Activity for Troels Schwarz-Linnet
-
Troels Schwarz-Linnet posted a comment on discussion Model-free analysis
Hi David. It's many years ago i worked with relax, and I worked with dispersion data. But I recall that relax make a data pipe for each model, and then make a result pipe where it compares each model. Can you load the state file and see if there is other pipes ? Try changing to another pipe and then see if it works? Or, maybe it was a state file for each model and there is a result file, which have each model in its own pipe. Edward, do you know? On Thu, 17 Nov 2022, 17.39 L. David EMAIL Finger,...
-
Hi, What's the problem? On Tue, 5 Apr 2022, 17.29 Asokan Anbanandam, asokan05@users.sourceforge.net wrote: [tickets-support:#3] https://sourceforge.net/p/nmr-relax/tickets-support/3/ running R1 under GUI help request* Status: open Release: 4.2 Created: Tue Apr 05, 2022 03:29 PM UTC by Asokan Anbanandam Last Updated: Tue Apr 05, 2022 03:29 PM UTC Owner: nobody Attachments: Screen Shot 2022-04-05 at 11.22.33 AM.png https://sourceforge.net/p/nmr-relax/tickets-support/3/attachment/Screen%20Shot%202022-04-05%20at%2011.22.33%20AM.png...
-
Hi, have you started/restarted after changes? Try clean the temp file folder in installation.
-
Thanks Eric! I have already specified that foreign keys should not be created, since I only sync 20 out of 600 tables, and I found that creation of tables with referential integrity was not possible, since foreign keys was created for tables not synced/created. So, in this case, I read the worst case scenario is a moment of time where data can exist in B, before A, but eventually will fixed after sync. Could there be other concerns to be aware of ? Could application triggers be prevented in their...
-
Reference 1: http://wiki.openbravo.com/wiki/Projects:DB_Synchronization_With_Symmetric_DS/Trigger_Execution_Order
-
Hi, I have successful tried Symmetricds replication from Oracle to PostgreSQL. Before going into Production, I have been concerned about tables in Oracle, which already have application triggers. From what I have found in Manual and Google, Symmetricds triggers needs to run first. In MS SQL, there is a parameter that can order the order of trigger execution. But on Oracle, this does not exists. I found that for Oracle, it would require to delete the application triggers and instead specify them as...
-
Arghhhh.... It was my ODBC installation for PostgreSQL that had "Max Varchar" set to 255. This limited the data in DBeaver, when I looked at TEXT column. When I modified to 65k, I can easily see the data.
-
If I create a table in postresql CREATE TABLE test.rawdata ( id numeric NOT NULL, rawdata text, CONSTRAINT rawdata_pkey PRIMARY KEY (id) ); And then do: INSERT INTO test.rawdata (id, rawdata) VALUES(1, 'aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa...
-
-
If I create a table in postresql CREATE TABLE test.rawdata ( id numeric NOT NULL, rawdata text, CONSTRAINT rawdata_pkey PRIMARY KEY (id) ); And then do: INSERT INTO test.rawdata (id, rawdata) VALUES(1, 'aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa...
-
Also incoming files in tmp contains the whole information
-
I can see after a Reload Request, that files in tmp/xxx/outgoing contains the whole information of of the column. So it must be the reading in postgresql?
-
https://stackoverflow.com/questions/18394691/oracle-clob-cant-insert-beyond-4000-characters This says 4000 characters (the maximum string literal in Oracle). 4000 bytes/256 characters =15.6 bytes/character
-
https://sourceforge.net/p/symmetricds/discussion/739236/thread/7fa53f0d/?limit=25#e1f7 And this one: contains_big_lob=0 is equal doing dbms_lob.substr(column, 4000, 1 )
-
https://sourceforge.net/p/symmetricds/discussion/739236/thread/c0d84d75/?limit=25#7878 This looks related. Something with 256 and TEMP of Oracle.
-
Hi, I am syncing a table "RAWDATA" that has a column "RAWDATA" which is CLOB, and contains around 3.000 to 10.000 characters. When they are synced to PostgreSQL the column data is truncated to 256 characters. The table is created as: On oracle, the DDL for CREATE TABLE is: RAWDATA CLOB DEFAULT empty_clob(), On PostgreSQL the DDL for CREATE TABLE is: rawdata text DEFAULT 'empty_clob()'::text, I have played around with: Triggers: insert into sym_trigger use_capture_lobs, use_stream_lobs = 1,1 Channel:...
-
For ubuntu, I had to fix the path to java to an absolute path cd /opt/symmetric-server # Fix absolute path to JAVA java=`which java` if [ ! -f ./conf/sym_service.conf_bck ]; then echo "Making copy";cp ./conf/sym_service.conf ./conf/sym_service.conf_bck; fi sed -i "s@java.command=java@java.command=$java@" ./conf/sym_service.conf diff ./conf/sym_service.conf ./conf/sym_service.conf_bck ./bin/sym_service install # Check setup cat /lib/systemd/system/SymmetricDS.service systemctl status SymmetricDS systemctl...
-
Success! Setting in Oracle source: oracle.use.ntypes.for.sync=true I now get 3.198.950 at both source and target. In Oracle, in SYM_DATA, the columns ROW_DATA PK_DATA OLD_DATA are of NLOB type
-
Focusing just on s_sample, I rerun again. Source: 3.198.950 Target: 3.100.000 Since all batches seems to be transferred, it must then be batch creation?
-
In Oracle, in SYM_DATA, the columns ROW_DATA PK_DATA OLD_DATA are of CLOB type
-
Reading: https://medium.com/data-weekly/character-sets-and-symmetricds-c8f4c3ac6d4 Should I use: oracle.use.ntypes.for.sync Appendix B: Parameter List -> B.2. Runtime Parameters
-
Hi Eric, I tried to add some more tables to see if all tables were hit. Table; Source count; Target count OK: paramlist; 2052 : 2052 paramlistitem; 4319 : 4319 s_product; 1629 : 1629 study; 63 : 63 trackitem; 5958 : 5958 u_method_dept; 4970 : 4970 u_qcsample; 2072 : 2072 u_qcsample_res; 626707 : 626707 u_sla; 19553 : 19553 Difference: s_sample; 3198950 : 6 (Only 6 !) sdidata; 4.091.240 : 3.891.240 sdidataitem; 12.173.627 : 11.923.627 Source Oracle: select * from SYMMETRICDS.sym_outgoing_batch where...
-
The SYMMETRICDS was only given 100 MB quota on the node db. ALTER USER SYMMETRICDS QUOTA 100M ON USERS; But SELECT * FROM DBA_TS_QUOTAS WHERE TABLESPACE_NAME = 'USERS'; Shows that only 10MB is used out of the 100MB given.
-
Hi, I am testing setup of replication between Oracle (19C) to PostgreSQL (10). Using SymmetricDS 3.12.10. SymmetricDS is running on a VM, with JDBC connection to the 2 databases. The postgreSQL is registration server. Reload is from node to registration server. After a reload_request, a count on rows on source table says 12.173.416 , while target table says 12.050.000 There is no errors in the log showing. Table "sym_incoming_error" is empty. Is there an explanation for this? Node engine group.id=myapp...
-
https://www.symmetricds.org/doc/3.12/html/user-guide.html#_known_limitations The LONG data type is not supported with change capture, since LONG columns cannot be accessed from triggers. The data in LONG columns can still be kept in sync by enabling the use_stream_lobs feature in TRIGGER. With the default settings a database row cannot exceed 4k. If the error 'ORA-01489: result of string concatenation is too long' occurs then set use_capture_lobs to 1 in the TRIGGER table and contains_big_lobs to...
-
Trying to add MPI for Windows. It does not seem to work
-
Adding python 3.6 and adding test of mpirun.
-
Ignoring Windows C extensions
-
Trying MPI on Windows does not work
-
Added python as overall language to travis
-
Activating windows python 3.7 32bit for travis
-
Added travis-ci support for Python 3.7 and OSX.
-
Fixing a bug for running scons.
-
Fix for building on mac osx python 3.7
-
Simplify Travis file
-
Adding sending mails to nmr-relax-devel@lists.sourceforge.net.
-
Fixing the return value of execution of unit and systemtests
-
Adding exit codes for Unit and Systemtests.
1
