April 2010 - teiid-issues - Jboss List Archives

[JBoss JIRA] Created: (TEIID-1043) Invalid serialized attachments should be ingored.

by Steven Hawkins (JIRA)

Invalid serialized attachments should be ingored. ------------------------------------------------- Key: TEIID-1043 URL: https://jira.jboss.org/jira/browse/TEIID-1043 Project: Teiid Issue Type: Bug Affects Versions: 7.0 Reporter: Steven Hawkins Assignee: Steven Hawkins Priority: Minor Fix For: 7.0 Changing from one ms to another in the same server instance will through exceptions loading attachment information if the classes have changed. Instead errors should just be logged and the information recreated. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

2
2
0 / 0

[JBoss JIRA] Created: (TEIID-877) Membership Domain Documentation is Out of Date

by Nestor D. Rodriguez (JIRA)

Membership Domain Documentation is Out of Date ---------------------------------------------- Key: TEIID-877 URL: https://jira.jboss.org/jira/browse/TEIID-877 Project: Teiid Issue Type: Task Components: Server Affects Versions: 6.2.0 Environment: N/A Reporter: Nestor D. Rodriguez Assignee: Steven Hawkins Priority: Minor The Server Extensions Guide (membership_domain_guide.pdf) for the 6.2.0 release has not been updated for the new version and uses obsolete code in its description of the features, thus making it uselss as reference for developers. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

3
4
0 / 0

[JBoss JIRA] Created: (TEIID-1058) SessionCleanupThread causes deadlocks when server is clustered

by Jerry Helbling (JIRA)

SessionCleanupThread causes deadlocks when server is clustered -------------------------------------------------------------- Key: TEIID-1058 URL: https://jira.jboss.org/jira/browse/TEIID-1058 Project: Teiid Issue Type: Bug Components: Server Reporter: Jerry Helbling Assignee: Steven Hawkins Priority: Minor Fix For: 8.0 Note: The following problem exists in MetaMatrix 5.5.x. I am creating this Teiid JIRA so that a fix for it (if necessary) can be considered in that product. There are no plans to fix it in MetaMatrix 5.5.x. When SessionID cleanup is done, either at a scheduled interval or at server shutdown, there will sometimes be deadlocks if the MetaMatrix servers are clustered and/or have multiple processes. This is because all of the processes will do the full cleanup on whatever inactive and aged sessionids are in the database. And each of them uses a single transaction to do all of the deletes, no matter how many sessionids need to be deleted. This is hugely redundant and subject to deadlocks. If we had a concept of a 'lead' host in a cluster, it could be assigned this task, but we don't. Possible solutions: 1) Adding an ORDER BY to the SELECT we use to retrieve the old sessionids. This forces the processes to delete the sessionids in the same order, reducing the likelihood of true deadlocks. However, in testing it did not prevent the deadlocks. 2) Refactor the cleanup code so that it does a smaller amount of work per transaction. Currently each host will attempt to do the entire cleanup process in a single transaction. With a default TTL of 10 hours, that could be hundreds or thousands of deletable sessionids in a busy environment. The process could be rewritten to discover/delete/commit a few (say 10, or even just 1) at a time. 3) Add "FOR UPDATE NOWAIT" to the SELECT that is used to retrieve the sessionids that are ready to delete. But this syntax is only legal for Oracle and PostgreSQL. So that may not be such good solution. We have decided not to pursue this fix in 5.5.4. Whether the conditions will be the same in Teiid is not clear, but if they are this note is offered as a contribution toward solving it in Teiid. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

3
2
0 / 0

[JBoss JIRA] Created: (TEIID-429) Use DataSources in creating the ConnectorBindings instead of creating one using the Driver

by Ramesh Reddy (JIRA)

Use DataSources in creating the ConnectorBindings instead of creating one using the Driver ------------------------------------------------------------------------------------------ Key: TEIID-429 URL: https://jira.jboss.org/jira/browse/TEIID-429 Project: Teiid Issue Type: Task Affects Versions: 6.x Reporter: Ramesh Reddy Assignee: Ramesh Reddy Fix For: 6.0.0 Currently when we have separate configurations for XA sources and Non-XA sources, these should be combined into one definition, and by default we should make attempt use the data sources instead of Drivers. This will make it easy on the user in specifying the properties rather than coming up with unique URL for the data source. This will add DatabaseName ServerName Port properties in addition to already provided, User Password URL -> expert=true properties. The URL property will be set to "expert", so that if one needs to use the Driver, you can still use it. In addition this needs to provide support to add additional properties that may be needed by the specific data source. For this add OptionalProperties This should be supplied in the format [name=value;name=value] and runtime engines should support these to add these properties as DataSource bean properties. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

3
10
0 / 0

[JBoss JIRA] Created: (TEIID-921) Queries containing a UNION with an ORDER BY are producing unexpected results in CTC Query Testing

by Warren Gibson (JIRA)

Queries containing a UNION with an ORDER BY are producing unexpected results in CTC Query Testing -------------------------------------------------------------------------------------------------- Key: TEIID-921 URL: https://jira.jboss.org/jira/browse/TEIID-921 Project: Teiid Issue Type: Bug Components: Integration Tests Affects Versions: 7.0 Reporter: Warren Gibson Assignee: Steven Hawkins I am getting unexpected results (all sources) on 17 queries containing a UNION and ORDER BY. The ORDER BY isn't working when there is a UNION involved. These queries were producing expected results prior to the holidays. Below is a sample query: RDBMS_Test_Queries_6500_TestQuery32" value="SELECT StringKey FROM BQT1.SmallA WHERE IntKey >= 0 AND IntKey < 15 UNION SELECT StringKey FROM BQT1.SmallB WHERE IntKey >= 10 AND IntKey < 30 ORDER BY StringKey DESC The interesting thing is the queries do work with a UNION ALL. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

3
2
0 / 0

[JBoss JIRA] Created: (TEIID-790) Teiid Server uses Sun JDK specific MX server, thus portability to other JDKs is not possible

by Ramesh Reddy (JIRA)

Teiid Server uses Sun JDK specific MX server, thus portability to other JDKs is not possible -------------------------------------------------------------------------------------------- Key: TEIID-790 URL: https://jira.jboss.org/jira/browse/TEIID-790 Project: Teiid Issue Type: Task Components: Server Affects Versions: 6.2.0 Reporter: Ramesh Reddy Priority: Critical Fix For: 6.3.0 Teiid Server assumes the JMX server is the one that is provided by the SUN JDK, thus if user uses other JDK implementations like Iced-tea it will not be supported. This dependency must be removed. Possibly need to use other independent open source based JMX servers like MX4J or JBoss MC. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

4
3
0 / 0

[JBoss JIRA] Created: (TEIID-881) SOAP Connector should expose property to control HTTP Header Transfer-Encoding: chunked

by Larry O'Leary (JIRA)

SOAP Connector should expose property to control HTTP Header Transfer-Encoding: chunked --------------------------------------------------------------------------------------- Key: TEIID-881 URL: https://jira.jboss.org/jira/browse/TEIID-881 Project: Teiid Issue Type: Feature Request Components: XML Connector Affects Versions: 6.2.0, 6.1.0, 6.0.0 Reporter: Larry O'Leary Assignee: Steven Hawkins A user has reported that results from a SOAP service using the SOAP-Relational connector are sometimes invalid. This is due to the results from an earlier invocation being returned. After further investigation this turned out to be a bug with the web server that was hosting the web service and its inability to properly handle chunked requests. By default, Axis 1.3/2 (not sure what we are using lately) uses chunked Transfer-Encoding. By disabling chunked transfer encoding it allows the faulty web server to properly serve the request and return the expected results. Ideally, chunked requests should be used as per the default Axis configuration but we should expose a means of overriding this behaviour in the event a user runs across a buggy web server. For example, in SOAPExecutor chunked requests can be enabled/disabled via a setting in the Call properties. Boolean isChunked = ((HTTPConnectorState) this.getState()).isHttpHeaderTransferEncodingChunked(); if (isChunked != null) { Hashtable httpProps = (Hashtable) call.getProperty(HTTPConstants.REQUEST_HEADERS); if (httpProps == null) { httpProps = new Hashtable(); } httpProps.put(HTTPConstants.HEADER_TRANSFER_ENCODING_CHUNKED, isChunked); call.setProperty(HTTPConstants.REQUEST_HEADERS, httpProps); } Please note, this code was written against Axis 1.3 which is used in XML-Relational SOAP Connector for MMx 5.5.3. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

4
6
0 / 0

[JBoss JIRA] Created: (TEIID-1019) Request ability to turn on/off Auto Staging tables

by Debbie Steigner (JIRA)

Request ability to turn on/off Auto Staging tables -------------------------------------------------- Key: TEIID-1019 URL: https://jira.jboss.org/jira/browse/TEIID-1019 Project: Teiid Issue Type: Feature Request Components: Server, XML Planner/Processor Reporter: Debbie Steigner Assignee: Steven Hawkins Fix For: 7.0 Please add the ability to turn on/off Auto Staging tables. Setting the Cardinality to 0 is not always a workaround if the customer wants the cardinality set for better query planning on other models. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

4
9
0 / 0

[JBoss JIRA] Created: (TEIID-940) COUNT function in Postgres push down producting exception

by Warren Gibson (JIRA)

COUNT function in Postgres push down producting exception --------------------------------------------------------- Key: TEIID-940 URL: https://jira.jboss.org/jira/browse/TEIID-940 Project: Teiid Issue Type: Bug Components: Query Engine Affects Versions: 7.0 Reporter: Warren Gibson Assignee: Steven Hawkins Postgres push down is producing unexpected CTC results with queries containing COUNT(*). Example: RDBMS_Test_Queries_4500_TestQuery69" value="SELECT COUNT(*) FROM BQT1.SmallA Error: com.metamatrix.api.exception.MetaMatrixComponentException SQLState: 38000 ErrorCode: 0 -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

3
3
0 / 0

[JBoss JIRA] Created: (TEIID-878) Fix capabilities caching

by Steven Hawkins (JIRA)

Fix capabilities caching ------------------------ Key: TEIID-878 URL: https://jira.jboss.org/jira/browse/TEIID-878 Project: Teiid Issue Type: Task Components: Query Engine Affects Versions: 6.0.0 Reporter: Steven Hawkins Assignee: Steven Hawkins Fix For: 6.3 The caching of connector capabilities should be closer to the data layer to ensure that capabilities changes are picked up with connector restarts irregardless of vdb state. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

14 years, 7 months

2
2
0 / 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

teiid-issues April 2010