Kepler: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362012-08-28T00:15:15ZEcoinformatics Redmine
Redmine Bug #5699 (New): Change the workflow scheduler server documentation link in the workflow-schedule...https://projects.ecoinformatics.org/ecoinfo/issues/56992012-08-28T00:15:15ZJing Taotao@nceas.ucsb.edu
<p>Currently the the workflow scheduler server documentation points to the svm trunk version. In the next workflow-scheduler-gui release, the documenation should be a branch or tag version.</p> Bug #5586 (New): Get newest version of local files in EML actorhttps://projects.ecoinformatics.org/ecoinfo/issues/55862012-04-16T18:54:00ZJing Taotao@nceas.ucsb.edu
<p>Here is an email from kepler user:</p>
<p>--------------------------------------------------------------<br />Running v2.3 on Mac OSX 10.7.3 Fairly new to Kepler.</p>
<p>I'm trying to use the EML 2 Dataset Actor to read in a local copy (in a folder on my laptop drive) of a sample metadata (.xml) file using the EML 2.1.0 schema and a simple 4-column data (.txt) file. I have checked the files using the <a class="external" href="http://knb.ecoinformatics.org/emlparser/">http://knb.ecoinformatics.org/emlparser/</a> site, and both were validated.</p>
<p>Two problems:<br />i) If I change either file the Actor doesn't keep up with the changes. It seems to revert to a cached version of the file as it was when I first specified it to the Actor (via the EML File: and Data File: fields in Configure Actor).<br />ii) I wondered about checking the "Check for latest version" checkbox. However doing this generates an error dialog with the message "null Please enter a new value:" I have to uncheck to stop the error.<br />----------------------------------------------------------------</p>
<p>So we may need a feature:<br />When the "Check for latest version" box is checked, Kepler will check local files as well.</p> Bug #5483 (New): Another Kepler GUI will nuke the ~/.kepler/cache generated by the kepler in work...https://projects.ecoinformatics.org/ecoinfo/issues/54832011-09-08T16:14:50ZJing Taotao@nceas.ucsb.edu
<p>The workflow run engine embeds a kepler at workflowrunengine/kepler. Now the workflow run engine runs the kepler by calling the API of the kepler.</p>
<p>I found an issue that if we start another kepler GUI in another location, e.g. ~/project/kepler, the kepler gui will nuke the ~/.kepler/cache directory even though they are in the same suite. This may cause a problem to run the workflow run engine. However, if we start the kepler gui from workflowrunengine/kepler, the kepler gui wouldn't nuke the cache directory generated by the run engine.</p> Bug #5474 (New): Make the eml actor on kepler handle multiple physical representations in an entityhttps://projects.ecoinformatics.org/ecoinfo/issues/54742011-08-19T16:17:49ZJing Taotao@nceas.ucsb.edu
<p>Wade Sheldon from GCE-LTER brought an question if the eml actor can handle mutiple physical representations in an entity.</p>
<p>I took a look at code and found kepler can't handle this correctly. In org.ecoinformatics.seek.datasource.eml.eml2.Eml200Parser class, we always use</p>
<p>physical/dataFormat/textFormat/simpleDelimited/collapseDelimiters, physical/distribution/online/url/@function and et al directly. And also we only get the value from the node which index is 0.</p>
<p>This will handle entity with one physical representation correctly. But it will failed when an entity has more than one physical representation.</p>
<p>For example, a dataTable has two physicals. The first one isn't textFormat and has a downloadable url. The second one is textFormat and has a downloadable url.</p>
<p>Kepler will download the first non-text data object and ignore the second text data object. And the worst case is that kepler will mismatch the non-text data object with the text format information from the second text data object.</p> Bug #5444 (In Progress): Kepler can't run workflow if two instances exist at same timehttps://projects.ecoinformatics.org/ecoinfo/issues/54442011-07-30T00:21:27ZJing Taotao@nceas.ucsb.edu
<p>Both vanilla and reporting suites (2.3) have problems. But the behaviors are different:<br />1. Vanilla kepler:</p>
<p>Start a kepler then start antoher one. The second one runs a workflow. Then close the first kepler. Finally close the second one, it shows the error:<br />java.sql.SQLException: Connection is broken: java.io.EOFException<br />at org.hsqldb.jdbc.Util.sqlException(Unknown Source)<br />at org.hsqldb.jdbc.jdbcStatement.fetchResult(Unknown Source)<br />at org.hsqldb.jdbc.jdbcStatement.execute(Unknown Source)<br />at org.kepler.util.sql.DatabaseType._executeSQL(DatabaseType.java:930)<br />at org.kepler.util.sql.HSQL.disconnect(HSQL.java:79)<br />at org.kepler.objectmanager.lsid.LSIDGenerator.closeDatabase(LSIDGenerator.java:400)<br />at org.kepler.module.core.Shutdown.shutdownModule(Shutdown.java:48)<br />at org.kepler.Kepler._initializeModules(Kepler.java:572)<br />at org.kepler.Kepler.access$100(Kepler.java:81)<br />at org.kepler.Kepler$Shutdown.shutdown(Kepler.java:663)<br />at org.kepler.util.ShutdownNotifier.shutdown(ShutdownNotifier.java:48)<br />at org.kepler.gui.KeplerGraphFrame.dispose(KeplerGraphFrame.java:1146)</p>
<p>2. Reporting kepler.<br />Two kepers are running. One runs a workflow and get an error:<br /><jing> ptolemy.kernel.util.IllegalActionException: Error recording workflow structure: Error connecting or initializing database.
{DB User Name=sa, Password=, DB Host=, DB Name=provenanceDB, DB Type=HSQL, DB Port=9002, DB Table Prefix=, JDBC URL=}<br />The database is already in use by another process: lockFile: org.hsqldb.persist.LockFile@aef9f24b[file =/home/tao/KeplerData/modules/provenance/db-2.3/provenanceDB.lck, exists=true, locked=false, valid=false, ] method: checkHeartbeat read: 2011-07-28 13:32:04.979 heartbeat - read: -8463 ms.<br />in .testWithReport.TOP Provenance Recorder<br />Because:<br />Error connecting or initializing database.
{DB User Name=sa, Password=, DB Host=, DB Name=provenanceDB, DB Type=HSQL, DB Port=9002, DB Table Prefix=, JDBC URL=}<br />The database is already in use by another process: lockFile: org.hsqldb.persist.LockFile@aef9f24b[file =/home/tao/KeplerData/modules/provenance/db-2.3/provenanceDB.lck, exists=true, locked=false, valid=false, ] method: checkHeartbeat read: 2011-07-28 13:32:04.979 heartbeat - read: -8463 ms.<br />at org.kepler.provenance.ProvenanceRecorder._recordWorkflowContents(ProvenanceRecorder.java:2618)<br />at org.kepler.provenance.ProvenanceRecorder.preinitialize(ProvenanceRecorder.java:583)<br />at ptolemy.actor.CompositeActor.preinitialize(CompositeActor.java:1682)<br />at ptolemy.actor.Manager.preinitializeAndResolveTypes(Manager.java:930)<br />at ptolemy.actor.Manager.initialize(Manager.java:636)<br />at ptolemy.actor.Manager.execute(Manager.java:340)<br />at ptolemy.actor.Manager.run(Manager.java:1111)<br />at ptolemy.actor.Manager$PtolemyRunThread.run(Manager.java:1641)</p> Bug #5423 (New): Put kepler-server war file and kepler executatlbe as part of normal installation...https://projects.ecoinformatics.org/ecoinfo/issues/54232011-06-13T21:31:38ZJing Taotao@nceas.ucsb.edu
<p>We will have a kepler-server component, it consists of three parts:<br />1. scheduler<br />2. workflow run engine <br />3. kepler itself.</p>
<p>First we should put them into a servlet war file.</p>
<p>Then will design and implement installer for these to put war and kepler<br />executable in place as part of the normal installation procedure.</p>
<p>Note: we should NOT have the duplicated kepler executables in the installer.</p>
<p>The priority order of the installer including the server components is</p>
<p>linux > mac > windows</p> Bug #5401 (New): Workflow archiving dataturbine data to metacat should check the last archiving t...https://projects.ecoinformatics.org/ecoinfo/issues/54012011-05-12T21:14:57ZJing Taotao@nceas.ucsb.edu
<p>Currently the workflow check the last archiving time through a local database.<br />If the workflow run on different machine, it will cause conflict. If we search the metadata (EML) documents in the repository and get the last archiving time by pass them, the process will be more robust. However, cost is high.</p>
<p>If this workflow only be run by a scheduler and execution engine (this means the workflow will be run on one machine), checking the last archiving time through the local DB will be fine.</p> Bug #5400 (New): User can specify the interval of data chunk in the workflow archiving dataturbi...https://projects.ecoinformatics.org/ecoinfo/issues/54002011-05-12T21:05:06ZJing Taotao@nceas.ucsb.edu
<p>In the current workflow, it chops the data from the last archiving time to latest data stamp if there is no metadata change.</p>
<p>However, the workflow should have a feature that let user to specify the interval of dataset. For example, user can specify the interval to be month. So one dataset will have the all data generated in one month, such as March. The next dataset will contain the data in April.</p> Bug #5362 (New): Create a GUI so user can easily archive a senosr's data from a dataturbine to a ...https://projects.ecoinformatics.org/ecoinfo/issues/53622011-03-31T21:39:54ZJing Taotao@nceas.ucsb.eduBug #5340 (New): Miss a dollar sing ($) when an actor references a parameter.https://projects.ecoinformatics.org/ecoinfo/issues/53402011-03-04T23:05:26ZJing Taotao@nceas.ucsb.edu
<p>On Figure 3.20 (page 66) of KeplerUserManual-2.1.0.doc, it shows how to reference a parameter OutputDir. It shows:</p>
<p>OutputDir+"ccov6190"</p>
<p>But it doesn't work in this way. It should be:</p>
<p>$OutputDir+"ccov6190"</p>
<p>A dollar sign is missed.</p> Bug #5339 (New): In Database Query actor, the output port will be string type if you choose the o...https://projects.ecoinformatics.org/ecoinfo/issues/53392011-03-04T02:05:33ZJing Taotao@nceas.ucsb.edu
<p>Derik and I found that the type of the output port is string even though we chose "record" as the output type.</p> Bug #5338 (New): In Database Query actor, the output behavior will be different if there is no re...https://projects.ecoinformatics.org/ecoinfo/issues/53382011-03-04T02:02:36ZJing Taotao@nceas.ucsb.edu
<p>If there is result set found and the output type is "array":<br />no token will be sent if you select "outputEachRowSeparately" in configuration panel;<br />no token will be send but a message "no match result for query ..." will be shown if you don't select "outputEachRowSeparately".</p>
<p>If there is result set found and the output type is "no metadata".<br />no token will be sent if you select "outputEachRowSeparately" in configuration panel;<br />A token ( I guess "/n") will be sent out if you don't select "outputEachRowSeparately".</p> Bug #5336 (New): If a user chooses "no metadata" as output format in Database Query actor, the qu...https://projects.ecoinformatics.org/ecoinfo/issues/53362011-03-04T01:45:15ZJing Taotao@nceas.ucsb.edu
<p>If user chooses the output type as "no metadata" (no column name will be shown in the result), and query looks like "select id, lastTime from sensors where id='sensor0', the result only shows:</p>
<pre><code>sensor0</code></pre>
<p>The lastTime column is missed even though the sensor0 does have a value at the lastTime column</p> Bug #5284 (New): Set uploadToServer to be true at configuration.xmlhttps://projects.ecoinformatics.org/ecoinfo/issues/52842011-01-28T17:37:51ZJing Taotao@nceas.ucsb.edu
<p>When we set up kepler run engine, we always have to change the value from false to true. Why we set it to be false? I think user always wants to upload the run kar file to the repository when they pass the repository name to the kepler.</p> Bug #1587 (In Progress): Define and implement EcoGrid "dataQuery" methodhttps://projects.ecoinformatics.org/ecoinfo/issues/15872004-06-07T16:46:10ZJing Taotao@nceas.ucsb.edu
<p>This ecogrid server side data query task. Bug 1586 will implement dataQuery in<br />kepler side (client side, query local data). In task, kepler will send a sql<br />query from kepler to ecogrid node. Ecogrid will excute this query and return<br />partial data object to client.<br />This bug will be done on the base of bug 1584.</p>