q: could not connect to local URI: vm:?network=true...

[Paul Tomsic <pt...@yahoo.com>]

I'm having trouble starting 3 instances of activemq on
3 different boxes.

I'm getting "Could not connect to local URI:
vm:?network=true: java.io.IOException: Journal is
already opened by this application" when i try to
start the third box.

box 1,2 start up fine and i'm using multicast
discovery for the cluster.  1,2 find one another
successfully, but upon starting the 3rd  box, i get
the problem above.

I'm using RC2, and i've just recently tried this w/
the incubator-activemq snapshot w/o success.

thoughts/help appreciated

Re: q: could not connect to local URI: vm:?network=true...

[Javier Leyba <xl...@gmail.com>]

On 5/9/06, Paul Tomsic <pt...@yahoo.com> wrote:
>
>
> Your brokers are named in WHAT way?  what solved your
> problem?  renaming your brokers?  what did you rename
> them to?
>
> are you able to start more than 2 in a cluster?
> would you mind posting your startup xml file, please?
>
>


Ouch, sorry, let me explain.

I'm having the problem about the journal in use since january (AFAIR)
and never get help to the get the solution.

One month ago I downlaoded RC2 and changed my config files and the
problem was apparently solved. But few days ago I started to configure
network connectors and the problem comes back again.

Now I've saw this post and noticed that my brokers are named with an
underscore ie: brokerName="MASTER_UBUNTU" then I changed to
brokerName="MASTERUBUNTU" and the problem is solved.

Currently, my scenario consist in two servers (two physical different
machines) running JBoss with a Master/Slave brokers. Each one has a
master server and the slave of the other one to allow HA availability.

Nevertheless, after receive your mail I started a master broker and a
slave broker in the same machine, same JBoss instance, without
problems.


J




--
Javier Leyba
Barcelona - Spain
http://blog.leyba.com.ar

Re: q: could not connect to local URI: vm:?network=true...

[Paul Tomsic <pt...@yahoo.com>]

--- Javier Leyba <xl...@gmail.com> wrote:
> 
> Well, after several months finally I got a solution.
> 
> I've had same problem because all my brokers are
> named in this way. I
> renamed them and now I don't receive the error
> message about the
> journal in use.


Your brokers are named in WHAT way?  what solved your
problem?  renaming your brokers?  what did you rename
them to?

are you able to start more than 2 in a cluster?
would you mind posting your startup xml file, please?

Re: q: could not connect to local URI: vm:?network=true...

[Javier Leyba <xl...@gmail.com>]

On 5/8/06, Paul Tomsic <pt...@yahoo.com> wrote:
> i can post my config files if you think that'd help,
> but it really doesn't make sense that it's saying a
> journal's already in use when in fact, there's no
> other  activemq running on that box.
>
>


Well, after several months finally I got a solution.

I've had same problem because all my brokers are named in this way. I
renamed them and now I don't receive the error message about the
journal in use.

Thanks a lot !!!




--
Javier Leyba
Barcelona - Spain
http://blog.leyba.com.ar

Re: q: could not connect to local URI: vm:?network=true...

[Paul Tomsic <pt...@yahoo.com>]

eghost01.dev.mycompany.com

eghost02.dev.mycompany.com

...


--- Hiram Chirino <hi...@hiramchirino.com> wrote:

> What's the hostnames of the boxes?
> 
> We have had previous reports of similar errors when
> the machine is
> configured with hostname like "broker_box" which is
> an invalid
> hostname (according to the URI specification).
> 
> Regards,
> Hiram
> 
> On 5/8/06, Paul Tomsic <pt...@yahoo.com> wrote:
> > I'm having trouble starting 3 instances of
> activemq on
> > 3 different boxes.
> >
> > I'm getting "Could not connect to local URI:
> > vm:?network=true: java.io.IOException: Journal is
> > already opened by this application" when i try to
> > start the third box.
> >
> > box 1,2 start up fine and i'm using multicast
> > discovery for the cluster.  1,2 find one another
> > successfully, but upon starting the 3rd  box, i
> get
> > the problem above.
> >
> > I'm using RC2, and i've just recently tried this
> w/
> > the incubator-activemq snapshot w/o success.
> >
> > thoughts/help appreciated
> >
> >
> 
> 
> --
> Regards,
> Hiram
> 

Re: q: could not connect to local URI: vm:?network=true...

[Paul Tomsic <pt...@yahoo.com>]

i can post my config files if you think that'd help,
but it really doesn't make sense that it's saying a 
journal's already in use when in fact, there's no
other  activemq running on that box.



--- Hiram Chirino <hi...@hiramchirino.com> wrote:

> What's the hostnames of the boxes?
> 
> We have had previous reports of similar errors when
> the machine is
> configured with hostname like "broker_box" which is
> an invalid
> hostname (according to the URI specification).
> 
> Regards,
> Hiram
> 
> On 5/8/06, Paul Tomsic <pt...@yahoo.com> wrote:
> > I'm having trouble starting 3 instances of
> activemq on
> > 3 different boxes.
> >
> > I'm getting "Could not connect to local URI:
> > vm:?network=true: java.io.IOException: Journal is
> > already opened by this application" when i try to
> > start the third box.
> >
> > box 1,2 start up fine and i'm using multicast
> > discovery for the cluster.  1,2 find one another
> > successfully, but upon starting the 3rd  box, i
> get
> > the problem above.
> >
> > I'm using RC2, and i've just recently tried this
> w/
> > the incubator-activemq snapshot w/o success.
> >
> > thoughts/help appreciated
> >
> >
> 
> 
> --
> Regards,
> Hiram
> 

Re: q: could not connect to local URI: vm:?network=true...

[Hiram Chirino <hi...@hiramchirino.com>]

What's the hostnames of the boxes?

We have had previous reports of similar errors when the machine is
configured with hostname like "broker_box" which is an invalid
hostname (according to the URI specification).

Regards,
Hiram

On 5/8/06, Paul Tomsic <pt...@yahoo.com> wrote:
> I'm having trouble starting 3 instances of activemq on
> 3 different boxes.
>
> I'm getting "Could not connect to local URI:
> vm:?network=true: java.io.IOException: Journal is
> already opened by this application" when i try to
> start the third box.
>
> box 1,2 start up fine and i'm using multicast
> discovery for the cluster.  1,2 find one another
> successfully, but upon starting the 3rd  box, i get
> the problem above.
>
> I'm using RC2, and i've just recently tried this w/
> the incubator-activemq snapshot w/o success.
>
> thoughts/help appreciated
>
>


--
Regards,
Hiram