[JBoss JIRA] (JGRP-1797) NakackTest testReceptionOfAllMessages fails to receive all messages
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1797?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz edited comment on JGRP-1797 at 2/23/14 7:55 PM:
--------------------------------------------------------------------
There is something very wrong with this test. Sending messages seems to happen correctly but receiving messages does not.
Some issues:
- when I add logging to the receivers, no output is generated, which means that no messages are being received - and this happens when the test passes too!
- if I get the channels to connect after setting the receiver, I see different behaviour (if I run the test case as is, I see no messages received; if I connect after setting the receivers, I see messages received by C only)
Still looking into it...
was (Author: rachmato):
There is something very wrong with this test. Sending messages seems to happen correctly but receiving messages does not.
Some issues:
- when I add logging to the receivers, no output is generated, which means that no messages are being received - and this happens when the test passes too!
- if I get the channels to connect after setting the receiver, I see different behaviour
Still looking into it...
> NakackTest testReceptionOfAllMessages fails to receive all messages
> -------------------------------------------------------------------
>
> Key: JGRP-1797
> URL: https://issues.jboss.org/browse/JGRP-1797
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Environment: RHEL (5/24), Solaris (6/24), HPUX (1/2) where x/y means x failures over y test executions.
> Reporter: Richard Achmatowicz
> Assignee: Bela Ban
> Fix For: 3.2.13
>
> Attachments: stdout.txt
>
>
> This test does the following:
> - involves channels A, B, and C where only B and C are senders and all are receivers
> - B and C send 1000 multicast messages
> - checks that all messages are received and in the correct order
> This test is failing intermittently but quite often.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (DROOLS-439) Package builder unable to find declared array class
by Davide Sottara (JIRA)
[ https://issues.jboss.org/browse/DROOLS-439?page=com.atlassian.jira.plugin... ]
Davide Sottara commented on DROOLS-439:
---------------------------------------
It is indeed a bug, thanks for reporting!
Tentative fix here : https://github.com/sotty/drools/commit/5e28d652e6310e570b33ee011d17f904ef...
> Package builder unable to find declared array class
> ---------------------------------------------------
>
> Key: DROOLS-439
> URL: https://issues.jboss.org/browse/DROOLS-439
> Project: Drools
> Issue Type: Bug
> Security Level: Public(Everyone can see)
> Affects Versions: 6.0.1.Final
> Reporter: Mark Sullivan
> Assignee: Mark Proctor
>
> When declaring a fact type with an array field, where the array type is declared in the same package, for example:
> {code}
> declare Owner
> name : String
> end
> declare Pet
> owners : Owner[]
> end
> {code}
> The following build error occurs:
> {code}
> [Message [id=1, level=ERROR, path=pets.drl, line=5, column=0
> text=Unable to find class 'Owner']]
> {code}
> This error doesn't always occur when an array field is declared. It's dependent on the order in which each fact type is processed.
> PackageBuilder attempts to order fact types for processing based on the declared super type and attribute types.
> The following loop in *PackageBuilder.sortByHierarchy()* processes each field looking for field type dependencies:
> {code}
> for (TypeFieldDescr field : tdescr.getFields().values()) {
> QualifiedName typeName = new QualifiedName(field.getPattern().getObjectType());
> if (!hasCircularDependency(name, typeName, taxonomy)) {
> supers.add(typeName);
> }
> }
> {code}
> However I don't think it correctly spots the dependency when the field type is an array.
> This issue can be worked around by declaring an unused field of the same type as the array:
> {code}
> declare Owner
> name : String
> end
> declare Pet
> owners : Owner[]
> _unused: Owner
> end
> {code}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1797) NakackTest testReceptionOfAllMessages fails to receive all messages
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1797?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz updated JGRP-1797:
--------------------------------------
Attachment: stdout.txt
Sample of test run with "Got message" in receiver.
> NakackTest testReceptionOfAllMessages fails to receive all messages
> -------------------------------------------------------------------
>
> Key: JGRP-1797
> URL: https://issues.jboss.org/browse/JGRP-1797
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Environment: RHEL (5/24), Solaris (6/24), HPUX (1/2) where x/y means x failures over y test executions.
> Reporter: Richard Achmatowicz
> Assignee: Bela Ban
> Fix For: 3.2.13
>
> Attachments: stdout.txt
>
>
> This test does the following:
> - involves channels A, B, and C where only B and C are senders and all are receivers
> - B and C send 1000 multicast messages
> - checks that all messages are received and in the correct order
> This test is failing intermittently but quite often.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1797) NakackTest testReceptionOfAllMessages fails to receive all messages
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1797?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz commented on JGRP-1797:
-------------------------------------------
There is something very wrong with this test. Sending messages seems to happen correctly but receiving messages does not.
Some issues:
- when I add logging to the receivers, no output is generated, which means that no messages are being received - and this happens when the test passes too!
- if I get the channels to connect after setting the receiver, I see different behaviour
Still looking into it...
> NakackTest testReceptionOfAllMessages fails to receive all messages
> -------------------------------------------------------------------
>
> Key: JGRP-1797
> URL: https://issues.jboss.org/browse/JGRP-1797
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Environment: RHEL (5/24), Solaris (6/24), HPUX (1/2) where x/y means x failures over y test executions.
> Reporter: Richard Achmatowicz
> Assignee: Bela Ban
> Fix For: 3.2.13
>
>
> This test does the following:
> - involves channels A, B, and C where only B and C are senders and all are receivers
> - B and C send 1000 multicast messages
> - checks that all messages are received and in the correct order
> This test is failing intermittently but quite often.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1797) NakackTest testReceptionOfAllMessages fails to receive all messages
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1797?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz updated JGRP-1797:
--------------------------------------
Description:
This test does the following:
- involves channels A, B, and C where only B and C are senders and all are receivers
- B and C send 1000 multicast messages
- checks that all messages are received and in the correct order
This test is failing intermittently but quite often.
was:
This test does the following:
- sends multicast messages from senders
- checks that all messages are received and in the correct order
This test is failing intermittently but quite often.
> NakackTest testReceptionOfAllMessages fails to receive all messages
> -------------------------------------------------------------------
>
> Key: JGRP-1797
> URL: https://issues.jboss.org/browse/JGRP-1797
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Environment: RHEL (5/24), Solaris (6/24), HPUX (1/2) where x/y means x failures over y test executions.
> Reporter: Richard Achmatowicz
> Assignee: Bela Ban
> Fix For: 3.2.13
>
>
> This test does the following:
> - involves channels A, B, and C where only B and C are senders and all are receivers
> - B and C send 1000 multicast messages
> - checks that all messages are received and in the correct order
> This test is failing intermittently but quite often.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1797) NakackTest testReceptionOfAllMessages fails to receive all messages
by Richard Achmatowicz (JIRA)
Richard Achmatowicz created JGRP-1797:
-----------------------------------------
Summary: NakackTest testReceptionOfAllMessages fails to receive all messages
Key: JGRP-1797
URL: https://issues.jboss.org/browse/JGRP-1797
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2.13
Environment: RHEL (5/24), Solaris (6/24), HPUX (1/2) where x/y means x failures over y test executions.
Reporter: Richard Achmatowicz
Assignee: Bela Ban
Fix For: 3.2.13
This test does the following:
- sends multicast messages from senders
- checks that all messages are received and in the correct order
This test is failing intermittently but quite often.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1796) Fix flush problem with integration of commons-logging output
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1796?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz commented on JGRP-1796:
-------------------------------------------
I've just linked the PR for 3.2.13.Final.
> Fix flush problem with integration of commons-logging output
> ------------------------------------------------------------
>
> Key: JGRP-1796
> URL: https://issues.jboss.org/browse/JGRP-1796
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Reporter: Richard Achmatowicz
> Assignee: Richard Achmatowicz
> Priority: Minor
> Fix For: 3.2.13, 3.3.6, 3.5
>
>
> The commons-logging output which appears in test case output files does not have line termination processed correctly. We end up with text like this:
> {noformat}
> 120665 [TRACE] GMS: - A: no initial members discovered: creating group as first member120666 [DEBUG] GMS: - A: installing view [A|0] [A]120666 [DEBUG] GMS: - created group (first member). My view is [A|0], impl is org.jgroups.protocols.pbcast.CoordGmsImpl
> {noformat}
> This makes reading the logging output hard to read.
> We want it to look like this:
> {noformat}
> 120665 [TRACE] GMS: - A: no initial members discovered: creating group as first member
> 120666 [DEBUG] GMS: - A: installing view [A|0] [A]
> 120666 [DEBUG] GMS: - created group (first member). My view is [A|0], impl is org.jgroups.protocols.pbcast.CoordGmsImpl
> {noformat}
> Th solution is to implement flush() as a method which simply appends a null string with CR/LF applied as well to the current collected sysout and syserr.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1796) Fix flush problem with integration of commons-logging output
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1796?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz updated JGRP-1796:
--------------------------------------
Git Pull Request: https://github.com/belaban/JGroups/pull/125
> Fix flush problem with integration of commons-logging output
> ------------------------------------------------------------
>
> Key: JGRP-1796
> URL: https://issues.jboss.org/browse/JGRP-1796
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Reporter: Richard Achmatowicz
> Assignee: Richard Achmatowicz
> Priority: Minor
> Fix For: 3.2.13, 3.3.6, 3.5
>
>
> The commons-logging output which appears in test case output files does not have line termination processed correctly. We end up with text like this:
> {noformat}
> 120665 [TRACE] GMS: - A: no initial members discovered: creating group as first member120666 [DEBUG] GMS: - A: installing view [A|0] [A]120666 [DEBUG] GMS: - created group (first member). My view is [A|0], impl is org.jgroups.protocols.pbcast.CoordGmsImpl
> {noformat}
> This makes reading the logging output hard to read.
> We want it to look like this:
> {noformat}
> 120665 [TRACE] GMS: - A: no initial members discovered: creating group as first member
> 120666 [DEBUG] GMS: - A: installing view [A|0] [A]
> 120666 [DEBUG] GMS: - created group (first member). My view is [A|0], impl is org.jgroups.protocols.pbcast.CoordGmsImpl
> {noformat}
> Th solution is to implement flush() as a method which simply appends a null string with CR/LF applied as well to the current collected sysout and syserr.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JGRP-1796) Fix flush problem with integration of commons-logging output
by Richard Achmatowicz (JIRA)
[ https://issues.jboss.org/browse/JGRP-1796?page=com.atlassian.jira.plugin.... ]
Richard Achmatowicz edited comment on JGRP-1796 at 2/23/14 4:13 PM:
--------------------------------------------------------------------
I'm sure this was working in 2008 when I first implemented it!
was (Author: rachmato):
I'm sure this was fixed in 2008 when I first implemented it!
> Fix flush problem with integration of commons-logging output
> ------------------------------------------------------------
>
> Key: JGRP-1796
> URL: https://issues.jboss.org/browse/JGRP-1796
> Project: JGroups
> Issue Type: Bug
> Affects Versions: 3.2.13
> Reporter: Richard Achmatowicz
> Assignee: Richard Achmatowicz
> Priority: Minor
> Fix For: 3.2.13, 3.3.6, 3.5
>
>
> The commons-logging output which appears in test case output files does not have line termination processed correctly. We end up with text like this:
> {noformat}
> 120665 [TRACE] GMS: - A: no initial members discovered: creating group as first member120666 [DEBUG] GMS: - A: installing view [A|0] [A]120666 [DEBUG] GMS: - created group (first member). My view is [A|0], impl is org.jgroups.protocols.pbcast.CoordGmsImpl
> {noformat}
> This makes reading the logging output hard to read.
> We want it to look like this:
> {noformat}
> 120665 [TRACE] GMS: - A: no initial members discovered: creating group as first member
> 120666 [DEBUG] GMS: - A: installing view [A|0] [A]
> 120666 [DEBUG] GMS: - created group (first member). My view is [A|0], impl is org.jgroups.protocols.pbcast.CoordGmsImpl
> {noformat}
> Th solution is to implement flush() as a method which simply appends a null string with CR/LF applied as well to the current collected sysout and syserr.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months