[
https://issues.apache.org/jira/browse/PROTON-1344?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15852097#comment-15852097
]
ASF subversion and git services commented on PROTON-1344:
-
Commit 66d8f0e9c493419f27ff63f47dd7a707efc28e80 in qpid-proton's branch
refs/heads/master from [~astitcher]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton.git;h=66d8f0e ]
PROTON-1344: Added missed PN_LISTENER_ACCEPT event from pn_event_type_name()
> Provide C "proactor" API for multi-threaded proton applications
> ---
>
> Key: PROTON-1344
> URL: https://issues.apache.org/jira/browse/PROTON-1344
> Project: Qpid Proton
> Issue Type: Improvement
> Components: proton-c
>Reporter: Alan Conway
>Assignee: Alan Conway
> Fix For: 0.16.0
>
>
> This work will pull together work that has been going on with the
> experimental pn_connection_engine_t and work on a new C multi-threaded libuv
> C driver. This work has matured to the point of putting together a coherent
> and complete pn_proactor API to (eventually) replace the pn_reactor.
> The big differences:
> - proactive not reactive: e.g. void proactor_connect() does not return a
> bound connection, it starts an asynchronous connection process. All results
> are reported asynchronously as events, not as synchrouns return values.
> - application controls threading: application thread(s) call wait() to block
> for events. There are no callbacks to application code from unknown threading
> contexts.
> - enforces threading model to ensure that events from one connection engine
> are only handled in one thread at a time. The application does not need locks
> to or tables to manage this.
> - provides simple inter-thread singalling - any thread can "wake" a
> connection, causing the proactor to return an event in a thread-safe way for
> application-triggered work.
> The big similarities:
> Same proton protocol engine API is unchanged - the pn_connection_engine will
> dissappear as a separate entity but it's functions will be folded back into
> pn_connection and pn_transport.
> Same event model: existing code that processes events is unchanged, instead
> of registering `void my_handler(pn_event_t *)` with the reactor, the
> application calls my_handler(pn_driver_wait()) in its own threads. There will
> be a handful of new proactor-related events.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-629?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Ross resolved DISPATCH-629.
---
Resolution: Fixed
> Add a protocol-version to the inter-router protocol to enable backward
> compatibility
>
>
> Key: DISPATCH-629
> URL: https://issues.apache.org/jira/browse/DISPATCH-629
> Project: Qpid Dispatch
> Issue Type: Improvement
> Components: Routing Engine
>Reporter: Ted Ross
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> Add a protocol version field to the messages that are passed between routers
> in a network. This will allow future development for backward compatibility
> as new features are added and changes to the protocol occur.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-629?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15852052#comment-15852052
]
ASF subversion and git services commented on DISPATCH-629:
--
Commit 3bb2c53d8dcd708e50340e458b84a6507dbb8b13 in qpid-dispatch's branch
refs/heads/master from [~tr...@redhat.com]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-dispatch.git;h=3bb2c53 ]
DISPATCH-629 - Added protocol-version to inter-router messages for future
backward compatibility.
> Add a protocol-version to the inter-router protocol to enable backward
> compatibility
>
>
> Key: DISPATCH-629
> URL: https://issues.apache.org/jira/browse/DISPATCH-629
> Project: Qpid Dispatch
> Issue Type: Improvement
> Components: Routing Engine
>Reporter: Ted Ross
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> Add a protocol version field to the messages that are passed between routers
> in a network. This will allow future development for backward compatibility
> as new features are added and changes to the protocol occur.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
Ted Ross created DISPATCH-629:
-
Summary: Add a protocol-version to the inter-router protocol to
enable backward compatibility
Key: DISPATCH-629
URL: https://issues.apache.org/jira/browse/DISPATCH-629
Project: Qpid Dispatch
Issue Type: Improvement
Components: Routing Engine
Reporter: Ted Ross
Assignee: Ted Ross
Fix For: 0.8.0
Add a protocol version field to the messages that are passed between routers in
a network. This will allow future development for backward compatibility as
new features are added and changes to the protocol occur.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Ross resolved DISPATCH-628.
---
Resolution: Fixed
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851759#comment-15851759
]
ASF subversion and git services commented on DISPATCH-628:
--
Commit 7d6c053764d454449e997425ac695a2189ef6908 in qpid-dispatch's branch
refs/heads/master from [~lulf]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-dispatch.git;h=7d6c053 ]
DISPATCH-628: Multiple connections per autoLink and linkRoute overwrites
connection handle
This ensures that an autolink and linkroute can be bound to more than one
connection. The assert checking this is disabled in optimized builds and was
never triggered.
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851760#comment-15851760
]
ASF GitHub Bot commented on DISPATCH-628:
-
Github user asfgit closed the pull request at:
https://github.com/apache/qpid-dispatch/pull/142
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
Github user asfgit closed the pull request at:
https://github.com/apache/qpid-dispatch/pull/142
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851750#comment-15851750
]
ASF GitHub Bot commented on DISPATCH-628:
-
Github user lulf commented on a diff in the pull request:
https://github.com/apache/qpid-dispatch/pull/142#discussion_r99380500
--- Diff: src/router_core/route_control.c ---
@@ -396,7 +408,18 @@ void qdr_route_connection_closed_CT(qdr_core_t *core,
qdr_connection_t *conn)
al = DEQ_NEXT_N(REF, al);
}
-cid->open_connection = 0;
+//
+// Remove our own entry in the connection list
+//
+qdr_connection_ref_t * cref = DEQ_HEAD(cid->connection_refs);
+while (cref) {
+if (cref->conn == conn) {
+DEQ_REMOVE(cid->connection_refs, cref);
--- End diff --
Agreed, I think I didn't know about that one. Just pushed an update.
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
Github user lulf commented on a diff in the pull request:
https://github.com/apache/qpid-dispatch/pull/142#discussion_r99380500
--- Diff: src/router_core/route_control.c ---
@@ -396,7 +408,18 @@ void qdr_route_connection_closed_CT(qdr_core_t *core,
qdr_connection_t *conn)
al = DEQ_NEXT_N(REF, al);
}
-cid->open_connection = 0;
+//
+// Remove our own entry in the connection list
+//
+qdr_connection_ref_t * cref = DEQ_HEAD(cid->connection_refs);
+while (cref) {
+if (cref->conn == conn) {
+DEQ_REMOVE(cid->connection_refs, cref);
--- End diff --
Agreed, I think I didn't know about that one. Just pushed an update.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/PROTON-1384?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robbie Gemmell updated PROTON-1384:
---
Fix Version/s: (was: proton-j-0.17.0)
proton-j-0.17.1
> [proton-j] 'Data' utility incorrectly encodes str32 type length
> ---
>
> Key: PROTON-1384
> URL: https://issues.apache.org/jira/browse/PROTON-1384
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: 0.16.0
>Reporter: Robbie Gemmell
>Assignee: Robbie Gemmell
> Fix For: proton-j-0.17.1
>
>
> The 'Data' utility incorrectly encodes str32 type, due to a (probable c)
> error in StringElement, it encodes the length as a [u]byte when it should be
> a [u]int.
> Something like below should fix things (will make it after the reorg is done,
> and I add a test):
> {noformat}
> @@ -127,7 +127,7 @@ class StringElement extends AtomicElement
> else
> {
> b.put((byte)0xb1);
> -b.put((byte)length);
> +b.putInt(length);
> }
> b.put(bytes);
> return size;
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPIDJMS-256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Timothy Bish resolved QPIDJMS-256.
--
Resolution: Fixed
> Closing a consumer that was used in a transaction does not stop message
> dispatching unitl next commit or rollback
> -
>
> Key: QPIDJMS-256
> URL: https://issues.apache.org/jira/browse/QPIDJMS-256
> Project: Qpid JMS
> Issue Type: Bug
> Components: qpid-jms-client
>Affects Versions: 0.11.1, 0.20.0
>Reporter: Timothy Bish
>Assignee: Timothy Bish
> Fix For: 0.21.0
>
>
> When a consumer close is initiated but the consumer is active in an ongoing
> transaction the close is deferred until the next commit or rollback is
> performed however the consumer resource is not stopped or drained of credit
> and so it continues to dispatch messages and gather prefetch until the next
> commit or rollback is performed.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPIDJMS-231?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Timothy Bish resolved QPIDJMS-231.
--
Resolution: Fixed
Fix Version/s: 0.21.0
> closing a consumer used since prior commit/rollback holds its prefetched
> messages until the next commit/rollback
>
>
> Key: QPIDJMS-231
> URL: https://issues.apache.org/jira/browse/QPIDJMS-231
> Project: Qpid JMS
> Issue Type: Bug
> Components: qpid-jms-client
>Affects Versions: 0.11.0, 0.20.0
>Reporter: Alex Rudyy
>Assignee: Timothy Bish
> Fix For: 0.21.0
>
>
> The messages prefetched by the consumer and not yet delivered to the
> application are not released on consumer close. As result, the new consumer
> cannot consume the messages prefetched by closed consumer.
> This issue was discovered for a transacted consumer. I did not investigate
> whether other acknowledgment modes are affected by it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPIDJMS-231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851667#comment-15851667
]
ASF subversion and git services commented on QPIDJMS-231:
-
Commit d22a22f64df447152b78a575e5b2ce44c0b89087 in qpid-jms's branch
refs/heads/master from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-jms.git;h=d22a22f ]
QPIDJMS-256 QPIDJMS-231 Better deal with deffered consumer close
Handle consumer close in TX better by moving the deferment handling to
the AmqpConsumer where close decide if it should hold until the TX has
completed and ensure that on deferred close that the consumer is drained
and prefetched messages are released back to the remote for redispatch.
Ensure that when a consumer is closed but deferred due to being in a
transaction or having pending delivered messages that are not yet ack'd
the client does not dispatch any more inbound messages to the consumer.
> closing a consumer used since prior commit/rollback holds its prefetched
> messages until the next commit/rollback
>
>
> Key: QPIDJMS-231
> URL: https://issues.apache.org/jira/browse/QPIDJMS-231
> Project: Qpid JMS
> Issue Type: Bug
> Components: qpid-jms-client
>Affects Versions: 0.11.0, 0.20.0
>Reporter: Alex Rudyy
>Assignee: Timothy Bish
>
> The messages prefetched by the consumer and not yet delivered to the
> application are not released on consumer close. As result, the new consumer
> cannot consume the messages prefetched by closed consumer.
> This issue was discovered for a transacted consumer. I did not investigate
> whether other acknowledgment modes are affected by it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPIDJMS-256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851666#comment-15851666
]
ASF subversion and git services commented on QPIDJMS-256:
-
Commit d22a22f64df447152b78a575e5b2ce44c0b89087 in qpid-jms's branch
refs/heads/master from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-jms.git;h=d22a22f ]
QPIDJMS-256 QPIDJMS-231 Better deal with deffered consumer close
Handle consumer close in TX better by moving the deferment handling to
the AmqpConsumer where close decide if it should hold until the TX has
completed and ensure that on deferred close that the consumer is drained
and prefetched messages are released back to the remote for redispatch.
Ensure that when a consumer is closed but deferred due to being in a
transaction or having pending delivered messages that are not yet ack'd
the client does not dispatch any more inbound messages to the consumer.
> Closing a consumer that was used in a transaction does not stop message
> dispatching unitl next commit or rollback
> -
>
> Key: QPIDJMS-256
> URL: https://issues.apache.org/jira/browse/QPIDJMS-256
> Project: Qpid JMS
> Issue Type: Bug
> Components: qpid-jms-client
>Affects Versions: 0.11.1, 0.20.0
>Reporter: Timothy Bish
>Assignee: Timothy Bish
> Fix For: 0.21.0
>
>
> When a consumer close is initiated but the consumer is active in an ongoing
> transaction the close is deferred until the next commit or rollback is
> performed however the consumer resource is not stopped or drained of credit
> and so it continues to dispatch messages and gather prefetch until the next
> commit or rollback is performed.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851660#comment-15851660
]
ASF GitHub Bot commented on DISPATCH-628:
-
Github user ted-ross commented on a diff in the pull request:
https://github.com/apache/qpid-dispatch/pull/142#discussion_r99370107
--- Diff: src/router_core/route_control.c ---
@@ -396,7 +408,18 @@ void qdr_route_connection_closed_CT(qdr_core_t *core,
qdr_connection_t *conn)
al = DEQ_NEXT_N(REF, al);
}
-cid->open_connection = 0;
+//
+// Remove our own entry in the connection list
+//
+qdr_connection_ref_t * cref = DEQ_HEAD(cid->connection_refs);
+while (cref) {
+if (cref->conn == conn) {
+DEQ_REMOVE(cid->connection_refs, cref);
--- End diff --
I believe this will leak one qdr_connection_ref_t each time it is run. Why
did you not use qdr_del_connection_ref?
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
Github user ted-ross commented on a diff in the pull request:
https://github.com/apache/qpid-dispatch/pull/142#discussion_r99370107
--- Diff: src/router_core/route_control.c ---
@@ -396,7 +408,18 @@ void qdr_route_connection_closed_CT(qdr_core_t *core,
qdr_connection_t *conn)
al = DEQ_NEXT_N(REF, al);
}
-cid->open_connection = 0;
+//
+// Remove our own entry in the connection list
+//
+qdr_connection_ref_t * cref = DEQ_HEAD(cid->connection_refs);
+while (cref) {
+if (cref->conn == conn) {
+DEQ_REMOVE(cid->connection_refs, cref);
--- End diff --
I believe this will leak one qdr_connection_ref_t each time it is run. Why
did you not use qdr_del_connection_ref?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Ross updated DISPATCH-628:
--
Component/s: Router Node
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Ross reassigned DISPATCH-628:
-
Assignee: Ted Ross
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
> Components: Router Node
>Reporter: Ulf Lilleengen
>Assignee: Ted Ross
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Ross updated DISPATCH-628:
--
Fix Version/s: 0.8.0
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
>Reporter: Ulf Lilleengen
> Fix For: 0.8.0
>
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPID-7558?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Keith Wall updated QPID-7558:
-
Attachment: QPID-7558_v6.1.patch
> [Java Broker] Allow Multiple JDBC Virtual Hosts / Message Stores to share the
> same database
> ---
>
> Key: QPID-7558
> URL: https://issues.apache.org/jira/browse/QPID-7558
> Project: Qpid
> Issue Type: Improvement
> Components: Java Broker
>Reporter: Rob Godfrey
> Fix For: qpid-java-7.0
>
> Attachments:
> 0001-QPID-7558-Java-Broker-whitelist-validValuePattern-fo.patch,
> QPID-7558_v6.1.patch
>
>
> As per the e-mail discussion on the users list
> [here|http://mail-archives.apache.org/mod_mbox/qpid-users/201611.mbox/%3CCACMpbT81LATE1d2M4L%2B%3D1uq_CDPnx8y56Md0w362xS54-yGVeg%40mail.gmail.com%3E]
> it may sometimes be useful for the user to be able to configure a JDBC
> store to use a per-instance table name prefix to allow multiple instances to
> be run against the same database/schema.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/PROTON-1385?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851408#comment-15851408
]
ASF subversion and git services commented on PROTON-1385:
-
Commit 2ce5570973864946f454802bae7247de181e50d1 in qpid-proton-j's branch
refs/heads/master from Robert Gemmell
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=2ce5570 ]
PROTON-1385: add build status badges to readme
> make proton-j independently releasable
> --
>
> Key: PROTON-1385
> URL: https://issues.apache.org/jira/browse/PROTON-1385
> Project: Qpid Proton
> Issue Type: Task
> Components: build, proton-j, release
>Reporter: Robbie Gemmell
>Assignee: Robbie Gemmell
> Fix For: 0.17.0, proton-j-0.17.0
>
>
> As previously discussed on the mailing lists, we are moving proton-j to be an
> independently released component. As part of this a new qpid-proton-j
> repository was created in December as a copy of the existing proton repo, it
> can be found at:
> https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git
> This JIRA will track the changes made to leave only proton-j in its new repo,
> and also remove it from the original qpid-proton repository where proton-c
> and its language bindings will remain.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPID-7658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851372#comment-15851372
]
Rob Godfrey commented on QPID-7658:
---
It just seems like an accident waiting to happen - if the "localContainerId" is
ever anything but the container id of the virtual host, what does that mean? A
couple of random thoughts:
* should the link registry also record some sort of ttl for the links... i.e.
if a client never reconnects is it not the link registry's job to remove the
link from itself... or are you expecting that the link registry will do a deep
inspection of the local teminus to determine when it should auto-remove the link
* you don't provide an explicit mechanism for updating the data in the
registry ... are the terminii objects immutable? If not how does the registry
know when the state of the terminii has changed and from that whether the
change requires persisting
Personally I still feel your model is wrong in that I think there is a Link
object that encapsulates the local terminus, the remote terminus, and the link
endpoint (if any) currently associated with the local terminus.
> [Java Broker] Improve LinkRegistry
> --
>
> Key: QPID-7658
> URL: https://issues.apache.org/jira/browse/QPID-7658
> Project: Qpid
> Issue Type: Task
> Components: Java Broker
>Reporter: Lorenz Quack
> Fix For: qpid-java-7.0
>
>
> Currently the AbstractVirtualHost has a Map.
> The handling of the remoteContainerId should also be encapsulated in the
> LinkRegistry giving each VH only a single LinkRegistry.
> Furthermore the following change to the LinkRegistry is proposed:
> * LinkRegistry.putSendingLinkIfAbsent(String localContainerId, String
> remoteContrainerId, String linkName, Source source, Target target) ->
> Pair
> * LinkRegistry.putReceivingLinkIfAbsent(String localContainerId, String
> remoteContrainerId, String linkName, Source source, Target target) ->
> Pair
> * LinkRegistry.getSendvingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.getReceivingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.removeSendingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.removeReceivingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> The implementation of which must be thread-safe because the LinkRegistry will
> be accessed on multiple threads concurrently.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPID-7658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851350#comment-15851350
]
Lorenz Quack commented on QPID-7658:
You are right, in the HA case we could not share a single LinkRegistry.
I felt that because the localContainerId is part of the definition of the Link
it felt reasonable to pass it in.
Whether the implementation then actually uses it should not be the concern of
the interface.
But maybe you are right and this will never be useful.
> [Java Broker] Improve LinkRegistry
> --
>
> Key: QPID-7658
> URL: https://issues.apache.org/jira/browse/QPID-7658
> Project: Qpid
> Issue Type: Task
> Components: Java Broker
>Reporter: Lorenz Quack
> Fix For: qpid-java-7.0
>
>
> Currently the AbstractVirtualHost has a Map.
> The handling of the remoteContainerId should also be encapsulated in the
> LinkRegistry giving each VH only a single LinkRegistry.
> Furthermore the following change to the LinkRegistry is proposed:
> * LinkRegistry.putSendingLinkIfAbsent(String localContainerId, String
> remoteContrainerId, String linkName, Source source, Target target) ->
> Pair
> * LinkRegistry.putReceivingLinkIfAbsent(String localContainerId, String
> remoteContrainerId, String linkName, Source source, Target target) ->
> Pair
> * LinkRegistry.getSendvingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.getReceivingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.removeSendingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.removeReceivingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> The implementation of which must be thread-safe because the LinkRegistry will
> be accessed on multiple threads concurrently.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/PROTON-1385?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851343#comment-15851343
]
ASF subversion and git services commented on PROTON-1385:
-
Commit e12b8373a56acd86eceaf9047a18131da012e2d4 in qpid-proton-j's branch
refs/heads/master from Robert Gemmell
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=e12b837 ]
PROTON-1385: add tag name format config for the release plugin, tweak pom to
avoid irrelevant change during version rewrites
> make proton-j independently releasable
> --
>
> Key: PROTON-1385
> URL: https://issues.apache.org/jira/browse/PROTON-1385
> Project: Qpid Proton
> Issue Type: Task
> Components: build, proton-j, release
>Reporter: Robbie Gemmell
>Assignee: Robbie Gemmell
> Fix For: 0.17.0, proton-j-0.17.0
>
>
> As previously discussed on the mailing lists, we are moving proton-j to be an
> independently released component. As part of this a new qpid-proton-j
> repository was created in December as a copy of the existing proton repo, it
> can be found at:
> https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git
> This JIRA will track the changes made to leave only proton-j in its new repo,
> and also remove it from the original qpid-proton repository where proton-c
> and its language bindings will remain.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/PROTON-1396?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851338#comment-15851338
]
ASF subversion and git services commented on PROTON-1396:
-
Commit 99e516b5c3cc62090f5d69ea4a9af433014541de in qpid-proton's branch
refs/heads/master from Robert Gemmell
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton.git;h=99e516b ]
PROTON-1396: restore the so version to 10 on master for 0.17.0.
Checking (thanks Andrew!) shows no ABI relevant change between 0.16.0 and now.
This reverts commit 2389a2179e39fd53912a3e28369b1b409d94c996 for PROTON-1374
> 0.17.0 release tasks
>
>
> Key: PROTON-1396
> URL: https://issues.apache.org/jira/browse/PROTON-1396
> Project: Qpid Proton
> Issue Type: Task
> Components: release
>Reporter: Robbie Gemmell
>Assignee: Robbie Gemmell
> Fix For: 0.17.0
>
>
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/PROTON-1374?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851339#comment-15851339
]
ASF subversion and git services commented on PROTON-1374:
-
Commit 99e516b5c3cc62090f5d69ea4a9af433014541de in qpid-proton's branch
refs/heads/master from Robert Gemmell
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton.git;h=99e516b ]
PROTON-1396: restore the so version to 10 on master for 0.17.0.
Checking (thanks Andrew!) shows no ABI relevant change between 0.16.0 and now.
This reverts commit 2389a2179e39fd53912a3e28369b1b409d94c996 for PROTON-1374
> 0.16.0 release tasks
>
>
> Key: PROTON-1374
> URL: https://issues.apache.org/jira/browse/PROTON-1374
> Project: Qpid Proton
> Issue Type: Task
> Components: release
>Reporter: Robbie Gemmell
>Assignee: Robbie Gemmell
> Fix For: 0.16.0
>
>
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/QPID-7658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851330#comment-15851330
]
Rob Godfrey commented on QPID-7658:
---
Surely the link registry state will need to be persisted, and thus replicated
when using HA virtual hosts, which precludes the sharing of a link registry
between different virtual hosts on the same broker (unless you propose there is
a single shared link registry across all Java Brokers in existence). As such I
fail to se the utility in supplying the localContainerId in to these methods.
I think by definition a link registry is specific to the (local) container.
> [Java Broker] Improve LinkRegistry
> --
>
> Key: QPID-7658
> URL: https://issues.apache.org/jira/browse/QPID-7658
> Project: Qpid
> Issue Type: Task
> Components: Java Broker
>Reporter: Lorenz Quack
> Fix For: qpid-java-7.0
>
>
> Currently the AbstractVirtualHost has a Map.
> The handling of the remoteContainerId should also be encapsulated in the
> LinkRegistry giving each VH only a single LinkRegistry.
> Furthermore the following change to the LinkRegistry is proposed:
> * LinkRegistry.putSendingLinkIfAbsent(String localContainerId, String
> remoteContrainerId, String linkName, Source source, Target target) ->
> Pair
> * LinkRegistry.putReceivingLinkIfAbsent(String localContainerId, String
> remoteContrainerId, String linkName, Source source, Target target) ->
> Pair
> * LinkRegistry.getSendvingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.getReceivingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.removeSendingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> * LinkRegistry.removeReceivingLink(String localContainerId, String
> remoteContrainerId, String linkName) -> Pair
> The implementation of which must be thread-safe because the LinkRegistry will
> be accessed on multiple threads concurrently.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[
https://issues.apache.org/jira/browse/DISPATCH-628?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15851312#comment-15851312
]
ASF GitHub Bot commented on DISPATCH-628:
-
GitHub user lulf opened a pull request:
https://github.com/apache/qpid-dispatch/pull/142
DISPATCH-628: Multiple connections per autoLink and linkRoute overwrites
connection handle
This ensures that an autolink and linkroute can be bound to more than one
connection. The assert checking this is disabled in optimized builds and was
never triggered.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/lulf/qpid-dispatch DISPATCH-628
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/qpid-dispatch/pull/142.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #142
commit 49d95b767ca5ee8fc9d0dadb3951c799e4ec89ba
Author: Ulf Lilleengen
Date: 2017-02-03T09:58:19Z
DISPATCH-628: Multiple connections per autoLink and linkRoute overwrites
connection handle
This ensures that an autolink and linkroute can be bound to more than one
connection. The assert checking this is disabled in optimized builds and was
never triggered.
> [PATCH] Multiple connections per autoLink and linkRoute overwrites connection
> handle
>
>
> Key: DISPATCH-628
> URL: https://issues.apache.org/jira/browse/DISPATCH-628
> Project: Qpid Dispatch
> Issue Type: Bug
>Reporter: Ulf Lilleengen
>
> When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
> across an issue with an autoLink or linkRoute only being able to refer to
> more than 1 connection, which I assumed would originaly cause an assert in
> route_control.c:350 to trigger when I configured an autoLink to trigger on a
> containerId and use the same container id when sharding a queue.
> Since the assert didn't trigger, I assumed that I must have misread the code,
> but nontheless as part of the patch I made it possible for a
> linkRoute/autoLink to be associated with multiple connections.
> Recently, I have seen some router crashes when scaling up and down queues in
> EnMasse, and I just figured that the reason for this assert not triggering is
> that it is disabled in the RelWithDebInfo configuration.
> If built without optimizations (Debug), I'm able to trigger the issue with
> the configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
> To trigger, I startup the router and broker as so:
> qdrouterd -c router_incoming.conf
> export MESSAGING_SERVICE_HOST=localhost
> export MESSAGING_SERVICE_PORT_INTERNAL=55673
> export QUEUE_NAME=myqueue
> export CONTAINER_ID=myqueue
> ./br/bin/simplebroker &
> ./br/bin/simplebroker
> I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
GitHub user lulf opened a pull request:
https://github.com/apache/qpid-dispatch/pull/142
DISPATCH-628: Multiple connections per autoLink and linkRoute overwrites
connection handle
This ensures that an autolink and linkroute can be bound to more than one
connection. The assert checking this is disabled in optimized builds and was
never triggered.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/lulf/qpid-dispatch DISPATCH-628
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/qpid-dispatch/pull/142.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #142
commit 49d95b767ca5ee8fc9d0dadb3951c799e4ec89ba
Author: Ulf Lilleengen
Date: 2017-02-03T09:58:19Z
DISPATCH-628: Multiple connections per autoLink and linkRoute overwrites
connection handle
This ensures that an autolink and linkroute can be bound to more than one
connection. The assert checking this is disabled in optimized builds and was
never triggered.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
Ulf Lilleengen created DISPATCH-628:
---
Summary: [PATCH] Multiple connections per autoLink and linkRoute
overwrites connection handle
Key: DISPATCH-628
URL: https://issues.apache.org/jira/browse/DISPATCH-628
Project: Qpid Dispatch
Issue Type: Bug
Reporter: Ulf Lilleengen
When implementing https://github.com/apache/qpid-dispatch/pull/111 I stumbled
across an issue with an autoLink or linkRoute only being able to refer to more
than 1 connection, which I assumed would originaly cause an assert in
route_control.c:350 to trigger when I configured an autoLink to trigger on a
containerId and use the same container id when sharding a queue.
Since the assert didn't trigger, I assumed that I must have misread the code,
but nontheless as part of the patch I made it possible for a linkRoute/autoLink
to be associated with multiple connections.
Recently, I have seen some router crashes when scaling up and down queues in
EnMasse, and I just figured that the reason for this assert not triggering is
that it is disabled in the RelWithDebInfo configuration.
If built without optimizations (Debug), I'm able to trigger the issue with the
configuration in http://people.redhat.com/~ulilleen/bugs/routerconn/
To trigger, I startup the router and broker as so:
qdrouterd -c router_incoming.conf
export MESSAGING_SERVICE_HOST=localhost
export MESSAGING_SERVICE_PORT_INTERNAL=55673
export QUEUE_NAME=myqueue
export CONTAINER_ID=myqueue
./br/bin/simplebroker &
./br/bin/simplebroker
I have extracted the patch from DISPATCH-476 and will create a PR with it.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org