Hi Michal,
You can start the HSQL servers before you execute the workflows. To do this, all the databases need to be configured to use a server instead of directly accessing the file. By default, the databases for the cache and core use a server, but not for provenance. To make the provenance database use a server, update configuration.xml in the provenance module so that DB Host is set to localhost. More information on configuring the databases can be found here: https://kepler-project.org/developers/reference/accessing-hsql/ To start the servers: kepler.sh -hsql start After the workflows finish execution, you can stop the servers: kepler.sh -hsql stop It looks like the location of KeplerData and .kepler can be changed by setting $HOME, but I have not exhaustively tested this. --dan On 5/22/12 2:24 AM, Michal Owsiak wrote:
Hello Kepler team, I have following issue. I'd like to run multiple instances of the workflow at the same time. However, this execution mode fails. I am attaching simple workflow (simple_workflow.xml) I am starting it following way: --- CUT start_workflow.sh --- #!/bin/bash $KEPLER/kepler.sh -runwf -nogui -nocache simple_workflow.xml & $KEPLER/kepler.sh -runwf -nogui -nocache simple_workflow.xml & $KEPLER/kepler.sh -runwf -nogui -nocache simple_workflow.xml & $KEPLER/kepler.sh -runwf -nogui -nocache simple_workflow.xml & --- CUT --- If you try to execute it in multiple parallel instances, you will receive errors similar to what I have: [null] 980 ms. Memory: 493632K Free: 479782K (97%) [null] java.sql.SQLException: File input/output error: /Users/owsiak/KeplerData/modules/core/coreDB.script.new [null] at org.hsqldb.jdbc.Util.sqlException(Unknown Source) ... ... ... ... [null] Command failed: ptolemy.kernel.util.IllegalActionException: RecordingException: Error generating new LSID for workflow execution : null [null] in .simple_workflow.TOP Provenance Recorder [null] ptolemy.kernel.util.IllegalActionException: RecordingException: Error generating new LSID for workflow execution : null [null] in .simple_workflow.TOP Provenance Recorder [null] at org.kepler.provenance.ProvenanceRecorder.preinitialize(ProvenanceRecorder.java:605) Question: is it possible to run multiple instances of the Kepler at the same time? How can one achieve that? Question 2: is it possible to use database in server mode? How can I prepare this kind of setup? Question 3: is it possible to change location of "KeplerData" and ".kepler" for each running Kepler process? Thanks in advance for the support Cheers Michal _______________________________________________ Kepler-users mailing list Kepler-users@kepler-project.org http://lists.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-users
_______________________________________________ Kepler-users mailing list Kepler-users@kepler-project.org http://lists.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-users
