5:17 a.m.
Clarify request context lifecycle during remote method invocation of EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:59 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Carlo de Wolf commented on CDI-180:
-----------------------------------
The same arguments I posted in
http://java.net/projects/ejb-spec/lists/jsr345-experts/archive/2011-09/me... apply
here.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
5:56 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Pete Muir commented on CDI-180:
-------------------------------
Decomposing this:
However it doesn't cover various invocation scenarios. For
example it's
possible to do in-VM invocation (via remote interface) from the same
deployment or from a different deployment.
I don't see why the spec implies this is treated any differently.
Is the request context shared in this case?
The spec doesn't specify how the request context is shared. We should tidy this up.
Will be destroyed after the EJB remote method invocation?
I think this is fairly unambiguous given "The request context is destroyed: after the
EJB remote method invocation"!
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:02 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Stuart Douglas commented on CDI-180:
------------------------------------
> However it doesn't cover various invocation scenarios. For
example it's
> possible to do in-VM invocation (via remote interface) from the same
> deployment or from a different deployment.
I don't see why the spec implies this is treated any differently.
I think the real question here is that if you have 2 CDI applications, and they
communicate via in-vm method calls, does the bean in the second application live for a
single invocation, or does it last for the life of the calling request scope. I would
assume the former.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:08 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Pete Muir commented on CDI-180:
-------------------------------
I think we need the exact definition of a remote method invocation lifespan from EJB to
clear this up properly. Does anyone have this?
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:41 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Mark Struberg commented on CDI-180:
-----------------------------------
I'm not sure if this falls under this category, but do we generally like to define
what happens with remote and also with _async_ parts?
E.g. if I have a bean which starts an async EJB task. At first it sounds very tempting to
just promote the current RequestContext and SessionContext also over to this async task.
But now think a bit about it. The user starts an async EJB method which needs 30 minutes
(or even longer) to succeed. Immediately after that, the user logs out and the
logOutHandler therefor destroys the Session. What will happen?
Also please be aware that the methods to programmatically access a Session with a given Id
got removed from the Servlet spec long time ago. Plus there are serialization issues.
Back to your original problem (which is different, but the effect is pretty much the
same): whenever we cross boundaries (threads, servers), this might work in some
situations, but we will fail to support it in a generally valid manner.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:47 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Mark Struberg edited comment on CDI-180 at 10/26/11 7:46 AM:
-------------------------------------------------------------
I'm not sure if this falls under this category, but do we generally like to define
what happens with remote and also with _async_ parts?
E.g. if I have a bean which starts an async EJB task. At first it sounds very tempting to
just promote the current RequestContext and SessionContext also over to this async task.
But now think a bit about it. The user starts an async EJB method which needs 30 minutes
(or even longer) to succeed. Immediately after kicking off the task (long before it
finishes), the user logs out and the logOutHandler therefor destroys the Session. What
will happen?
Also please be aware that the methods to programmatically access a Session with a given Id
got removed from the Servlet spec long time ago. Plus there are serialization issues.
Back to your original problem (which is different, but the effect is pretty much the
same): whenever we cross boundaries (threads, servers), this might work in some
situations, but we will fail to support it in a generally valid manner.
was (Author: struberg):
I'm not sure if this falls under this category, but do we generally like to define
what happens with remote and also with _async_ parts?
E.g. if I have a bean which starts an async EJB task. At first it sounds very tempting to
just promote the current RequestContext and SessionContext also over to this async task.
But now think a bit about it. The user starts an async EJB method which needs 30 minutes
(or even longer) to succeed. Immediately after that, the user logs out and the
logOutHandler therefor destroys the Session. What will happen?
Also please be aware that the methods to programmatically access a Session with a given Id
got removed from the Servlet spec long time ago. Plus there are serialization issues.
Back to your original problem (which is different, but the effect is pretty much the
same): whenever we cross boundaries (threads, servers), this might work in some
situations, but we will fail to support it in a generally valid manner.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:47 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Martin Kouba commented on CDI-180:
----------------------------------
I assume async parts (EJB async method, timer) should run in isolated request context and
with no active session context.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:59 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Mark Struberg commented on CDI-180:
-----------------------------------
@Martin might work, but that's nowhere defined that way atm, right?
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
7:19 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Martin Kouba commented on CDI-180:
----------------------------------
Exactly it's not defined in spec. As Pete commented "...spec doesn't specify
how the request context is shared. We should tidy this up...". Thus CDI spec should
clarify *isolation of built-in contexts during various invocation scenarios*.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
1:11 p.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Pete Muir updated CDI-180:
--------------------------
Fix Version/s: 1.1 (Proposed)
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
Fix For: 1.1 (Proposed)
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:46 p.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Pete Muir reassigned CDI-180:
-----------------------------
Assignee: Joe Bergmark
Another closely related issue
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
Assignee: Joe Bergmark
Fix For: 1.1 (Proposed)
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:54 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Pete Muir updated CDI-180:
--------------------------
Fix Version/s: 1.1.PRD
(was: 1.1 (Proposed))
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
Assignee: Joe Bergmark
Fix For: 1.1.PRD
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
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
7:05 a.m.
[
https://issues.jboss.org/browse/CDI-180?page=com.atlassian.jira.plugin.sy...
]
Pete Muir resolved CDI-180.
---------------------------
Resolution: Done
Merged Joe's patch.
Clarify request context lifecycle during remote method invocation of
EJB
------------------------------------------------------------------------
Key: CDI-180
URL: https://issues.jboss.org/browse/CDI-180
Project: CDI Specification Issues
Issue Type: Clarification
Affects Versions: 1.0
Reporter: Martin Kouba
Assignee: Joe Bergmark
Fix For: 1.1.PRD
CDI 1.0 spec states in *6.7.1. Request context lifecycle*:
"The request scope is active: during any remote method invocation of any EJB"
and "The request context is destroyed: after the EJB remote method invocation"
However it doesn't cover various invocation scenarios. For example it's possible
to do in-VM invocation (via remote interface) from the same deployment or from a different
deployment. Is the request context shared in this case? Will be destroyed after the EJB
remote method invocation?
Also take into account that application servers often do optimization and handle remote
interface calls in a local interface manner within the same JVM.
--
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
4253
days inactive
4567
days old
13 comments
8 participants
participants (8)
-
Carlo de Wolf (Commented) (JIRA) -
Mark Struberg (Commented) (JIRA)
-
Mark Struberg (Issue Comment Edited) (JIRA)
-
Martin Kouba (Commented) (JIRA)
-
Martin Kouba (Created) (JIRA)
-
Pete Muir (Commented) (JIRA)
-
Pete Muir (JIRA)
-
Stuart Douglas (Commented) (JIRA)