Releases: Particular/NServiceBus.RabbitMQ
2.1.2
2.1.1
2.1.0
As part of this release we had 32 commits which resulted in 13 issues being closed.
Feature
- Make CircuitBreaker intermittent failure delay configurable#31
Improvements
- Only open the admin connection when we need to use it#71
- Pluggable message id strategy#69
- Allow consumer tag to be set#54
- Support RMQ client 3.4#53
- Extend the receive timeouts when debugging#24
- Pass connection purpose to Rabbit API#7
- Better default for the RabbitMQ prefetch count#1
Bugs
- Reply to message incoming via gateway#62
- Creation of exchanges should honor durability settings #49
- Exchange creation failing if local address overridden#29
Where to get it
You can download this release from nuget
2.0.5
2.0.4
As part of this release we had 27 commits which resulted in 2 issues being closed.
Who's affected
Any user publishing/sending messages with publisher confirms enabled(default)
Symptoms
Causes RabbitMQ resources to stay active for longer then necessary.
After a long period all running services fail to connect to the Broker and the endpoint consumes all the CPU of the machine.
Bugs
- #58 Channel not disposed if WaitForConfirmsOrDie throws an exception
- #55 Race condition in RabbitMqDequeueStrategy can cause deadlock
Where to get it
You can download this release from nuget
2.0.3
As part of this release we had 2 commits which resulted in 1 issue being closed.
Bugs
#52 Binary incompatibility with RMQ client 3.4
Hello,
RMQ client 3.4 appears to be binary incompatible with NSB RMQ at least of 1.1.5 version - please see https://groups.google.com/forum/#!topic/rabbitmq-users/xfFEQSYCt2Q
Is it possible to build and publish recompiled Nuget version (probably 1.1.6) with corrected dependencies? Thanks in advance.
best regards,
Alexander Alexeev
Kaspersky Lab
Where to get it
You can download this release from nuget
2.0.2
As part of this release we had 2 commits which resulted in 1 issue being closed.
Bugs
#51 ChannelAllocationException with RabbitMQ while publishing a lot of messages
We are using NServiceBus v5 with RabbitMQ Transport.
We have 6 services that run on the same machine/bus, using NServiceBus.Host.
One of them (InputService) process XML feeds and publish a lot of messages (about 1500) in a foreach loop.
Since the v5 update, this exception is raised in the publish loop of our InputService:
RabbitMQ.Client.Exceptions.ChannelAllocationException: The connection cannot support any more channels. Consider creating a new connection
at RabbitMQ.Client.Impl.SessionManager.Create()
at RabbitMQ.Client.Impl.ConnectionBase.CreateModel()
at NServiceBus.Transports.RabbitMQ.ChannelProvider.GetNewPublishChannel()
in c:\BuildAgent\work\41ea81d808fdfd62\src\NServiceBus.RabbitMQ\ChannelProvider.cs:line 35
Content trimmed. See full issue
Where to get it
You can download this release from nuget
2.0.1
As part of this release we had 5 commits which resulted in 1 issue being closed.
Bugs
#45 When auditing a reply message, the message is sent to the wrong queue
Instead they are sent to the destination specified in the callback header.
This bug was introduced in v2.0.0
Where to get it
2.0.0
As part of this release we had 4 issues closed.
Improvements
#32 Move to .net 4.5
#28 Update to RabbitMq.Client 3.3.5
Bugs
#35 Scale out on RabbitMQ causes duplicate deliveries to handlers
Raised by @fhalim
Migrated from by Particular/NServiceBus#2182
I’ve noticed that the NSB scale out feature defaults to true on RabbitMQ. This causes funky behavior with the fanout exchange being used in the conventional routing topology. For instance, if Message1 is subscribed to on machine1 and machine2, the routing will end up looking like
+---------+
+-->Message1 +--------------+
| +---------+ |
| |
| |
+-----------------+ +-------------------+
Content trimmed. See full issue
#6 RabbitMQ Transport seems to leak threads
For certain integration and acceptance tests we're starting the bus up in a self hosted scenario. It appears that the RabbitMQ transport is leaking threads which causes VSTest to not want to shut down.
This might be related to the same PersistentConnection problem that plagued the RabbitMQ installer for creating queues.
I'll provide a sample in the near future.
Where to get it
1.1.5
As part of this release we had 1 issue closed.
Bugs
#37 Invalid attempt made to decrement the event's count below zero.
Using v1.1.3 and getting this error on RabbitMqDequeuStrategy, line 194.
I don't have a great case to reproduce, only that it's happening on a specific machine repeatedly.
2014-08-11 13:05:54 [Fatal] Repeated failures when communicating with the RabbitMq broker
System.InvalidOperationException: Invalid attempt made to decrement the event's count below zero.
at System.Threading.CountdownEvent.Signal()
at NServiceBus.Transports.RabbitMQ.RabbitMqDequeueStrategy.Action(Object obj) in y:\BuildAgent\work\1ccdfa5b068fb66c\src\NServiceBus.RabbitMQ\RabbitMqDequeueStrategy.cs:line 194
at System.Threading.Tasks.Task.Execute()
Where to get it
You can download this release from nuget