You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by Travis Klotz <tr...@gmail.com> on 2013/01/09 16:30:35 UTC
Issues with stand alone web console in 5.7 using a master/slave architecture
We are running the web console as stand alone application in a
glassfish container using system properties to configure connection
info to point to our two server master/slave setup. We do this so
that we easily have a single URL to look at regardless of which server
is the master and which is the slave.
This has worked great for us in 5.5 and 5.6 but when upgrading to 5.7
we have run into a problem. When server1 is the master, everything
works great, but when server1 is the slave, we always get redirected
to slave.jsp and are informed that the server is currently in slave
mode. This in in spite of the fact that we have configured the web
app to point to both servers using both JMX and JMS.
For Example:
-Dwebconsole.jms.url=failover:(tcp://server1:61616,tcp://server2:61616)
-Dwebconsole.jmx.url=service:jmx:rmi:///jndi/rmi://server1:1099/jmxrmi,service:jmx:rmi:///jndi/rmi://server2:1099/jmxrmi
Is there a new configuration option we need to set? Or perhaps
something didn't get updated properly in the 5.7 web console?
Thanks,
Travis
Re: Issues with stand alone web console in 5.7 using a master/slave architecture
Posted by Tim Bain <tb...@alumni.duke.edu>.
That's in ActiveMQ, which now uses Git at
https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=summary. See
https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=tree;f=activemq-web-console;h=04a8a829a6107322ba3eedf2ae5028a2120fdf93;hb=refs/heads/master
for the master branch of the web console.
Tim
On Fri, Jan 15, 2016 at 8:20 AM, ruffp <pr...@gmail.com> wrote:
> In fact in case I was not clear enough, the activemqweb application now
> behave correctly (meaning does not say "slave" whichever node I am querying
> but forwards to the active one) with the new servicemix (5.1.4) which use
> ActiveMQ 5.10.
>
> Now I don't even know where are the real source code of this bundle
> (activemqweb) in ServiceMix, in ActiveMQ or somewhere else. If you can
> point
> me to the right version (I found
> http://activemq.apache.org/activemq-web-console which gives 404 and
> another
> SVN repo in ActiveMQ where the version is 5.8)
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/Issues-with-stand-alone-web-console-in-5-7-using-a-master-slave-architecture-tp4661538p4706039.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>
Re: Issues with stand alone web console in 5.7 using a master/slave
architecture
Posted by ruffp <pr...@gmail.com>.
In fact in case I was not clear enough, the activemqweb application now
behave correctly (meaning does not say "slave" whichever node I am querying
but forwards to the active one) with the new servicemix (5.1.4) which use
ActiveMQ 5.10.
Now I don't even know where are the real source code of this bundle
(activemqweb) in ServiceMix, in ActiveMQ or somewhere else. If you can point
me to the right version (I found
http://activemq.apache.org/activemq-web-console which gives 404 and another
SVN repo in ActiveMQ where the version is 5.8)
--
View this message in context: http://activemq.2283324.n4.nabble.com/Issues-with-stand-alone-web-console-in-5-7-using-a-master-slave-architecture-tp4661538p4706039.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: Issues with stand alone web console in 5.7 using a master/slave architecture
Posted by Tim Bain <tb...@alumni.duke.edu>.
Note that what you're asking for is not a fix, but rather an enhancement.
Nothing's broken: the failover transport is successfully connecting to the
first responsive URI in the list, it's just that that's not the behavior
you actually want. Which simply means that the failover transport isn't
the appropriate way to try to achieve this, even if it did happen to work
in previous versions of ActiveMQ.
I personally don't know if anyone has implemented the capability you're
hoping for, though you could easily go through all the resolved JIRA
entries in each of the versions between your current one and the one you're
planning to upgrade to, to determine whether the new version will have the
capability.
Tim
On Jan 14, 2016 1:46 AM, "ruffp" <pr...@gmail.com> wrote:
> We have a serviceMix 4.3 which use the same ActiveMQ (embedded) and we also
> have the issue.
>
> In another deployment, we use ActiveMQ 5.10.x and apparently the
> activemqweb
> is working better (still few bugs when you browse queue from the slave
> instance) but quite more stable anyway.
>
> We plan to migrate our ActiveMQ to 5.10.x but in the meantime is there any
> fix for the activemqweb on version 5.7?
>
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/Issues-with-stand-alone-web-console-in-5-7-using-a-master-slave-architecture-tp4661538p4705980.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>
Re: Issues with stand alone web console in 5.7 using a master/slave
architecture
Posted by ruffp <pr...@gmail.com>.
We have a serviceMix 4.3 which use the same ActiveMQ (embedded) and we also
have the issue.
In another deployment, we use ActiveMQ 5.10.x and apparently the activemqweb
is working better (still few bugs when you browse queue from the slave
instance) but quite more stable anyway.
We plan to migrate our ActiveMQ to 5.10.x but in the meantime is there any
fix for the activemqweb on version 5.7?
--
View this message in context: http://activemq.2283324.n4.nabble.com/Issues-with-stand-alone-web-console-in-5-7-using-a-master-slave-architecture-tp4661538p4705980.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: Issues with stand alone web console in 5.7 using a master/slave architecture
Posted by Dejan Bosanac <de...@nighttale.net>.
Hi Travis,
I think the main change is that now slave broker exposes it's JMX and
shows that it's slave. In older version, the slave MBean server was
not running when broker was slave, so you were connected to the master
instead. I'm not sure there's a workaround for what are you trying to
achieve (other than using different consoles)
Regards
--
Dejan Bosanac
----------------------
Red Hat, Inc.
FuseSource is now part of Red Hat
dbosanac@redhat.com
Twitter: @dejanb
Blog: http://sensatic.net
ActiveMQ in Action: http://www.manning.com/snyder/
On Wed, Jan 9, 2013 at 4:30 PM, Travis Klotz <tr...@gmail.com> wrote:
> We are running the web console as stand alone application in a
> glassfish container using system properties to configure connection
> info to point to our two server master/slave setup. We do this so
> that we easily have a single URL to look at regardless of which server
> is the master and which is the slave.
>
> This has worked great for us in 5.5 and 5.6 but when upgrading to 5.7
> we have run into a problem. When server1 is the master, everything
> works great, but when server1 is the slave, we always get redirected
> to slave.jsp and are informed that the server is currently in slave
> mode. This in in spite of the fact that we have configured the web
> app to point to both servers using both JMX and JMS.
>
> For Example:
> -Dwebconsole.jms.url=failover:(tcp://server1:61616,tcp://server2:61616)
> -Dwebconsole.jmx.url=service:jmx:rmi:///jndi/rmi://server1:1099/jmxrmi,service:jmx:rmi:///jndi/rmi://server2:1099/jmxrmi
>
> Is there a new configuration option we need to set? Or perhaps
> something didn't get updated properly in the 5.7 web console?
>
> Thanks,
>
> Travis