[JBoss JIRA] (JBTM-2090) BlackTie process terminating with default action of signal 13 (SIGPIPE)
by Tom Jenkinson (JIRA)
[ https://issues.jboss.org/browse/JBTM-2090?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson updated JBTM-2090:
--------------------------------
Fix Version/s: 5.x.y
(was: 5.0.5)
> BlackTie process terminating with default action of signal 13 (SIGPIPE)
> -----------------------------------------------------------------------
>
> Key: JBTM-2090
> URL: https://issues.jboss.org/browse/JBTM-2090
> Project: JBoss Transaction Manager
> Issue Type: Bug
> Components: BlackTie
> Reporter: Tom Jenkinson
> Assignee: Amos Feng
> Labels: linux64el6, linux64el7
> Fix For: 5.x.y
>
>
> http://172.17.131.2/view/Narayana+BlackTie/job/narayana/419/PROFILE=BLACK...
> {code}
> ==11776== Process terminating with default action of signal 13 (SIGPIPE)
> ==11776== at 0x72734ED: ??? (in /lib64/libpthread-2.12.so)
> ==11776== by 0x4E4C136: apr_socket_send (in /usr/lib64/libapr-1.so.0.3.9)
> ==11776== by 0x54F05CB: HybridSocketEndpointQueue::_send(char const*, long, long, char*, int, long, long, char const*, char const*) (HybridSocketEndpointQueue.cxx:305)
> ==11776== by 0x54F081F: HybridSocketEndpointQueue::send(char const*, long, long, char*, int, long, long, char const*, char const*) (HybridSocketEndpointQueue.cxx:347)
> ==11776== by 0x54EB35C: HybridSocketSessionImpl::send(message_t&) (HybridSocketSessionImpl.cxx:536)
> ==11776== by 0x6D1894D: send(Session*, char const*, char*, long, int, long, long, message_t&, long, int, long, bool, char*) (XATMIc.cxx:147)
> ==11776== by 0x6D18D4F: send(Session*, char const*, char*, long, int, long, long, long, int, long, bool, char*) (XATMIc.cxx:183)
> ==11776== by 0x6D24122: tpdiscon (XATMIc.cxx:1220)
> ==11776== by 0x497211: TestTPConnect::tearDown() (TestTPConnect.cxx:66)
> ==11776== by 0x49B74A: CppUnit::TestCaller<TestTPConnect>::tearDown() (TestCaller.h:182)
> ==11776== by 0x3707A29B36: CppUnit::TestCaseMethodFunctor::operator()() const (TestCase.cpp:32)
> ==11776== by 0x3707A1BEA3: CppUnit::DefaultProtector::protect(CppUnit::Functor const&, CppUnit::ProtectorContext const&) (DefaultProtector.cpp:15)
> ==11776== by 0x3707A25C18: CppUnit::ProtectorChain::ProtectFunctor::operator()() const (ProtectorChain.cpp:20)
> ==11776== by 0x3707A2595B: CppUnit::ProtectorChain::protect(CppUnit::Functor const&, CppUnit::ProtectorContext const&) (ProtectorChain.cpp:77)
> ==11776== by 0x3707A3164F: CppUnit::TestResult::protect(CppUnit::Functor const&, CppUnit::Test*, std::string const&) (TestResult.cpp:178)
> ==11776== by 0x3707A2989B: CppUnit::TestCase::run(CppUnit::TestResult*) (TestCase.cpp:97)
> ==11776== by 0x3707A2A10B: CppUnit::TestComposite::doRunChildTests(CppUnit::TestResult*) (TestComposite.cpp:64)
> ==11776== by 0x3707A2A035: CppUnit::TestComposite::run(CppUnit::TestResult*) (TestComposite.cpp:23)
> ==11776== by 0x3707A2A10B: CppUnit::TestComposite::doRunChildTests(CppUnit::TestResult*) (TestComposite.cpp:64)
> ==11776== by 0x3707A2A035: CppUnit::TestComposite::run(CppUnit::TestResult*) (TestComposite.cpp:23)
> ==11776== by 0x3707A31429: CppUnit::TestResult::runTest(CppUnit::Test*) (TestResult.cpp:145)
> ==11776== by 0x3707A33A21: CppUnit::TestRunner::run(CppUnit::TestResult&, std::string const&) (TestRunner.cpp:96)
> ==11776== by 0x3707A369CA: CppUnit::TextTestRunner::run(std::string, bool, bool, bool) (TextTestRunner.cpp:64)
> ==11776== by 0x4F6F83: main (TestRunner.cxx:27)
> --11776-- Discarding syms at 0x973f480-0x973fec8 in /usr/lib64/gconv/ISO8859-1.so due to munmap()
> --11776-- Discarding syms at 0x953b580-0x953ccd8 in /usr/lib64/gconv/UTF-16.so due to munmap()
> --11776-- Discarding syms at 0xc9461f0-0xc94e648 in /lib64/libnss_files-2.12.so due to munmap()
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
9 years, 10 months
[JBoss JIRA] (JBTM-2316) Add missing ArjunaJTS/jtax and ArjunaJTA/jdbc report to code-coverage
by Tom Jenkinson (JIRA)
[ https://issues.jboss.org/browse/JBTM-2316?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson updated JBTM-2316:
--------------------------------
Summary: Add missing ArjunaJTS/jtax and ArjunaJTA/jdbc report to code-coverage (was: Add missing ArjunaJTS/jtax and ArjunaJTA/jdbc report to code-coverage and remove performance tool)
> Add missing ArjunaJTS/jtax and ArjunaJTA/jdbc report to code-coverage
> ---------------------------------------------------------------------
>
> Key: JBTM-2316
> URL: https://issues.jboss.org/browse/JBTM-2316
> Project: JBoss Transaction Manager
> Issue Type: Task
> Components: Testing
> Reporter: Tom Jenkinson
> Assignee: Michael Musgrove
> Fix For: 5.0.5
>
>
> I can't see the transactional driver or JTAx stuff in the code-coverage report and I also don't think its worth aggregating in the performance test harness into the coverage numbers.
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
9 years, 10 months
[JBoss JIRA] (JBTM-2329) Add a management API that an AbstractRecord provider can implement to clear the heuristic state of a transaction
by Tom Jenkinson (JIRA)
[ https://issues.jboss.org/browse/JBTM-2329?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson updated JBTM-2329:
--------------------------------
Forum Reference: https://developer.jboss.org/message/912282
> Add a management API that an AbstractRecord provider can implement to clear the heuristic state of a transaction
> ----------------------------------------------------------------------------------------------------------------
>
> Key: JBTM-2329
> URL: https://issues.jboss.org/browse/JBTM-2329
> Project: JBoss Transaction Manager
> Issue Type: Feature Request
> Components: Transaction Core
> Reporter: Tom Jenkinson
> Assignee: Tom Jenkinson
>
> If an AbstractRecord type has resulted in a heuristic state, it would be useful to provide some form of management API to clear the state on the resource manager.
> After discussion on the forum, it seems likely the API would be something like:
> Update AbstractRecord to change the return type of value() to the HeuristicManagement interface (rather than Object). Although value() is maybe not the most accurate term and it does appear that this is used internally by XAResourceRecord in an unintended manner (i.e. not dealing with HeuristicInformation.
> Add a new interface of modify the existing HeuristicInformation (the current version of which is to be deprecated in: JBTM-2328)
> {code}
> public interface HeuristicManagement
> {
> public void resolve() throws CouldNotResolveException;
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
9 years, 10 months
[JBoss JIRA] (JBTM-2330) Document the procedure to discover and resolve heuristically completed branches in Narayana
by Tom Jenkinson (JIRA)
Tom Jenkinson created JBTM-2330:
-----------------------------------
Summary: Document the procedure to discover and resolve heuristically completed branches in Narayana
Key: JBTM-2330
URL: https://issues.jboss.org/browse/JBTM-2330
Project: JBoss Transaction Manager
Issue Type: Feature Request
Components: Documentation
Reporter: Tom Jenkinson
Assignee: Michael Musgrove
Our documentation is not complete in this area. We need to describe how users can detect heuristically completed branches in the system. It would be useful to consider this in standalone and embedded wildfly modes.
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
9 years, 10 months
[JBoss JIRA] (JBTM-2329) Add a management API that an AbstractRecord provider can implement to clear the heuristic state of a transaction
by Tom Jenkinson (JIRA)
Tom Jenkinson created JBTM-2329:
-----------------------------------
Summary: Add a management API that an AbstractRecord provider can implement to clear the heuristic state of a transaction
Key: JBTM-2329
URL: https://issues.jboss.org/browse/JBTM-2329
Project: JBoss Transaction Manager
Issue Type: Feature Request
Components: Transaction Core
Reporter: Tom Jenkinson
Assignee: Tom Jenkinson
If an AbstractRecord type has resulted in a heuristic state, it would be useful to provide some form of management API to clear the state on the resource manager.
After discussion on the forum, it seems likely the API would be something like:
Update AbstractRecord to change the return type of value() to the HeuristicManagement interface (rather than Object). Although value() is maybe not the most accurate term and it does appear that this is used internally by XAResourceRecord in an unintended manner (i.e. not dealing with HeuristicInformation.
Add a new interface of modify the existing HeuristicInformation (the current version of which is to be deprecated in: JBTM-2328)
{code}
public interface HeuristicManagement
{
public void resolve() throws CouldNotResolveException;
}
{code}
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
9 years, 10 months