You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by khaing211 <kh...@gmail.com> on 2016/01/09 00:04:33 UTC
received backup update for live/backup pair X/Y but it didn't
belong to
Hi,
We have a cluster of 4 lives and 4 standbies (1 to 1). For our client, we
instantiate TransportConfiguration to each of lives hosts by our
configuration. When we create small number (i.e. 20-30) of connections from
a single client to the cluster, we don't see any problem.
When we create 4000 connections (1000 connection per node by roundrobin
policy), we see a lot variant of the following DEBUG log.
17:55:02.005 DEBUG [Thread-10 (ActiveMQ-client-netty-threads-10304615)]
org.apache.activemq.artemis.core.client - ClientSessionFactoryImpl received
backup update for live/backup pair = TransportConfiguration(name=artemis,
factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
?port=61616&host=prod-artemis1 / TransportConfiguration(name=artemis,
factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
?port=61616&host=prod-artemis-backup1 but it didn't belong to
TransportConfiguration(name=artemis,
factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
?port=61616&host=prod-artemis4
Can anyone explain why we get live/backup pair for incorrect connector?
https://github.com/apache/activemq-artemis/blob/master/artemis-core-client/src/main/java/org/apache/activemq/artemis/core/client/impl/ClientSessionFactoryImpl.java
--
View this message in context: http://activemq.2283324.n4.nabble.com/received-backup-update-for-live-backup-pair-X-Y-but-it-didn-t-belong-to-tp4705709.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: received backup update for live/backup pair X/Y but it didn't belong to
Posted by Clebert Suconic <cl...@gmail.com>.
Isn't that just debug logs? I'm trying to find what's he issue TBH.
Why don't u create an independent example showing your issue ?
Sent from my iPhone
> On Jan 8, 2016, at 8:17 PM, Khai Nguyen <kh...@gmail.com> wrote:
>
> Apache Artemis 1.1.0
>
> I can't show you the exact configuration i.e. ip info and bunch of layer of
> abstraction. Also, the log above is of the client api, not the server.
> Server is fine. Client keep logging that.
>
> I can give you a general sense of the setup.
>
> Server:
>
> 1. Cluster by broadcast group
> 1. master / slave in same broadcast group.
> 2. ha-policy by replication
> 1. master / slave
> 2. matching group-name
>
>
> Client configuration by java code
>
> 1. Netty connector
> 2. Array of TransportConfiguration with host and port
> 3. JMS ConnectionFactory interface
>
>
>
>
> On Fri, Jan 8, 2016 at 5:02 PM Clebert Suconic <cl...@gmail.com>
> wrote:
>
>> I have no idea TBH. I would need more information.. like an example
>> showing the issue (replicating the issue). It could be a configuration
>> issue. Or even if you hit a but anyone would need some help to
>> identify what's going on.
>>
>>
>> Start with what version you're using please? 1.0.0, 1.1.0, 1.2.0?
>>
>>> On Fri, Jan 8, 2016 at 6:04 PM, khaing211 <kh...@gmail.com> wrote:
>>> Hi,
>>>
>>> We have a cluster of 4 lives and 4 standbies (1 to 1). For our client, we
>>> instantiate TransportConfiguration to each of lives hosts by our
>>> configuration. When we create small number (i.e. 20-30) of connections
>> from
>>> a single client to the cluster, we don't see any problem.
>>>
>>> When we create 4000 connections (1000 connection per node by roundrobin
>>> policy), we see a lot variant of the following DEBUG log.
>>>
>>> 17:55:02.005 DEBUG [Thread-10 (ActiveMQ-client-netty-threads-10304615)]
>>> org.apache.activemq.artemis.core.client - ClientSessionFactoryImpl
>> received
>>> backup update for live/backup pair = TransportConfiguration(name=artemis,
>> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
>>> ?port=61616&host=prod-artemis1 / TransportConfiguration(name=artemis,
>> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
>>> ?port=61616&host=prod-artemis-backup1 but it didn't belong to
>>> TransportConfiguration(name=artemis,
>> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
>>> ?port=61616&host=prod-artemis4
>>>
>>>
>>> Can anyone explain why we get live/backup pair for incorrect connector?
>> https://github.com/apache/activemq-artemis/blob/master/artemis-core-client/src/main/java/org/apache/activemq/artemis/core/client/impl/ClientSessionFactoryImpl.java
>>>
>>>
>>>
>>> --
>>> View this message in context:
>> http://activemq.2283324.n4.nabble.com/received-backup-update-for-live-backup-pair-X-Y-but-it-didn-t-belong-to-tp4705709.html
>>> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>>
>>
>>
>> --
>> Clebert Suconic
>>
Re: received backup update for live/backup pair X/Y but it didn't
belong to
Posted by Khai Nguyen <kh...@gmail.com>.
Apache Artemis 1.1.0
I can't show you the exact configuration i.e. ip info and bunch of layer of
abstraction. Also, the log above is of the client api, not the server.
Server is fine. Client keep logging that.
I can give you a general sense of the setup.
Server:
1. Cluster by broadcast group
1. master / slave in same broadcast group.
2. ha-policy by replication
1. master / slave
2. matching group-name
Client configuration by java code
1. Netty connector
2. Array of TransportConfiguration with host and port
3. JMS ConnectionFactory interface
On Fri, Jan 8, 2016 at 5:02 PM Clebert Suconic <cl...@gmail.com>
wrote:
> I have no idea TBH. I would need more information.. like an example
> showing the issue (replicating the issue). It could be a configuration
> issue. Or even if you hit a but anyone would need some help to
> identify what's going on.
>
>
> Start with what version you're using please? 1.0.0, 1.1.0, 1.2.0?
>
> On Fri, Jan 8, 2016 at 6:04 PM, khaing211 <kh...@gmail.com> wrote:
> > Hi,
> >
> > We have a cluster of 4 lives and 4 standbies (1 to 1). For our client, we
> > instantiate TransportConfiguration to each of lives hosts by our
> > configuration. When we create small number (i.e. 20-30) of connections
> from
> > a single client to the cluster, we don't see any problem.
> >
> > When we create 4000 connections (1000 connection per node by roundrobin
> > policy), we see a lot variant of the following DEBUG log.
> >
> > 17:55:02.005 DEBUG [Thread-10 (ActiveMQ-client-netty-threads-10304615)]
> > org.apache.activemq.artemis.core.client - ClientSessionFactoryImpl
> received
> > backup update for live/backup pair = TransportConfiguration(name=artemis,
> >
> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
> > ?port=61616&host=prod-artemis1 / TransportConfiguration(name=artemis,
> >
> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
> > ?port=61616&host=prod-artemis-backup1 but it didn't belong to
> > TransportConfiguration(name=artemis,
> >
> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
> > ?port=61616&host=prod-artemis4
> >
> >
> > Can anyone explain why we get live/backup pair for incorrect connector?
> >
> >
> >
> https://github.com/apache/activemq-artemis/blob/master/artemis-core-client/src/main/java/org/apache/activemq/artemis/core/client/impl/ClientSessionFactoryImpl.java
> >
> >
> >
> > --
> > View this message in context:
> http://activemq.2283324.n4.nabble.com/received-backup-update-for-live-backup-pair-X-Y-but-it-didn-t-belong-to-tp4705709.html
> > Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>
>
>
> --
> Clebert Suconic
>
Re: received backup update for live/backup pair X/Y but it didn't
belong to
Posted by Clebert Suconic <cl...@gmail.com>.
I have no idea TBH. I would need more information.. like an example
showing the issue (replicating the issue). It could be a configuration
issue. Or even if you hit a but anyone would need some help to
identify what's going on.
Start with what version you're using please? 1.0.0, 1.1.0, 1.2.0?
On Fri, Jan 8, 2016 at 6:04 PM, khaing211 <kh...@gmail.com> wrote:
> Hi,
>
> We have a cluster of 4 lives and 4 standbies (1 to 1). For our client, we
> instantiate TransportConfiguration to each of lives hosts by our
> configuration. When we create small number (i.e. 20-30) of connections from
> a single client to the cluster, we don't see any problem.
>
> When we create 4000 connections (1000 connection per node by roundrobin
> policy), we see a lot variant of the following DEBUG log.
>
> 17:55:02.005 DEBUG [Thread-10 (ActiveMQ-client-netty-threads-10304615)]
> org.apache.activemq.artemis.core.client - ClientSessionFactoryImpl received
> backup update for live/backup pair = TransportConfiguration(name=artemis,
> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
> ?port=61616&host=prod-artemis1 / TransportConfiguration(name=artemis,
> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
> ?port=61616&host=prod-artemis-backup1 but it didn't belong to
> TransportConfiguration(name=artemis,
> factory=org-apache-activemq-artemis-core-remoting-impl-netty-NettyConnectorFactory)
> ?port=61616&host=prod-artemis4
>
>
> Can anyone explain why we get live/backup pair for incorrect connector?
>
>
> https://github.com/apache/activemq-artemis/blob/master/artemis-core-client/src/main/java/org/apache/activemq/artemis/core/client/impl/ClientSessionFactoryImpl.java
>
>
>
> --
> View this message in context: http://activemq.2283324.n4.nabble.com/received-backup-update-for-live-backup-pair-X-Y-but-it-didn-t-belong-to-tp4705709.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
--
Clebert Suconic