[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2004-01-15 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 20:06
Message generated for change (Settings changed) made by cazzius
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossCX
Group: v3.2
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

Comment By: Claudio Vesco (cazzius)
Date: 2004-01-15 16:20

Message:
Logged In: YES 
user_id=211618

fixed on Branch_3_2

--

Comment By: Elias Ross (genman)
Date: 2004-01-12 18:43

Message:
Logged In: YES 
user_id=556458


Is there any chance you can fix this for the next 3.2 release?

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 13:27

Message:
Logged In: YES 
user_id=211618

I agree with you, Adrian.

I think that registering and unregistering a JDBC Driver can be 
done by a MBean. This is what is done already by 
org.jboss.jdbc.HypersonicDatabase, for example :-)

But I propose another thing :-)

In Connector spec (1.1 or 1.5) if you want to initialize the 
resource adapter you can do the initialization in 
ManagedConnectionFactory.createConnectionFactory but 
there is no a callback to deinitialize the resource adapter.

I propose to add a shutdown (jboss only, obviously not in 
javax.resource.spi.ManagedConnectionFactory) to deinitialize 
the resource adapter.

This is not only for a Database MCF, but can be applied to 
every resource adapter if the developer/deployer want it.

In this moment I run the testsuite (in jboss head) with this 
semantic, I am waiting to see what happens :-)

If you don't like this proposal, please reassign to you this bug 
id.

Claudio

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:25

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:24

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 10:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.


[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2004-01-15 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 20:06
Message generated for change (Comment added) made by cazzius
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossCX
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Claudio Vesco (cazzius)
Date: 2004-01-15 16:20

Message:
Logged In: YES 
user_id=211618

fixed on Branch_3_2

--

Comment By: Elias Ross (genman)
Date: 2004-01-12 18:43

Message:
Logged In: YES 
user_id=556458


Is there any chance you can fix this for the next 3.2 release?

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 13:27

Message:
Logged In: YES 
user_id=211618

I agree with you, Adrian.

I think that registering and unregistering a JDBC Driver can be 
done by a MBean. This is what is done already by 
org.jboss.jdbc.HypersonicDatabase, for example :-)

But I propose another thing :-)

In Connector spec (1.1 or 1.5) if you want to initialize the 
resource adapter you can do the initialization in 
ManagedConnectionFactory.createConnectionFactory but 
there is no a callback to deinitialize the resource adapter.

I propose to add a shutdown (jboss only, obviously not in 
javax.resource.spi.ManagedConnectionFactory) to deinitialize 
the resource adapter.

This is not only for a Database MCF, but can be applied to 
every resource adapter if the developer/deployer want it.

In this moment I run the testsuite (in jboss head) with this 
semantic, I am waiting to see what happens :-)

If you don't like this proposal, please reassign to you this bug 
id.

Claudio

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:25

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:24

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 10:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.

---

[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2004-01-12 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 12:06
Message generated for change (Comment added) made by genman
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossCX
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Elias Ross (genman)
Date: 2004-01-12 10:43

Message:
Logged In: YES 
user_id=556458


Is there any chance you can fix this for the next 3.2 release?

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 05:27

Message:
Logged In: YES 
user_id=211618

I agree with you, Adrian.

I think that registering and unregistering a JDBC Driver can be 
done by a MBean. This is what is done already by 
org.jboss.jdbc.HypersonicDatabase, for example :-)

But I propose another thing :-)

In Connector spec (1.1 or 1.5) if you want to initialize the 
resource adapter you can do the initialization in 
ManagedConnectionFactory.createConnectionFactory but 
there is no a callback to deinitialize the resource adapter.

I propose to add a shutdown (jboss only, obviously not in 
javax.resource.spi.ManagedConnectionFactory) to deinitialize 
the resource adapter.

This is not only for a Database MCF, but can be applied to 
every resource adapter if the developer/deployer want it.

In this moment I run the testsuite (in jboss head) with this 
semantic, I am waiting to see what happens :-)

If you don't like this proposal, please reassign to you this bug 
id.

Claudio

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 04:25

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 04:24

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 02:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.

--

Comment By: Elias Ross (genman)
Date: 2003-11-20 10:11

Message:
Logged In: YES 
user_id=556458


The problem is the HSQLMBean isn't used for the cached D

[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-12-02 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 20:06
Message generated for change (Comment added) made by cazzius
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossCX
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 13:27

Message:
Logged In: YES 
user_id=211618

I agree with you, Adrian.

I think that registering and unregistering a JDBC Driver can be 
done by a MBean. This is what is done already by 
org.jboss.jdbc.HypersonicDatabase, for example :-)

But I propose another thing :-)

In Connector spec (1.1 or 1.5) if you want to initialize the 
resource adapter you can do the initialization in 
ManagedConnectionFactory.createConnectionFactory but 
there is no a callback to deinitialize the resource adapter.

I propose to add a shutdown (jboss only, obviously not in 
javax.resource.spi.ManagedConnectionFactory) to deinitialize 
the resource adapter.

This is not only for a Database MCF, but can be applied to 
every resource adapter if the developer/deployer want it.

In this moment I run the testsuite (in jboss head) with this 
semantic, I am waiting to see what happens :-)

If you don't like this proposal, please reassign to you this bug 
id.

Claudio

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:25

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:24

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 10:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.

--

Comment By: Elias Ross (genman)
Date: 2003-11-20 18:11

Message:
Logged In: YES 
user_id=556458


The problem is the HSQLMBean isn't used for the cached DS
configuration, it is just used to provide a HSQL DB
TCP/IP-enabled datasource.  There are three HSQL operating
modes:  in memory only, on disk, and network-enabled.

Anyway, the datasource configuration for JMS uses on disk,
which 

[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-12-02 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 20:06
Message generated for change (Comment added) made by ejort
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossCX
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:25

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:24

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 10:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.

--

Comment By: Elias Ross (genman)
Date: 2003-11-20 18:11

Message:
Logged In: YES 
user_id=556458


The problem is the HSQLMBean isn't used for the cached DS
configuration, it is just used to provide a HSQL DB
TCP/IP-enabled datasource.  There are three HSQL operating
modes:  in memory only, on disk, and network-enabled.

Anyway, the datasource configuration for JMS uses on disk,
which means the HSQLMBean service isn't configured or
deployed.  Take a look at this deployed file: 
$JBOSS/server/default/deploy/hsqldb-ds.xml

The default configuration is broken.

It might be possible to have the HSQLMBean take care of both
issuing SHUTDOWN to a HSQL cached table configuration and
the network configuration.  However, looking at the source,
it didn't seem elegant to write it as such.


--

Comment By: Heiko W.Rupp (pilhuhn)
Date: 2003-11-20 17:07

Message:
Logged In: YES 
user_id=217112

Shouldn't / couldn't that go into the stop() method of the
HSQLMBean? That is called by the server anyway and could do
the  right thing.
I think it is wrong to have two Mbeans for one service.

--

Comment By: Elias Ross (genman)
Date: 2003-11-18 23:18

Message:
Logged In: YES 
user_id=556458


I wrote an MBean to do this and it works okay.  I don't know
if it would be better to update the connection manager (RA
deployer???) or make peopl

[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-12-02 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 20:06
Message generated for change (Comment added) made by ejort
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossCX
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Adrian Brock (ejort)
Date: 2003-12-02 12:24

Message:
Logged In: YES 
user_id=9459

Hi Guys,

I disgree with the assumption that MCF <-> Database
There can be other services using the database that do not
go through
the ManagedConnectionFactory.
There might even be two deployments over the same database.

IMHO the use of DriverClass on the jdbc 
LocalManagedConnectionFactory is too cute.

What should really happen is that a separate MBean is
responsible
for registering (and unregistering) the jdbc driver.

For backwards compatibility (and cuteness), 
we can alter ConnectionFactoryTemplate.xsl 
to instantiate that MBean
if it sees 

Finally, to solve this problem an hsql specific version of
the MBean can add the extra processing in
stopService()
This would be a specific MBean deployment rather
than specifying  on the connection factory.

Regards,
Adrian

--

Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 10:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.

--

Comment By: Elias Ross (genman)
Date: 2003-11-20 18:11

Message:
Logged In: YES 
user_id=556458


The problem is the HSQLMBean isn't used for the cached DS
configuration, it is just used to provide a HSQL DB
TCP/IP-enabled datasource.  There are three HSQL operating
modes:  in memory only, on disk, and network-enabled.

Anyway, the datasource configuration for JMS uses on disk,
which means the HSQLMBean service isn't configured or
deployed.  Take a look at this deployed file: 
$JBOSS/server/default/deploy/hsqldb-ds.xml

The default configuration is broken.

It might be possible to have the HSQLMBean take care of both
issuing SHUTDOWN to a HSQL cached table configuration and
the network configuration.  However, looking at the source,
it didn't seem elegant to write it as such.


--

Comment By: Heiko W.Rupp (pilhuhn)
Date: 2003-11-20 17:07

Message:
Logged In: YES 
user_id=217112

Shouldn't / couldn't that go into the stop() method of the
HSQLMBean? That is called by the server anyway and could do
the  right thing.
I think it is wrong to have two Mbeans for one service.

--

Comment By: Elias Ross (genman)
Date: 2003-11-18 23:18

Message:
Logged In: YES 
user_id=556458


I wrote an MBean to do this and it works okay.  I don't know
if it would be better to update the connection manager (RA
deployer???) or make people use this externally declared
MBean instead.  I'm starting to to think this is the wrong
approach:


   
  ...
  jboss:service=HSQLShutdown
   

   
   
 jdbc:hsqldb:${jboss.server.data.dir}/jms
 org.hsqldb.jdbcDriver
 sa
 
 SHUTDOWN COMPACT
   



--

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
JBoss-Development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development


[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-12-02 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 20:06
Message generated for change (Comment added) made by cazzius
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

>Category: JBossCX
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
>Assigned to: Claudio Vesco (cazzius)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Claudio Vesco (cazzius)
Date: 2003-12-02 10:49

Message:
Logged In: YES 
user_id=211618

In Connector spec there is no a clean way to shutdown a 
ManagedConnectionFactory.

I propose to add to RARDeployer MBean the 
ManagedConnectionFactory shutdown handling.

For example:

public interface ShutdownHook {
   void shutdown(ManagedConnectionFactory mcf);
}

shutdown(mcf) method can be invoked in stopService() of 
RARDeployer MBean.

For the HSQL problem, when we have this hook then we can 
implement a class which send SHUTDOWN COMPACT to HSQL.

I'll commit it in jboss head branch and if there are no 
problems, in jboss 3.2 branch.

--

Comment By: Elias Ross (genman)
Date: 2003-11-20 18:11

Message:
Logged In: YES 
user_id=556458


The problem is the HSQLMBean isn't used for the cached DS
configuration, it is just used to provide a HSQL DB
TCP/IP-enabled datasource.  There are three HSQL operating
modes:  in memory only, on disk, and network-enabled.

Anyway, the datasource configuration for JMS uses on disk,
which means the HSQLMBean service isn't configured or
deployed.  Take a look at this deployed file: 
$JBOSS/server/default/deploy/hsqldb-ds.xml

The default configuration is broken.

It might be possible to have the HSQLMBean take care of both
issuing SHUTDOWN to a HSQL cached table configuration and
the network configuration.  However, looking at the source,
it didn't seem elegant to write it as such.


--

Comment By: Heiko W.Rupp (pilhuhn)
Date: 2003-11-20 17:07

Message:
Logged In: YES 
user_id=217112

Shouldn't / couldn't that go into the stop() method of the
HSQLMBean? That is called by the server anyway and could do
the  right thing.
I think it is wrong to have two Mbeans for one service.

--

Comment By: Elias Ross (genman)
Date: 2003-11-18 23:18

Message:
Logged In: YES 
user_id=556458


I wrote an MBean to do this and it works okay.  I don't know
if it would be better to update the connection manager (RA
deployer???) or make people use this externally declared
MBean instead.  I'm starting to to think this is the wrong
approach:


   
  ...
  jboss:service=HSQLShutdown
   

   
   
 jdbc:hsqldb:${jboss.server.data.dir}/jms
 org.hsqldb.jdbcDriver
 sa
 
 SHUTDOWN COMPACT
   



--

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
JBoss-Development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development


[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-11-20 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 12:06
Message generated for change (Comment added) made by genman
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossMQ
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Nobody/Anonymous (nobody)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Elias Ross (genman)
Date: 2003-11-20 10:11

Message:
Logged In: YES 
user_id=556458


The problem is the HSQLMBean isn't used for the cached DS
configuration, it is just used to provide a HSQL DB
TCP/IP-enabled datasource.  There are three HSQL operating
modes:  in memory only, on disk, and network-enabled.

Anyway, the datasource configuration for JMS uses on disk,
which means the HSQLMBean service isn't configured or
deployed.  Take a look at this deployed file: 
$JBOSS/server/default/deploy/hsqldb-ds.xml

The default configuration is broken.

It might be possible to have the HSQLMBean take care of both
issuing SHUTDOWN to a HSQL cached table configuration and
the network configuration.  However, looking at the source,
it didn't seem elegant to write it as such.


--

Comment By: Heiko W.Rupp (pilhuhn)
Date: 2003-11-20 09:07

Message:
Logged In: YES 
user_id=217112

Shouldn't / couldn't that go into the stop() method of the
HSQLMBean? That is called by the server anyway and could do
the  right thing.
I think it is wrong to have two Mbeans for one service.

--

Comment By: Elias Ross (genman)
Date: 2003-11-18 15:18

Message:
Logged In: YES 
user_id=556458


I wrote an MBean to do this and it works okay.  I don't know
if it would be better to update the connection manager (RA
deployer???) or make people use this externally declared
MBean instead.  I'm starting to to think this is the wrong
approach:


   
  ...
  jboss:service=HSQLShutdown
   

   
   
 jdbc:hsqldb:${jboss.server.data.dir}/jms
 org.hsqldb.jdbcDriver
 sa
 
 SHUTDOWN COMPACT
   



--

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
JBoss-Development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development


[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-11-20 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 21:06
Message generated for change (Comment added) made by pilhuhn
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossMQ
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Nobody/Anonymous (nobody)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Heiko W.Rupp (pilhuhn)
Date: 2003-11-20 18:07

Message:
Logged In: YES 
user_id=217112

Shouldn't / couldn't that go into the stop() method of the
HSQLMBean? That is called by the server anyway and could do
the  right thing.
I think it is wrong to have two Mbeans for one service.

--

Comment By: Elias Ross (genman)
Date: 2003-11-19 00:18

Message:
Logged In: YES 
user_id=556458


I wrote an MBean to do this and it works okay.  I don't know
if it would be better to update the connection manager (RA
deployer???) or make people use this externally declared
MBean instead.  I'm starting to to think this is the wrong
approach:


   
  ...
  jboss:service=HSQLShutdown
   

   
   
 jdbc:hsqldb:${jboss.server.data.dir}/jms
 org.hsqldb.jdbcDriver
 sa
 
 SHUTDOWN COMPACT
   



--

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
JBoss-Development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development


[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-11-18 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 12:06
Message generated for change (Comment added) made by genman
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossMQ
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Nobody/Anonymous (nobody)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

>Comment By: Elias Ross (genman)
Date: 2003-11-18 15:18

Message:
Logged In: YES 
user_id=556458


I wrote an MBean to do this and it works okay.  I don't know
if it would be better to update the connection manager (RA
deployer???) or make people use this externally declared
MBean instead.  I'm starting to to think this is the wrong
approach:


   
  ...
  jboss:service=HSQLShutdown
   

   
   
 jdbc:hsqldb:${jboss.server.data.dir}/jms
 org.hsqldb.jdbcDriver
 sa
 
 SHUTDOWN COMPACT
   



--

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
JBoss-Development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development


[JBoss-dev] [ jboss-Bugs-844654 ] HSQL datasource -- should issue "SHUTDOWN"

2003-11-18 Thread SourceForge.net
Bugs item #844654, was opened at 2003-11-18 12:06
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866

Category: JBossMQ
Group: v3.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Nobody/Anonymous (nobody)
Summary: HSQL datasource -- should issue "SHUTDOWN"

Initial Comment:

This may or may not be JMS specific, but it relates to
the default Persistence Manager used in JBoss 3.2.2
which uses "cached" tables.

It appears that SHUTDOWN is not issued to the HSQL DB
when JBoss is taken down.  

It should be easy to make the HSQL plug service work
with cached tables and then do the right thing upon
shutdown.  Or, create a new service which issues the
shutdown request.

--

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=844654&group_id=22866


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
JBoss-Development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development