL:
    To fix this bug, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 4.7.3/31 (2016-03-11)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/31 (2016-03-11)
    ——————————————-

    [HVR-1267] F_JD0A42 DURING CAPTURE FROM SQL SERVER IF HVR_TLOG_BACKUP_DIR POINTS TO THE ROOT FOLDER OF A NETWORK DRIVE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1239] FIXED “NO SUCH FILE OR DIRECTORY AT HVRALERT LINE 64” WITH POSTGRESQL HUB:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1171] FIXED F_JC04C3 WHEN USING GENERATED COMMAND LINE STRING FOR POSTGRESQL HUB:
    To fix this bug, upgrade HVR on the gui machine(s).

    [HVR-1332] FIXED F_JD2272: RELATION ALREADY EXISTS DURING GREENPLUM REFRESH FOR TABLE WITH MIXED CASES:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1320] FIXED F_JD22A1: EXCEEDING MAXIMUM ROWHASHLOCKS INTO TERADATA:
    To fix this bug, upgrade HVR on the hub and integrate machine(s).

    [HVR-1310] FIXED F_JD2918 SYNTAX ERROR DURING TERADATA BULKLOAD TO TABLE WITH “/” IN THE NAME:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1298] FIXED F_JD2918: CAN NOT BULK LOAD DATA IN TERADATA TABLE WITH SECONDARY INDEX:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1209] FIXED F_JD2918: TPT ERROR (SYNTAX ERROR) BECAUSE OF UN-QUOTED COLUMN NAME:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1291] FIXED F_JG200A DURING REFRESH OR INTEGRATE TO POSTGRESQL:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1124] FIXED F_JG358E DURING BURST INTEGRATE INTO POSTGRESQL:
    To fix this bug, upgrade HVR on the capture, hub and integrate machine(s).

    [HVR-1143] FIXED F_JT1449: ‘HVR_REC_TSTAMP’ WAS NOT MATCHED:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1115] FIXED HVR CONTROL SKIPPING FOR HVR_TX_SEQ:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1207] FIXED HVR UNEXPECTED TRUNCATES FIXED-LENGTH CHARACTER STRINGS DURING REFRESH:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1315] FIXED ILLEGAL DIVISION BY ZERO IN HVRSTATISTICS.PL:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1278] FIXED SOFTDELETE MIGHT CAUSE MEMORY CORRUPTION, SPURIOUS F_JG221D:
    To fix this bug, upgrade HVR on the hub and integrate machine(s).

    [HVR-1305] FIXED DISAPPEARING OF LATENCY INFO FOR RUNNING CAP JOBS IF TOO MANY FILES IN HVR_CONFIG/ROUTER:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1256] FIXED FAILED JOBS GOING TO PENDING STATE ON BUSY WINDOWS HUBS:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1263] FIXED FILE REPLICATION FOR SSH,FTP,WEBDAV WHEN NON-DEFAULT PORT IS USED:
    To fix this bug, upgrade HVR on the capture and integrate machine(s).

    [HVR-1312] FIXED GUARDIANS NOT TERMINATING CHILD PROCESSES:
    To fix this bug, upgrade HVR on the hub, capture and integrate machine(s).

    [HVR-1321] FIXED LEFTOVER DEBUG MESSAGE IN DB2 FOR I LBC:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1301] FIXED MISSING UPDATES WITH /BURST OR /COALESCE:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1294] HVR REFRESH FAILS WITH F_JD21B5: [ORA-00904: HVR_CAP_USER]:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1200] SUPPORT FOR APPEND HINT INSERTS FOR COMPRESSED TABLES:
    To fix this bug, upgrade HVR on the capture machine(s).

    NEW FEATURES IN HVR 4.7.3/30 (2016-02-23)
    —————————————–

    [HVR-1219] ADD COMPARECONDITION TO RESTRICT ACTION

    NEW FEATURES IN HVR 4.7.3/30 (2016-02-23)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/30 (2016-02-23)
    ——————————————-

    [HVR-1249] FIXED F_JD2503 INVALID PATH FOR VASM CAPTURE:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1251] FIXED ZIZ_FILE_READ AND ZIZ_FILE_READ_LIVE WITH READDIRECT:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1174] ORACLE CAPTURE FAILED WITH F_JC0018: INTERNAL ERROR: UNREGISTERED EVENT:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1233] SUPPORT FOR EC2 ROLE CREDENTIALS FOR S3 DURING REDSHIFT BULK LOAD:
    To fix this bug, upgrade HVR on the integrate machine(s).

    NEW FEATURES IN HVR 4.7.3/29 (2016-02-12)
    —————————————–

    [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).

    PROBLEMS FIXED IN HVR 4.7.3/29 (2016-02-12)
    ——————————————-
    [HVR-1213] ORACLE CAPTURE FAILED WITH SEGMENTATION VIOLATION WHILE HANDLING URP RECORD
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1211] FIXED TERADATA BULK LOAD FOR BIGGER DECIMALS AND NUMBER:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-999] ORACLE CAPTURE MISSING VALUE FOR UPDATE WHEN LAST COLUMN(S) IS NULLIFIED:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1051] SLOW CATALOG INITIALIZE WHEN LOTS OF LOCATIONS AND CHANNELS:
    To fix this bug, upgrade HVR on the hub machine(s).

    NEW FEATURES IN HVR 4.7.3/28 (2016-02-01)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/28 (2016-02-01)
    ——————————————-

    [HVR-1110] FIXED ILLEGAL INSERT-INSERT PAIR (F_JT0527) WITH BURST INTEGRATE ON HVR_STATS CHANNEL:
    To fix this bug, upgrade HVR on the hub machine(s).

    [HVR-1173] ORACLE CAPTURE FAILED WITH F_JZ1105: SHIFT VALUE EXTRACTED FROM SHIFTS ARRAY OF QUICK MULTIPLE INSERT/DELETE FOR COMPRESSED COLUMN IS INVALID:
    To fix this bug, upgrade HVR on the capture machine(s).

    [HVR-1193] TABLE SELECT FILTER LIMITED TO UNDER 10 TABLE ENTRIES:
    To fix this bug, upgrade HVR on the gui machine(s).

    NEW FEATURES IN HVR 4.7.3/26 (2016-01-26)
    —————————————–

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

    PROBLEMS FIXED IN HVR 4.7.3/26 (2016-01-26)
    ——————————————-

    [HVR-1179] FIXED REDSHIFT CAN’T PARSE INSTANCE PROFILE ROLE CREDENTIALS:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1176] GREENPLUM EXTERNAL TABLE FOR BURST IS RECREATED WHEN DTYPES ARE CHANGED:
    To fix this bug, upgrade HVR on the integrate machine(s).

    [HVR-1142] ORACLE CAPTURE FAILED WITH F_JZ194F: PASTE ENCOUNTERS ZFRAG_11_17_LOB1 BUT ITS TABLE ID DIFFERS FROM EXISTING CHANGE TABLE ID:
    To fix this bug, upgrade HVR on the capture machine(s).

    in reply to: Knowledge Base #8750
    Simon Faas
    Keymaster

    In order to replicate tables with columns with datatype XMLTYPE you will need to do the the following:

    1. Add action Environment /Name=HVR_VIRTUAL_COLS /Value=1 (define this for location group ‘*’ and tables ‘*’) to your channel.

    2. (Re)run Table Select from the GUI (you can optionally specify the names of the tables containing XMLTYPE using Table Filter. this will make Table Select run faster)

    The XMLTYPE columns should now be visible in the Properties window of the tables containing XMLTYPE columns.
    Note that these columns will be shown with datatype CLOB.

    3. Actions

    For Log-based capture:

    Define 2 ColumnProperties actions; The first one will be used during replication. The 2nd will be used durin refresh

    a. Add action ColumnProperties /Name=xml /CaptureExpression=(select xmltype.getClobVal(t.xml) from document_xml t where t.document_id = {document_id}) /Context=!refresh
    Only define this for the capture location.

    b. Add action ColumnProperties /CaptureExpression=xmltype.getClobVal(xml) /Context=refresh
    Define this for capture and integrate locations

    For trigger-based capture

    Define action ColumnProperties /CaptureExpression=xmltype.getClobVal({xml})
    Define this for capture and integrate locations

    Notes:

    • Define this for each table containing an XMLTYPE datatype column.
    • Replace xml with the name of the column with the XMLTYPE datatype.
    • Replace document_xml with the name of the table.
    • Replace document_id with the name of the key column of the table.

    4. Add action ColumnProperties /Name=xml /IntegrateExpression=xmltype.createXml({xml})

    Define this for each table containing an XMLTYPE datatype column.
    Only define this for the integrate location.
    Replace xml with the name of the column with the XMLTYPE datatype.

    5. Add action DbIntegrate /DbProc /DbProcDuringRefresh

    Define this action for each table containing an XMLTYPE datatype column.
    Only define this for the integrate location.

    in reply to: Community Discussions #8867
    Mark
    Keymaster

    How to resolve the following error?

    F_JG202F: Attempt to load dynamic shared library ‘[some hvr connectivity library].so’ failed with error: ‘[referenced library].so: cannot open shared object file: No such file or directory’. This problem could be because this library or a library it referenced was not contained in library search path variable ‘LD_LIBRARY_PATH’.

    in reply to: How to resolve F_JG202F? #8869
    Mark
    Keymaster

    Error F_JG202F occurs every now and again when the first attempt to connect to a database is made on a Linux or Unix system. The issues is generally that [referenced library].so is not installed on the system, or cannot be found.

    Here is how to resolve the issue:

    1) Double check the HVR release notes $HVR_HOME/hvr.rel on the machine that connects to the database to make sure that the required connectivity libraries for your database have been installed. If connectivity libraries are missing then install these.

    2) With all libraries installed on the system you may still see F_JG202F. If that is the case then find out what is missing by using for example ldd on Linux to find the linked dynamic dependencies. E.g.:

    ldd hvr_or120.so

    This command will output the list of referenced libraries for the HVR Oracle 12c connectivity library. It will be clear from the output if one of the referenced libraries cannot be found. You may then have to set or adjust the LD_LIBRARY_PATH (on Linux; Unix systems have their own equivalent) environment variable to ensure all dynamically linked dependencies can be found.

    3) The last step is to make sure that when HVR initiates the connection the environment is set correctly. This generally involves restarting the HVR remote listener, and depending on how it is started you may have to edit the xinetd setup you created or add the environment settings to the hvrtab file when using hvr_boot to start the remote listener upon server (re)start.

    Note that you may need to go back through these steps if you initially chose to run the hvrremotelistener process on the command line and later switched to an xinetd or hvr_boot automatic startup.

    in reply to: Knowledge Base #8701
    Simon Faas
    Keymaster

    An memory problem can occur if HVR has to capture changes from SQL Server 2008’s backup transaction files.
    It seems to be due to a bug inside SQL Server 2008 which causes a serious memory leak.
    This problem does not occur in Microsoft SQL Server 2012 and higher.

    It only occurs if HVR is using the SQL read method (aka FN), and if the latency is so large that HVR must read changes from backup transaction files.
    The SQL mode is used in HVR 5.0 when Capture /LogReadMethod=SQL is specified and also used in HVR 4.7 and 4.8 (as default).

    The main workaround is to use DIRECT (aka FD) mode.
    In HVR 5.0, define Capture /LogReadMethod=DIRECT.
    In HVR 4.8 a special environment variable can be used to switch to this same method. Contact Technical Support for details.
    Another workaround is to upgrade from SQL Server 2008 to 2012 or higher.

    A disadvantage of DIRECT mode is that HVR needs higher (Windows system) privileges. However HVR does not support the “SQL” method with “minimal privileges” for SQL Server 2008 anyway (due a different issue).
    The DIRECT transaction log read method is much faster, but requires that HVR agent is installed on the SQL Server source database machine.

    It is strongly recommended to switch to DIRECT mode when capturing from a SQL Server 2008 database because the memory ballooning can kill a production DBMS system (not just HVR itself).

    in reply to: Knowledge Base #8745
    Herman Verheul
    Keymaster

    Problem:

    Capture fails with error F_JD2822 Unable to find logfiles or archives with valid non-zero sequences in the system tables while searching for timestamp and additional error F_JT0528: The previous error occurred while initializing cluster thread in reader

    Cause:

    Specific redo thread mentioned in error F_JT0528 is not used. The status in view v$log shows UNUSED for the specific thread

    Solution:

    On the location group for the capture (source database) add a new action with /Name=HVR_ENROLL_VLOG_RESTRICT /Value=”thread# <> ” , then run HVR LOAD/HVR Initialize with options “Table enrollment” and “Scripts and Jobs” or Drop the logfile groups belonging to the specific thread , e.g if groups 7,8,9 are belonging to thread 2 and status for those groups in v$log is UNUSED;

    ALTER DATABASE DROP LOGFILE GROUP 7;
    ALTER DATABASE DROP LOGFILE GROUP 8;
    ALTER DATABASE DROP LOGFILE GROUP 9;

    then run HVR LOAD/HVR Initialize with options “Table enrollment” and “Scripts and Jobs”

    in reply to: Knowledge Base #8748
    Mark
    Keymaster

    HVR uses Oracle’s native connectivity to connect to the Oracle Database. Oracle’s native connectivity has many options, and with that, there are many ways to connect HVR to the Oracle Database.

    The most efficient connection to an Oracle Database is to connect locally to the database when running on the same server. Such a local connection bypasses the TNS listener and should be used if possible i.e. if HVR runs on the database server(s). The way to connect in this scenario is to provide the path to the ORACLE_HOME out of which the database runs and the SID. DBAs will typically know what these values are.

    To get to the ORACLE_HOME you can use the remote browse capabilities in the HVR GUI. Note that the HVR OS user must have access to the location of the ORACLE_HOME in order for this connection to succeed. Make sure to use the ORACLE_HOME that the database runs out of, and not some other ORACLE_HOME e.g. the one that is used for the Grid/ASM installation or if there are multiple installations of Oracle on the server. The SID is case sensitive on Linux/Unix environments and a typo in the SID name will cause the connection to fail.

    In an Oracle RAC setup it is strongly recommended you use the SCAN listener (Oracle Clusterware 11gR2 introduced this feature) in HVR with the ability to run HVR on every node, maybe because HVR was installed on every node, or because the software was installed on a shared disk accessible on every node. In a RAC setup the HVR remote listener must be running on the same port (e.g. 4343) on every node. When using the SCAN listener HVR does not know ahead of time which node in the cluster it will be connected to and initiate the connection to one of the nodes based on the response from the SCAN listener. From there HVR will connect to the database using the service name that is provided as part of the connection information. For Oracle RAC this is the preferred way to connect because it is independent of the availability of any one of the servers in the cluster, and if the cluster is expanded or shrunk then there is no need to make any changes to the HVR connectivity information.

    In some cases however you may be connecting to the Oracle Database using TNS. For example:

    • The hub database schema is in a database on a different server.
    • The connection is to an Oracle RAC that does not use the SCAN listener (typically an old cluster).
    • The connection is to an Oracle Database service such as Amazon RDS (Relational Database Service).
    • The system administrator or DBA does not allow installation of any third-party software on the database server.

    When HVR connects to the Oracle Database using TNS it still needs Oracle client libraries to connect. It is best to start with a regular client (or a full server Oracle home if it is available on the machine that is running HVR) to avoid a currently required workaround to use the Oracle Instant client.

    The connection to the Oracle Database will be using:

    • Environment variable TNS_ADMIN to find the required files to connect to the Oracle Database, starting with sqlnet.ora, specifying the one or more ways the client can initiate a connection to the database and the order in which they are considered.
    • Common entries in this file include TNSNAMES and EZCONNECT. If TNSNAMES is in the list then TNS_ADMIN is also where the connection will be looking for tnsnames.ora.
    • ORACLE_HOME value to find within this in the directory network/admin sqlnet.ora, and if applicable tnsnames.ora.

    Note that TNS_ADMIN may have to be provided as an environment variable in order for HVR to use it. This can be done multiple ways:

    • As part of the environment when running the HVR remote listener.
    • In the connection dialog when connecting to the hub database.
    • Using an Environment action in the context of a channel.

    Please refer to the Oracle Documentation for more details on how to construct a connect string.

Viewing 15 posts - 16 through 30 (of 356 total)
Test drive
Contact us