Re: [VOTE] Accept Code donations: Sheldon and Chatterbox

2017-09-09 Thread Elder Moraes
+1

+999,999,999 for a Sheldon Cooper class! ;-)

Elder

Twitter: @elderjava 
Blog: http://eldermoraes.com




2017-09-09 16:02 GMT-03:00 David Blevins :

>
> > On Sep 9, 2017, at 1:15 AM, Mark Struberg 
> wrote:
> >
> > But we should not forget to check the IP.
>
> Answers below, but I recall the Incubator IP checks being a little more
> formal.  At least there used to be a document to fill out.
>
> > Is this a clean room implementation? Have there been substantial
> contributions from others or does Tomitribe at least have a CLA for them?
>
> Clean room.
>
> > * As far as I've seen all of the committers are ASF TomEE committers,
> right? check!
>
> Minus one, Daniel Cunha.
>
> > * Just checked a few files, all of them are "Licensed to the Apache
> Software Foundation (ASF) under one or more..." ALv2, so this is fine from
> the beginning, even while not hosted at the ASF. David, are all files that
> way? Just checked a few and they have all been fine. check!
>
> Correct.  As Jean-Louis notes we have a checkstyle to enforce the header,
> so we’re good.
>
> > PS: Shedon Cooper might cause come trademark issues ;)
>
> We can maybe have a “cooper" command in there :)  Or make people submit
> their passwords three times before they can come in.
>
>
> -David
>
>


Re: [DISCUSS] Code donations: Sheldon and Chatterbox

2017-08-09 Thread Elder Moraes
You guys really rock!!

Elder

Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com




2017-08-09 23:42 GMT-03:00 David Blevins :

> > On Aug 9, 2017, at 7:18 PM, Elder Moraes  wrote:
> >
> > Don't know if I'm allowed to answer... and my apologies if I wasn't...
> :-)
>
> For 17 years this project has maintained a "users first, committers
> second” spirit.  Committers are one part of and servants to the community.
> So the short answer is, absolutely, *everyone* who has their heart in the
> project is welcome and *encouraged* to chime in.
>
> You don’t grow a project by treating people who are not committers like
> they are not committers :)  Treat people like they are committers and they
> soon will be.
>
> So a direct Thank You for having the bravery to speak up and I hope others
> follow your example :)
>
> Great to hear your support, Elder.
>
>
> -David
>
>


Re: [DISCUSS] Code donations: Sheldon and Chatterbox

2017-08-09 Thread Elder Moraes
Don't know if I'm allowed to answer... and my apologies if I wasn't... :-)

But from the point of view of the community, I think it would be awesome.
TomEE has been my default EE server for a long time and it would be great
to see it even more powerful.

Didn't know these projects, but just had a look at them and love it!


Cheers,


Elder 

Twitter: @elderjava 
Blog: http://eldermoraes.com




2017-08-09 22:58 GMT-03:00 David Blevins :

> Hi All,
>
> I’d like to open the topic of potentially donating a couple components
> we’ve built over at tomitribe.io to our beloved TomEE project:
>
>  - https://tomitribe.io/projects/sheldon  projects/sheldon>
>  - https://tomitribe.io/projects/chatterbox  projects/chatterbox>
>
> There are couple motivating factors:
>
>  - show the TomEE ecosystem is a bit bigger than just the server
>  - provide a clear pattern for “crazy new ideas” to start here that could
> potentially benefit TomEE or other EE related-projects
>  - more opportunities to earn commit
>  - give the community a boost
>
> We went from EJB container to EE server.  I’d be great to take another
> step and see us go to EE ecosystem.
>
> Thoughts?
>
>
> --
> David Blevins
> http://twitter.com/dblevins
> http://www.tomitribe.com
>
>


Re: Building a Static Discovery Cluster

2017-07-10 Thread Elder Moraes
Hello David, Romain and others,

I've just figured it out! Now I have a real static cluster, with session
being shared between nodes and avoiding unexpected members to joining in.

There are 3 key points:

   1. channelStartOptions="3" (to avoid using multicast);
   2. Receiver and Member using the same port;
   3. Turn one of the members into a LocalMember (instead of Member). It
   will work as the Replication Listener and Receiver Server.

By doing this you'll be able to see, after all the nodes are up and
running, the Group Channel showing confirmation of each node added. And
once you try to add another with unexpected IP, it is solemnly ignored...
;-)

How can I turn it into some documentation for Tomee? I have all the
details...



Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-10 14:40 GMT-03:00 Elder Moraes :

> Hey David!
>
> I've sent to the Tomcat list, but still didn't get any reply.
>
> Anyway I'm still trying to figure it out by myself and will be a pleasure
> to contribute with some docs once I do it.
>
>
> Tks!
>
> Twitter: @elderjava <https://twitter.com/elderjava>
> Blog: http://eldermoraes.com
>
>
>
>
> 2017-07-09 2:50 GMT-03:00 David Blevins :
>
>>  Hi Elder! just checking to see if you got the help you were looking for.
>>
>>  When you get it figured out any new docs would be amazing.
>>
>> On Fri, Jul 7, 2017 at 9:57 AM Elder Moraes 
>> wrote:
>>
>> > Ok, will try them.
>> >
>> > Thanks!
>> >
>> >
>> > Twitter: @elderjava <https://twitter.com/elderjava>
>> > Blog: http://eldermoraes.com
>> >
>> >
>> >
>> > 2017-07-06 18:02 GMT-03:00 Romain Manni-Bucau :
>> >
>> > > outch, no more, didnt resetup a cluster since 2 years. But if you
>> push a
>> > > tomee-maven-plugin sample it would be quick to help.
>> > >
>> > > Side note: Tomcat list can also be more relevant since they fully own
>> > that
>> > > part.
>> > >
>> > >
>> > > Romain Manni-Bucau
>> > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
>> > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
>> > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
>> > > rmannibucau> |
>> > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
>> > > <https://javaeefactory-rmannibucau.rhcloud.com>
>> > >
>> > > 2017-07-06 22:52 GMT+02:00 Elder Moraes :
>> > >
>> > > > Do you have some working example? Don't need to be dockerized.
>> > > >
>> > > >
>> > > > Twitter: @elderjava <https://twitter.com/elderjava>
>> > > > Blog: http://eldermoraes.com
>> > > >
>> > > >
>> > > > 2017-07-06 17:05 GMT-03:00 Romain Manni-Bucau <
>> rmannibu...@gmail.com>:
>> > > >
>> > > > > yes should help, you can
>> > > > > check org.apache.catalina.tribes.mem
>> bership.McastServiceImpl#start,
>> > > > > org.apache.catalina.tribes.group.ChannelCoordinator#internalStart
>> > > > > and org.apache.catalina.tribes.group.ChannelCoordinator#sendMess
>> age
>> > to
>> > > > see
>> > > > > it in action. It basically disable multicast services (threads)
>> and
>> > > > switch
>> > > > > it to send messages
>> > > > >
>> > > > >
>> > > > > Romain Manni-Bucau
>> > > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
>> > > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
>> > > > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
>> > > > > rmannibucau> |
>> > > > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE
>> Factory
>> > > > > <https://javaeefactory-rmannibucau.rhcloud.com>
>> > > > >
>> > > > > 2017-07-06 20:53 GMT+02:00 Elder Moraes :
>> > > > >
>> > > > > > The new logs are here:
>> > > > > >
>> > > > > > https://www.dropbox.com/s/1bdz8vf9pw0h42v/catalina.2017-
>> > > 07-06.log?dl=0
>> > > > > >
>> > > > > > Yes, it shows everything! ;-)
>> > >

Re: Building a Static Discovery Cluster

2017-07-10 Thread Elder Moraes
Hey David!

I've sent to the Tomcat list, but still didn't get any reply.

Anyway I'm still trying to figure it out by myself and will be a pleasure
to contribute with some docs once I do it.


Tks!

Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com




2017-07-09 2:50 GMT-03:00 David Blevins :

>  Hi Elder! just checking to see if you got the help you were looking for.
>
>  When you get it figured out any new docs would be amazing.
>
> On Fri, Jul 7, 2017 at 9:57 AM Elder Moraes 
> wrote:
>
> > Ok, will try them.
> >
> > Thanks!
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> >
> > 2017-07-06 18:02 GMT-03:00 Romain Manni-Bucau :
> >
> > > outch, no more, didnt resetup a cluster since 2 years. But if you push
> a
> > > tomee-maven-plugin sample it would be quick to help.
> > >
> > > Side note: Tomcat list can also be more relevant since they fully own
> > that
> > > part.
> > >
> > >
> > > Romain Manni-Bucau
> > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > rmannibucau> |
> > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > >
> > > 2017-07-06 22:52 GMT+02:00 Elder Moraes :
> > >
> > > > Do you have some working example? Don't need to be dockerized.
> > > >
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > > 2017-07-06 17:05 GMT-03:00 Romain Manni-Bucau  >:
> > > >
> > > > > yes should help, you can
> > > > > check org.apache.catalina.tribes.membership.McastServiceImpl#
> start,
> > > > > org.apache.catalina.tribes.group.ChannelCoordinator#internalStart
> > > > > and org.apache.catalina.tribes.group.ChannelCoordinator#
> sendMessage
> > to
> > > > see
> > > > > it in action. It basically disable multicast services (threads) and
> > > > switch
> > > > > it to send messages
> > > > >
> > > > >
> > > > > Romain Manni-Bucau
> > > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > > > rmannibucau> |
> > > > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE
> Factory
> > > > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > > > >
> > > > > 2017-07-06 20:53 GMT+02:00 Elder Moraes :
> > > > >
> > > > > > The new logs are here:
> > > > > >
> > > > > > https://www.dropbox.com/s/1bdz8vf9pw0h42v/catalina.2017-
> > > 07-06.log?dl=0
> > > > > >
> > > > > > Yes, it shows everything! ;-)
> > > > > >
> > > > > > The last four lines shows the unexpected member being added to
> the
> > > > > cluster.
> > > > > >
> > > > > > Don't seems like is a Docker problem, as the new member is
> allowed
> > by
> > > > the
> > > > > > cluster.
> > > > > >
> > > > > > The multicast setup that I've mentioned before has nothing to do?
> > > > > >
> > > > > >
> > > > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > > > Blog: http://eldermoraes.com
> > > > > >
> > > > > >
> > > > > >
> > > > > > 2017-07-06 15:42 GMT-03:00 Romain Manni-Bucau <
> > rmannibu...@gmail.com
> > > >:
> > > > > >
> > > > > > > What are the new logs? ;)
> > > > > > >
> > > > > > > Tomcat logs everything showing if connection works, if members
> > are
> > > > > > > accepted/connecting etc...
> > > > > > >
> > > > > > > From what i saw the docker setup is n

Re: Building a Static Discovery Cluster

2017-07-07 Thread Elder Moraes
Ok, will try them.

Thanks!


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-06 18:02 GMT-03:00 Romain Manni-Bucau :

> outch, no more, didnt resetup a cluster since 2 years. But if you push a
> tomee-maven-plugin sample it would be quick to help.
>
> Side note: Tomcat list can also be more relevant since they fully own that
> part.
>
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://blog-rmannibucau.rhcloud.com> | Old Blog
> <http://rmannibucau.wordpress.com> | Github <https://github.com/
> rmannibucau> |
> LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> <https://javaeefactory-rmannibucau.rhcloud.com>
>
> 2017-07-06 22:52 GMT+02:00 Elder Moraes :
>
> > Do you have some working example? Don't need to be dockerized.
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> > 2017-07-06 17:05 GMT-03:00 Romain Manni-Bucau :
> >
> > > yes should help, you can
> > > check org.apache.catalina.tribes.membership.McastServiceImpl#start,
> > > org.apache.catalina.tribes.group.ChannelCoordinator#internalStart
> > > and org.apache.catalina.tribes.group.ChannelCoordinator#sendMessage to
> > see
> > > it in action. It basically disable multicast services (threads) and
> > switch
> > > it to send messages
> > >
> > >
> > > Romain Manni-Bucau
> > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > rmannibucau> |
> > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > >
> > > 2017-07-06 20:53 GMT+02:00 Elder Moraes :
> > >
> > > > The new logs are here:
> > > >
> > > > https://www.dropbox.com/s/1bdz8vf9pw0h42v/catalina.2017-
> 07-06.log?dl=0
> > > >
> > > > Yes, it shows everything! ;-)
> > > >
> > > > The last four lines shows the unexpected member being added to the
> > > cluster.
> > > >
> > > > Don't seems like is a Docker problem, as the new member is allowed by
> > the
> > > > cluster.
> > > >
> > > > The multicast setup that I've mentioned before has nothing to do?
> > > >
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > >
> > > > 2017-07-06 15:42 GMT-03:00 Romain Manni-Bucau  >:
> > > >
> > > > > What are the new logs? ;)
> > > > >
> > > > > Tomcat logs everything showing if connection works, if members are
> > > > > accepted/connecting etc...
> > > > >
> > > > > From what i saw the docker setup is not matching the conf. Maybe
> > start
> > > by
> > > > > making it running without docker before dockerizing it.
> > > > >
> > > > > Side note: maybe you want local member in your cluster (depends
> your
> > > goal
> > > > > but was often done IIRC).
> > > > >
> > > > > Le 6 juil. 2017 19:50, "Elder Moraes"  a
> > > écrit :
> > > > >
> > > > > > Thanks guys.
> > > > > >
> > > > > > I'm not using docker-compose.
> > > > > >
> > > > > > Actually I've just fixed that errors logged just by creating a
> > > > > > docker-network for the cluster and defining a static IP for each
> > > > member.
> > > > > So
> > > > > > the cluster is working well at this point.
> > > > > >
> > > > > > But it is still working as a dynamic one. I've found somewhere
> > that I
> > > > > > should use channelStartOptions="3" to disable multicast and avoid
> > > > > > unexpected members. But once I do it, the session is not shared
> > thru
> > > > the
> > > > > > cluster anymore.
> > > > > >
> > > > > > Any thoughts?
> > > > > >
> > > > > >
> >

Re: Building a Static Discovery Cluster

2017-07-06 Thread Elder Moraes
Do you have some working example? Don't need to be dockerized.


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com


2017-07-06 17:05 GMT-03:00 Romain Manni-Bucau :

> yes should help, you can
> check org.apache.catalina.tribes.membership.McastServiceImpl#start,
> org.apache.catalina.tribes.group.ChannelCoordinator#internalStart
> and org.apache.catalina.tribes.group.ChannelCoordinator#sendMessage to see
> it in action. It basically disable multicast services (threads) and switch
> it to send messages
>
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://blog-rmannibucau.rhcloud.com> | Old Blog
> <http://rmannibucau.wordpress.com> | Github <https://github.com/
> rmannibucau> |
> LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> <https://javaeefactory-rmannibucau.rhcloud.com>
>
> 2017-07-06 20:53 GMT+02:00 Elder Moraes :
>
> > The new logs are here:
> >
> > https://www.dropbox.com/s/1bdz8vf9pw0h42v/catalina.2017-07-06.log?dl=0
> >
> > Yes, it shows everything! ;-)
> >
> > The last four lines shows the unexpected member being added to the
> cluster.
> >
> > Don't seems like is a Docker problem, as the new member is allowed by the
> > cluster.
> >
> > The multicast setup that I've mentioned before has nothing to do?
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> >
> > 2017-07-06 15:42 GMT-03:00 Romain Manni-Bucau :
> >
> > > What are the new logs? ;)
> > >
> > > Tomcat logs everything showing if connection works, if members are
> > > accepted/connecting etc...
> > >
> > > From what i saw the docker setup is not matching the conf. Maybe start
> by
> > > making it running without docker before dockerizing it.
> > >
> > > Side note: maybe you want local member in your cluster (depends your
> goal
> > > but was often done IIRC).
> > >
> > > Le 6 juil. 2017 19:50, "Elder Moraes"  a
> écrit :
> > >
> > > > Thanks guys.
> > > >
> > > > I'm not using docker-compose.
> > > >
> > > > Actually I've just fixed that errors logged just by creating a
> > > > docker-network for the cluster and defining a static IP for each
> > member.
> > > So
> > > > the cluster is working well at this point.
> > > >
> > > > But it is still working as a dynamic one. I've found somewhere that I
> > > > should use channelStartOptions="3" to disable multicast and avoid
> > > > unexpected members. But once I do it, the session is not shared thru
> > the
> > > > cluster anymore.
> > > >
> > > > Any thoughts?
> > > >
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > > 2017-07-06 10:03 GMT-03:00 Thiago Veronezi :
> > > >
> > > > > If you are using docker-compose, check the depends_on option. It
> will
> > > > > guarantee you have your name available. It does not guarantee you
> > have
> > > > the
> > > > > service fully up and running though. If you need that too, you will
> > > need
> > > > > the script to wait for the service behind that name to be ready.
> > > > >
> > > > > []s,
> > > > > Thiago.
> > > > >
> > > > >
> > > > > On 5 Jul 2017 6:16 pm, "Romain Manni-Bucau"  >
> > > > wrote:
> > > > >
> > > > > > looks like docker dns is not working that well and registration
> > > fails -
> > > > > or
> > > > > > the startup is not ordered properly but result lead to the same.
> > > Often
> > > > > saw
> > > > > > docker-compose having a small bash loop to wait for the host/port
> > > being
> > > > > > accessible. Maybe that's what you need (yes it is ugly)
> > > > > >
> > > > > >
> > > > > > Romain Manni-Bucau
> > > > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > > > > <http://rmannibucau

Re: Building a Static Discovery Cluster

2017-07-06 Thread Elder Moraes
The new logs are here:

https://www.dropbox.com/s/1bdz8vf9pw0h42v/catalina.2017-07-06.log?dl=0

Yes, it shows everything! ;-)

The last four lines shows the unexpected member being added to the cluster.

Don't seems like is a Docker problem, as the new member is allowed by the
cluster.

The multicast setup that I've mentioned before has nothing to do?


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-06 15:42 GMT-03:00 Romain Manni-Bucau :

> What are the new logs? ;)
>
> Tomcat logs everything showing if connection works, if members are
> accepted/connecting etc...
>
> From what i saw the docker setup is not matching the conf. Maybe start by
> making it running without docker before dockerizing it.
>
> Side note: maybe you want local member in your cluster (depends your goal
> but was often done IIRC).
>
> Le 6 juil. 2017 19:50, "Elder Moraes"  a écrit :
>
> > Thanks guys.
> >
> > I'm not using docker-compose.
> >
> > Actually I've just fixed that errors logged just by creating a
> > docker-network for the cluster and defining a static IP for each member.
> So
> > the cluster is working well at this point.
> >
> > But it is still working as a dynamic one. I've found somewhere that I
> > should use channelStartOptions="3" to disable multicast and avoid
> > unexpected members. But once I do it, the session is not shared thru the
> > cluster anymore.
> >
> > Any thoughts?
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> > 2017-07-06 10:03 GMT-03:00 Thiago Veronezi :
> >
> > > If you are using docker-compose, check the depends_on option. It will
> > > guarantee you have your name available. It does not guarantee you have
> > the
> > > service fully up and running though. If you need that too, you will
> need
> > > the script to wait for the service behind that name to be ready.
> > >
> > > []s,
> > > Thiago.
> > >
> > >
> > > On 5 Jul 2017 6:16 pm, "Romain Manni-Bucau" 
> > wrote:
> > >
> > > > looks like docker dns is not working that well and registration
> fails -
> > > or
> > > > the startup is not ordered properly but result lead to the same.
> Often
> > > saw
> > > > docker-compose having a small bash loop to wait for the host/port
> being
> > > > accessible. Maybe that's what you need (yes it is ugly)
> > > >
> > > >
> > > > Romain Manni-Bucau
> > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > > rmannibucau> |
> > > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > > >
> > > > 2017-07-06 0:13 GMT+02:00 Elder Moraes :
> > > >
> > > > > I've got the log and put here:
> > > > >
> > > > > https://www.dropbox.com/s/yxt3ym1ucv1hwb3/catalina.2017-
> > 07-05.log?dl=0
> > > > >
> > > > >
> > > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > > Blog: http://eldermoraes.com
> > > > >
> > > > >
> > > > >
> > > > > 2017-07-05 18:56 GMT-03:00 Romain Manni-Bucau <
> rmannibu...@gmail.com
> > >:
> > > > >
> > > > > > interesting, would be great to have debug log of that clustering
> > part
> > > > > > (something like org.apache.catalina.tribes.level = FINE), think
> > the
> > > > > static
> > > > > > members are actually excluded of the cluster reading this error
> > > > > >
> > > > > >
> > > > > > Romain Manni-Bucau
> > > > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > > > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > > > > rmannibucau> |
> > > > > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE
> > Factory
> > > > > > <https://javaeefactory-rmannibucau.rhcloud.com>

Re: Building a Static Discovery Cluster

2017-07-06 Thread Elder Moraes
Thanks guys.

I'm not using docker-compose.

Actually I've just fixed that errors logged just by creating a
docker-network for the cluster and defining a static IP for each member. So
the cluster is working well at this point.

But it is still working as a dynamic one. I've found somewhere that I
should use channelStartOptions="3" to disable multicast and avoid
unexpected members. But once I do it, the session is not shared thru the
cluster anymore.

Any thoughts?


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com


2017-07-06 10:03 GMT-03:00 Thiago Veronezi :

> If you are using docker-compose, check the depends_on option. It will
> guarantee you have your name available. It does not guarantee you have the
> service fully up and running though. If you need that too, you will need
> the script to wait for the service behind that name to be ready.
>
> []s,
> Thiago.
>
>
> On 5 Jul 2017 6:16 pm, "Romain Manni-Bucau"  wrote:
>
> > looks like docker dns is not working that well and registration fails -
> or
> > the startup is not ordered properly but result lead to the same. Often
> saw
> > docker-compose having a small bash loop to wait for the host/port being
> > accessible. Maybe that's what you need (yes it is ugly)
> >
> >
> > Romain Manni-Bucau
> > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > rmannibucau> |
> > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > <https://javaeefactory-rmannibucau.rhcloud.com>
> >
> > 2017-07-06 0:13 GMT+02:00 Elder Moraes :
> >
> > > I've got the log and put here:
> > >
> > > https://www.dropbox.com/s/yxt3ym1ucv1hwb3/catalina.2017-07-05.log?dl=0
> > >
> > >
> > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > Blog: http://eldermoraes.com
> > >
> > >
> > >
> > > 2017-07-05 18:56 GMT-03:00 Romain Manni-Bucau :
> > >
> > > > interesting, would be great to have debug log of that clustering part
> > > > (something like org.apache.catalina.tribes.level = FINE), think the
> > > static
> > > > members are actually excluded of the cluster reading this error
> > > >
> > > >
> > > > Romain Manni-Bucau
> > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > > rmannibucau> |
> > > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > > >
> > > > 2017-07-05 23:53 GMT+02:00 Elder Moraes :
> > > >
> > > > > Hmm, why not the tcpfailuredetector as a root (/>) instead of
> > wrapping
> > > > the
> > > > > static membership one?
> > > > >
> > > > > Done e got this error:
> > > > >
> > > > > 05-Jul-2017 21:51:32.845 WARNING
> > > > > [GroupChannel-Heartbeat[Catalina-Channel]-1]
> > > > > sun.reflect.NativeMethodAccessorImpl.invoke Unable to perform
> > > heartbeat
> > > > on
> > > > > the TcpFailureDetector.
> > > > >  java.lang.ArrayIndexOutOfBoundsException: 0
> > > > > at
> > > > > org.apache.catalina.tribes.membership.MemberImpl.
> > > > > hashCode(MemberImpl.java:570)
> > > > > at java.util.HashMap.hash(HashMap.java:338)
> > > > > at java.util.HashMap.containsKey(HashMap.java:595)
> > > > > at
> > > > > org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.
> > > > > performBasicCheck(TcpFailureDetector.java:258)
> > > > > at
> > > > > org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.
> > > > > checkMembers(TcpFailureDetector.java:225)
> > > > > at
> > > > > org.apache.catalina.tribes.group.interceptors.
> > > > > TcpFailureDetector.heartbeat(TcpFailureDetector.java:218)
> > > > > at
> > > > > org.apache.catalina.tribes.group.ChannelInterceptorBase.heartbeat(
> > > > > ChannelInterceptorBase.java:100)
> > > > >   

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
I've got the log and put here:

https://www.dropbox.com/s/yxt3ym1ucv1hwb3/catalina.2017-07-05.log?dl=0


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-05 18:56 GMT-03:00 Romain Manni-Bucau :

> interesting, would be great to have debug log of that clustering part
> (something like org.apache.catalina.tribes.level = FINE), think the static
> members are actually excluded of the cluster reading this error
>
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://blog-rmannibucau.rhcloud.com> | Old Blog
> <http://rmannibucau.wordpress.com> | Github <https://github.com/
> rmannibucau> |
> LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> <https://javaeefactory-rmannibucau.rhcloud.com>
>
> 2017-07-05 23:53 GMT+02:00 Elder Moraes :
>
> > Hmm, why not the tcpfailuredetector as a root (/>) instead of wrapping
> the
> > static membership one?
> >
> > Done e got this error:
> >
> > 05-Jul-2017 21:51:32.845 WARNING
> > [GroupChannel-Heartbeat[Catalina-Channel]-1]
> > sun.reflect.NativeMethodAccessorImpl.invoke Unable to perform heartbeat
> on
> > the TcpFailureDetector.
> >  java.lang.ArrayIndexOutOfBoundsException: 0
> > at
> > org.apache.catalina.tribes.membership.MemberImpl.
> > hashCode(MemberImpl.java:570)
> > at java.util.HashMap.hash(HashMap.java:338)
> > at java.util.HashMap.containsKey(HashMap.java:595)
> > at
> > org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.
> > performBasicCheck(TcpFailureDetector.java:258)
> > at
> > org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.
> > checkMembers(TcpFailureDetector.java:225)
> > at
> > org.apache.catalina.tribes.group.interceptors.
> > TcpFailureDetector.heartbeat(TcpFailureDetector.java:218)
> > at
> > org.apache.catalina.tribes.group.ChannelInterceptorBase.heartbeat(
> > ChannelInterceptorBase.java:100)
> > at
> > org.apache.catalina.tribes.group.GroupChannel.heartbeat(
> > GroupChannel.java:161)
> > at
> > org.apache.catalina.tribes.group.GroupChannel$HeartbeatThread.run(
> > GroupChannel.java:719)
> >
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> >
> > 2017-07-05 18:47 GMT-03:00 Romain Manni-Bucau :
> >
> > > Hmm, why not the tcpfailuredetector as a root (/>) instead of wrapping
> > the
> > > static membership one?
> > >
> > > Also did you manage to get debug log of tomcat cluster? it helps a lot
> in
> > > general
> > >
> > >
> > > Romain Manni-Bucau
> > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > rmannibucau> |
> > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > >
> > > 2017-07-05 23:43 GMT+02:00 Elder Moraes :
> > >
> > > > Did it. No errors found.
> > > >
> > > > If you want to have a look at the server.xml. Also I've undone the
> > > changes
> > > > I've made before.
> > > >
> > > > https://www.dropbox.com/s/p3iknyku2rb3f3w/server.xml?dl=0
> > > >
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > > 2017-07-05 17:08 GMT-03:00 Romain Manni-Bucau  >:
> > > >
> > > > > Gmail formatting doesnt help much but think it can miss some
> closing
> > > > tags.
> > > > > Autoformat it in an ide, it should show it.
> > > > >
> > > > > Le 5 juil. 2017 21:28, "Elder Moraes"  a
> > > écrit :
> > > > >
> > > > > > Actually I can't... :-(
> > > > > >
> > > > > > But if you could point me where I should look, maybe I can figure
> > it
> > > > out.
> > > > > >
> > > > > > Is that code that I've sent here correct? No issue?
> > > > > >
> > > > > > Did I put the StaticMembershipInterceptor at right place?
> > > > > >
> > >

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Hmm, why not the tcpfailuredetector as a root (/>) instead of wrapping the
static membership one?

Done e got this error:

05-Jul-2017 21:51:32.845 WARNING
[GroupChannel-Heartbeat[Catalina-Channel]-1]
sun.reflect.NativeMethodAccessorImpl.invoke Unable to perform heartbeat on
the TcpFailureDetector.
 java.lang.ArrayIndexOutOfBoundsException: 0
at
org.apache.catalina.tribes.membership.MemberImpl.hashCode(MemberImpl.java:570)
at java.util.HashMap.hash(HashMap.java:338)
at java.util.HashMap.containsKey(HashMap.java:595)
at
org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.performBasicCheck(TcpFailureDetector.java:258)
at
org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.checkMembers(TcpFailureDetector.java:225)
at
org.apache.catalina.tribes.group.interceptors.TcpFailureDetector.heartbeat(TcpFailureDetector.java:218)
at
org.apache.catalina.tribes.group.ChannelInterceptorBase.heartbeat(ChannelInterceptorBase.java:100)
at
org.apache.catalina.tribes.group.GroupChannel.heartbeat(GroupChannel.java:161)
at
org.apache.catalina.tribes.group.GroupChannel$HeartbeatThread.run(GroupChannel.java:719)



Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-05 18:47 GMT-03:00 Romain Manni-Bucau :

> Hmm, why not the tcpfailuredetector as a root (/>) instead of wrapping the
> static membership one?
>
> Also did you manage to get debug log of tomcat cluster? it helps a lot in
> general
>
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://blog-rmannibucau.rhcloud.com> | Old Blog
> <http://rmannibucau.wordpress.com> | Github <https://github.com/
> rmannibucau> |
> LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> <https://javaeefactory-rmannibucau.rhcloud.com>
>
> 2017-07-05 23:43 GMT+02:00 Elder Moraes :
>
> > Did it. No errors found.
> >
> > If you want to have a look at the server.xml. Also I've undone the
> changes
> > I've made before.
> >
> > https://www.dropbox.com/s/p3iknyku2rb3f3w/server.xml?dl=0
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> > 2017-07-05 17:08 GMT-03:00 Romain Manni-Bucau :
> >
> > > Gmail formatting doesnt help much but think it can miss some closing
> > tags.
> > > Autoformat it in an ide, it should show it.
> > >
> > > Le 5 juil. 2017 21:28, "Elder Moraes"  a
> écrit :
> > >
> > > > Actually I can't... :-(
> > > >
> > > > But if you could point me where I should look, maybe I can figure it
> > out.
> > > >
> > > > Is that code that I've sent here correct? No issue?
> > > >
> > > > Did I put the StaticMembershipInterceptor at right place?
> > > >
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > >
> > > > 2017-07-05 16:21 GMT-03:00 Romain Manni-Bucau  >:
> > > >
> > > > > Do you think you can do a setup without docker? Would be easier to
> > > check
> > > > > but sounds like a config issue.
> > > > >
> > > > > Le 5 juil. 2017 21:18, "Elder Moraes"  a
> > > écrit :
> > > > >
> > > > > > If you are talking about removing the McastService, I've already
> > done
> > > > and
> > > > > > didn't work. The cluster still works as a dynamic one.
> > > > > >
> > > > > >
> > > > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > > > Blog: http://eldermoraes.com
> > > > > >
> > > > > >
> > > > > > 2017-07-05 16:08 GMT-03:00 Romain Manni-Bucau <
> > rmannibu...@gmail.com
> > > >:
> > > > > >
> > > > > > > Think you ask to find node with this service so it behaves as
> > > > > configured.
> > > > > > > Remove it for static clusters.
> > > > > > >
> > > > > > > Le 5 juil. 2017 20:43, "Elder Moraes" 
> a
> > > > > écrit :
> > > > > > >
> > > > > > > Sorry, I don't get it...
> > > > > > >
> > > > > > >
> > > > > > > Twitter: @elderjava <https://t

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Did it. No errors found.

If you want to have a look at the server.xml. Also I've undone the changes
I've made before.

https://www.dropbox.com/s/p3iknyku2rb3f3w/server.xml?dl=0


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com


2017-07-05 17:08 GMT-03:00 Romain Manni-Bucau :

> Gmail formatting doesnt help much but think it can miss some closing tags.
> Autoformat it in an ide, it should show it.
>
> Le 5 juil. 2017 21:28, "Elder Moraes"  a écrit :
>
> > Actually I can't... :-(
> >
> > But if you could point me where I should look, maybe I can figure it out.
> >
> > Is that code that I've sent here correct? No issue?
> >
> > Did I put the StaticMembershipInterceptor at right place?
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> >
> > 2017-07-05 16:21 GMT-03:00 Romain Manni-Bucau :
> >
> > > Do you think you can do a setup without docker? Would be easier to
> check
> > > but sounds like a config issue.
> > >
> > > Le 5 juil. 2017 21:18, "Elder Moraes"  a
> écrit :
> > >
> > > > If you are talking about removing the McastService, I've already done
> > and
> > > > didn't work. The cluster still works as a dynamic one.
> > > >
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > > 2017-07-05 16:08 GMT-03:00 Romain Manni-Bucau  >:
> > > >
> > > > > Think you ask to find node with this service so it behaves as
> > > configured.
> > > > > Remove it for static clusters.
> > > > >
> > > > > Le 5 juil. 2017 20:43, "Elder Moraes"  a
> > > écrit :
> > > > >
> > > > > Sorry, I don't get it...
> > > > >
> > > > >
> > > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > > Blog: http://eldermoraes.com
> > > > >
> > > > >
> > > > >
> > > > > 2017-07-05 15:38 GMT-03:00 Romain Manni-Bucau <
> rmannibu...@gmail.com
> > >:
> > > > >
> > > > > > (disclaimer: read quickly on phone so can be wrong)
> > > > > >
> > > > > > seems you use membership discovery, can't it be simply that?
> > > > > >
> > > > > > 2017-07-05 20:07 GMT+02:00 Elder Moraes  >:
> > > > > >
> > > > > > > Sure! Here they are:
> > > > > > >
> > > > > > > Name: Catalina:type=Cluster,component=Member,name="tcp://{172,
> > 17,
> > > > 0,
> > > > > > > 3}:4000"
> > > > > > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > > > > > memberAliveTime: 59124
> > > > > > > suspect: false
> > > > > > > udpPort: -1
> > > > > > > local: false
> > > > > > > securePort: -1
> > > > > > > hostname: {172, 17, 0, 3}
> > > > > > > port: 4000
> > > > > > > serviceStartTime: 0
> > > > > > > ready: true
> > > > > > > failing: false
> > > > > > > name: tcp://{172, 17, 0, 3}:4000
> > > > > > > msgCount: 0
> > > > > > >
> > > > > > > Name: Catalina:type=Cluster,component=Member,name="tcp://1
> > > > > 72.17.0.4:4000
> > > > > > "
> > > > > > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > > > > > memberAliveTime: 8094
> > > > > > > suspect: false
> > > > > > > udpPort: -1
> > > > > > > local: true
> > > > > > > securePort: -1
> > > > > > > hostname: 172.17.0.4
> > > > > > > port: 4000
> > > > > > > serviceStartTime: 1499277570908
> > > > > > > ready: true
> > > > > > > failing: false
> > > > > > > name: tcp://172.17.0.4:4000
> > > > > > > msgCount: 108
> > > > > > >
> > > > > > > Name: Catalina:type=Cluster,component=Member,name="tcp://{172,
> >

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Actually I can't... :-(

But if you could point me where I should look, maybe I can figure it out.

Is that code that I've sent here correct? No issue?

Did I put the StaticMembershipInterceptor at right place?


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-05 16:21 GMT-03:00 Romain Manni-Bucau :

> Do you think you can do a setup without docker? Would be easier to check
> but sounds like a config issue.
>
> Le 5 juil. 2017 21:18, "Elder Moraes"  a écrit :
>
> > If you are talking about removing the McastService, I've already done and
> > didn't work. The cluster still works as a dynamic one.
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> > 2017-07-05 16:08 GMT-03:00 Romain Manni-Bucau :
> >
> > > Think you ask to find node with this service so it behaves as
> configured.
> > > Remove it for static clusters.
> > >
> > > Le 5 juil. 2017 20:43, "Elder Moraes"  a
> écrit :
> > >
> > > Sorry, I don't get it...
> > >
> > >
> > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > Blog: http://eldermoraes.com
> > >
> > >
> > >
> > > 2017-07-05 15:38 GMT-03:00 Romain Manni-Bucau :
> > >
> > > > (disclaimer: read quickly on phone so can be wrong)
> > > >
> > > > seems you use membership discovery, can't it be simply that?
> > > >
> > > > 2017-07-05 20:07 GMT+02:00 Elder Moraes :
> > > >
> > > > > Sure! Here they are:
> > > > >
> > > > > Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17,
> > 0,
> > > > > 3}:4000"
> > > > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > > > memberAliveTime: 59124
> > > > > suspect: false
> > > > > udpPort: -1
> > > > > local: false
> > > > > securePort: -1
> > > > > hostname: {172, 17, 0, 3}
> > > > > port: 4000
> > > > > serviceStartTime: 0
> > > > > ready: true
> > > > > failing: false
> > > > > name: tcp://{172, 17, 0, 3}:4000
> > > > > msgCount: 0
> > > > >
> > > > > Name: Catalina:type=Cluster,component=Member,name="tcp://1
> > > 72.17.0.4:4000
> > > > "
> > > > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > > > memberAliveTime: 8094
> > > > > suspect: false
> > > > > udpPort: -1
> > > > > local: true
> > > > > securePort: -1
> > > > > hostname: 172.17.0.4
> > > > > port: 4000
> > > > > serviceStartTime: 1499277570908
> > > > > ready: true
> > > > > failing: false
> > > > > name: tcp://172.17.0.4:4000
> > > > > msgCount: 108
> > > > >
> > > > > Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17,
> > 0,
> > > > > 5}:4000"
> > > > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > > > memberAliveTime: 52614
> > > > > suspect: false
> > > > > udpPort: -1
> > > > > local: false
> > > > > securePort: -1
> > > > > hostname: {172, 17, 0, 5}
> > > > > port: 4000
> > > > > serviceStartTime: 0
> > > > > ready: true
> > > > > failing: false
> > > > > name: tcp://{172, 17, 0, 5}:4000
> > > > > msgCount: 0
> > > > >
> > > > > Name: Catalina:type=Cluster
> > > > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > > > channelStartOptions: 15
> > > > > stateName: STARTED
> > > > > channelSendOptions: 8
> > > > > domain: Catalina
> > > > > clusterName: Catalina
> > > > > heartbeatBackgroundEnabled: false
> > > > > notifyLifecycleListenerOnFailure: false
> > > > > objectName: Catalina:type=Cluster
> > > > >
> > > > > Name: Catalina:type=Cluster,component=Deployer
> > > > > modelerType: org.apache.catalina.mbeans.ClassNameMBean
> > > > > watchDir: /tmp/war-listen/
> > > > > processDeployFrequency: 2
> > > > >

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
If you are talking about removing the McastService, I've already done and
didn't work. The cluster still works as a dynamic one.


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com


2017-07-05 16:08 GMT-03:00 Romain Manni-Bucau :

> Think you ask to find node with this service so it behaves as configured.
> Remove it for static clusters.
>
> Le 5 juil. 2017 20:43, "Elder Moraes"  a écrit :
>
> Sorry, I don't get it...
>
>
> Twitter: @elderjava <https://twitter.com/elderjava>
> Blog: http://eldermoraes.com
>
>
>
> 2017-07-05 15:38 GMT-03:00 Romain Manni-Bucau :
>
> > (disclaimer: read quickly on phone so can be wrong)
> >
> > seems you use membership discovery, can't it be simply that?
> >
> > 2017-07-05 20:07 GMT+02:00 Elder Moraes :
> >
> > > Sure! Here they are:
> > >
> > > Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17, 0,
> > > 3}:4000"
> > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > memberAliveTime: 59124
> > > suspect: false
> > > udpPort: -1
> > > local: false
> > > securePort: -1
> > > hostname: {172, 17, 0, 3}
> > > port: 4000
> > > serviceStartTime: 0
> > > ready: true
> > > failing: false
> > > name: tcp://{172, 17, 0, 3}:4000
> > > msgCount: 0
> > >
> > > Name: Catalina:type=Cluster,component=Member,name="tcp://1
> 72.17.0.4:4000
> > "
> > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > memberAliveTime: 8094
> > > suspect: false
> > > udpPort: -1
> > > local: true
> > > securePort: -1
> > > hostname: 172.17.0.4
> > > port: 4000
> > > serviceStartTime: 1499277570908
> > > ready: true
> > > failing: false
> > > name: tcp://172.17.0.4:4000
> > > msgCount: 108
> > >
> > > Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17, 0,
> > > 5}:4000"
> > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > memberAliveTime: 52614
> > > suspect: false
> > > udpPort: -1
> > > local: false
> > > securePort: -1
> > > hostname: {172, 17, 0, 5}
> > > port: 4000
> > > serviceStartTime: 0
> > > ready: true
> > > failing: false
> > > name: tcp://{172, 17, 0, 5}:4000
> > > msgCount: 0
> > >
> > > Name: Catalina:type=Cluster
> > > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > > channelStartOptions: 15
> > > stateName: STARTED
> > > channelSendOptions: 8
> > > domain: Catalina
> > > clusterName: Catalina
> > > heartbeatBackgroundEnabled: false
> > > notifyLifecycleListenerOnFailure: false
> > > objectName: Catalina:type=Cluster
> > >
> > > Name: Catalina:type=Cluster,component=Deployer
> > > modelerType: org.apache.catalina.mbeans.ClassNameMBean
> > > watchDir: /tmp/war-listen/
> > > processDeployFrequency: 2
> > > tempDir: /tmp/war-temp/
> > > maxValidTime: 300
> > > deployDir: /tmp/war-deploy/
> > > watchEnabled: false
> > >
> > >
> > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > Blog: http://eldermoraes.com
> > >
> > >
> > >
> > > 2017-07-05 13:40 GMT-03:00 Romain Manni-Bucau :
> > >
> > > > Hmm, can you have a look into JMX on some tomee instances to check
> the
> > > > tomcat cluster config please?
> > > >
> > > >
> > > > Romain Manni-Bucau
> > > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > > rmannibucau> |
> > > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > > >
> > > > 2017-07-05 18:34 GMT+02:00 Elder Moraes :
> > > >
> > > > > Hi Romain,
> > > > >
> > > > > Actually my problem is right the opposite (I forget to mention -
> > > sorry!):
> > > > > I'm trying to use a static discovery, but it is working as a
> dynamic
> > > one.
> > > > > In other words, when I run another container (say, 

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Sorry, I don't get it...


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-05 15:38 GMT-03:00 Romain Manni-Bucau :

> (disclaimer: read quickly on phone so can be wrong)
>
> seems you use membership discovery, can't it be simply that?
>
> 2017-07-05 20:07 GMT+02:00 Elder Moraes :
>
> > Sure! Here they are:
> >
> > Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17, 0,
> > 3}:4000"
> > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > memberAliveTime: 59124
> > suspect: false
> > udpPort: -1
> > local: false
> > securePort: -1
> > hostname: {172, 17, 0, 3}
> > port: 4000
> > serviceStartTime: 0
> > ready: true
> > failing: false
> > name: tcp://{172, 17, 0, 3}:4000
> > msgCount: 0
> >
> > Name: Catalina:type=Cluster,component=Member,name="tcp://172.17.0.4:4000
> "
> > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > memberAliveTime: 8094
> > suspect: false
> > udpPort: -1
> > local: true
> > securePort: -1
> > hostname: 172.17.0.4
> > port: 4000
> > serviceStartTime: 1499277570908
> > ready: true
> > failing: false
> > name: tcp://172.17.0.4:4000
> > msgCount: 108
> >
> > Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17, 0,
> > 5}:4000"
> > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > memberAliveTime: 52614
> > suspect: false
> > udpPort: -1
> > local: false
> > securePort: -1
> > hostname: {172, 17, 0, 5}
> > port: 4000
> > serviceStartTime: 0
> > ready: true
> > failing: false
> > name: tcp://{172, 17, 0, 5}:4000
> > msgCount: 0
> >
> > Name: Catalina:type=Cluster
> > modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
> > channelStartOptions: 15
> > stateName: STARTED
> > channelSendOptions: 8
> > domain: Catalina
> > clusterName: Catalina
> > heartbeatBackgroundEnabled: false
> > notifyLifecycleListenerOnFailure: false
> > objectName: Catalina:type=Cluster
> >
> > Name: Catalina:type=Cluster,component=Deployer
> > modelerType: org.apache.catalina.mbeans.ClassNameMBean
> > watchDir: /tmp/war-listen/
> > processDeployFrequency: 2
> > tempDir: /tmp/war-temp/
> > maxValidTime: 300
> > deployDir: /tmp/war-deploy/
> > watchEnabled: false
> >
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> >
> > 2017-07-05 13:40 GMT-03:00 Romain Manni-Bucau :
> >
> > > Hmm, can you have a look into JMX on some tomee instances to check the
> > > tomcat cluster config please?
> > >
> > >
> > > Romain Manni-Bucau
> > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > rmannibucau> |
> > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > >
> > > 2017-07-05 18:34 GMT+02:00 Elder Moraes :
> > >
> > > > Hi Romain,
> > > >
> > > > Actually my problem is right the opposite (I forget to mention -
> > sorry!):
> > > > I'm trying to use a static discovery, but it is working as a dynamic
> > one.
> > > > In other words, when I run another container (say, host4) it joins
> into
> > > the
> > > > cluster.
> > > >
> > > >
> > > >
> > > > Cheers,
> > > >
> > > > Twitter: @elderjava <https://twitter.com/elderjava>
> > > > Blog: http://eldermoraes.com
> > > >
> > > >
> > > >
> > > > 2017-07-05 12:38 GMT-03:00 Romain Manni-Bucau  >:
> > > >
> > > > > Hi Elder,
> > > > >
> > > > > are all the binding address (host)/ports opened between your
> > > containers?
> > > > > (can you telnet between them)
> > > > > do you use docker-compose to create the cluster (it helps a bit)?
> > > > > did you activate the debug log of tomcat clustering? it generally
> > helps
> > > > >
> > > > >
> > > > > Romain Manni-Bucau
> > > > > @rmannibucau <https://t

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Sure! Here they are:

Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17, 0,
3}:4000"
modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
memberAliveTime: 59124
suspect: false
udpPort: -1
local: false
securePort: -1
hostname: {172, 17, 0, 3}
port: 4000
serviceStartTime: 0
ready: true
failing: false
name: tcp://{172, 17, 0, 3}:4000
msgCount: 0

Name: Catalina:type=Cluster,component=Member,name="tcp://172.17.0.4:4000"
modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
memberAliveTime: 8094
suspect: false
udpPort: -1
local: true
securePort: -1
hostname: 172.17.0.4
port: 4000
serviceStartTime: 1499277570908
ready: true
failing: false
name: tcp://172.17.0.4:4000
msgCount: 108

Name: Catalina:type=Cluster,component=Member,name="tcp://{172, 17, 0,
5}:4000"
modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
memberAliveTime: 52614
suspect: false
udpPort: -1
local: false
securePort: -1
hostname: {172, 17, 0, 5}
port: 4000
serviceStartTime: 0
ready: true
failing: false
name: tcp://{172, 17, 0, 5}:4000
msgCount: 0

Name: Catalina:type=Cluster
modelerType: org.apache.tomcat.util.modeler.BaseModelMBean
channelStartOptions: 15
stateName: STARTED
channelSendOptions: 8
domain: Catalina
clusterName: Catalina
heartbeatBackgroundEnabled: false
notifyLifecycleListenerOnFailure: false
objectName: Catalina:type=Cluster

Name: Catalina:type=Cluster,component=Deployer
modelerType: org.apache.catalina.mbeans.ClassNameMBean
watchDir: /tmp/war-listen/
processDeployFrequency: 2
tempDir: /tmp/war-temp/
maxValidTime: 300
deployDir: /tmp/war-deploy/
watchEnabled: false


Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-05 13:40 GMT-03:00 Romain Manni-Bucau :

> Hmm, can you have a look into JMX on some tomee instances to check the
> tomcat cluster config please?
>
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://blog-rmannibucau.rhcloud.com> | Old Blog
> <http://rmannibucau.wordpress.com> | Github <https://github.com/
> rmannibucau> |
> LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> <https://javaeefactory-rmannibucau.rhcloud.com>
>
> 2017-07-05 18:34 GMT+02:00 Elder Moraes :
>
> > Hi Romain,
> >
> > Actually my problem is right the opposite (I forget to mention - sorry!):
> > I'm trying to use a static discovery, but it is working as a dynamic one.
> > In other words, when I run another container (say, host4) it joins into
> the
> > cluster.
> >
> >
> >
> > Cheers,
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
> >
> >
> > 2017-07-05 12:38 GMT-03:00 Romain Manni-Bucau :
> >
> > > Hi Elder,
> > >
> > > are all the binding address (host)/ports opened between your
> containers?
> > > (can you telnet between them)
> > > do you use docker-compose to create the cluster (it helps a bit)?
> > > did you activate the debug log of tomcat clustering? it generally helps
> > >
> > >
> > > Romain Manni-Bucau
> > > @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> > > <https://blog-rmannibucau.rhcloud.com> | Old Blog
> > > <http://rmannibucau.wordpress.com> | Github <https://github.com/
> > > rmannibucau> |
> > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> > > <https://javaeefactory-rmannibucau.rhcloud.com>
> > >
> > > 2017-07-05 17:30 GMT+02:00 Elder Moraes :
> > >
> > > > Hi everyone!
> > > >
> > > > Could you help me? I've been struggling to build a cluster using
> Docker
> > > and
> > > > Tomee based on the static discovery membership.
> > > >
> > > > Docker image: tomee:8-jre-7.0.3-plume
> > > >
> > > > Please refer to the "Cluster" node described bellow. Could you point
> me
> > > > what is wrong?
> > > >
> > > >  > > >channelSendOptions="8">
> > > >
> > > >> > >expireSessionsOnShutdown="false"
> > > >notifyListenersOnReplication="true"/>
> > > >
> > > >   
> > > >
> > > >> > > className="org.apache.catalina.tribes.membership.McastService"
> > > >   address="228.0.0.4"
> > > >   port="45564"
> > > > 

Re: Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Hi Romain,

Actually my problem is right the opposite (I forget to mention - sorry!):
I'm trying to use a static discovery, but it is working as a dynamic one.
In other words, when I run another container (say, host4) it joins into the
cluster.



Cheers,

Twitter: @elderjava <https://twitter.com/elderjava>
Blog: http://eldermoraes.com



2017-07-05 12:38 GMT-03:00 Romain Manni-Bucau :

> Hi Elder,
>
> are all the binding address (host)/ports opened between your containers?
> (can you telnet between them)
> do you use docker-compose to create the cluster (it helps a bit)?
> did you activate the debug log of tomcat clustering? it generally helps
>
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://blog-rmannibucau.rhcloud.com> | Old Blog
> <http://rmannibucau.wordpress.com> | Github <https://github.com/
> rmannibucau> |
> LinkedIn <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
> <https://javaeefactory-rmannibucau.rhcloud.com>
>
> 2017-07-05 17:30 GMT+02:00 Elder Moraes :
>
> > Hi everyone!
> >
> > Could you help me? I've been struggling to build a cluster using Docker
> and
> > Tomee based on the static discovery membership.
> >
> > Docker image: tomee:8-jre-7.0.3-plume
> >
> > Please refer to the "Cluster" node described bellow. Could you point me
> > what is wrong?
> >
> >  >channelSendOptions="8">
> >
> >>expireSessionsOnShutdown="false"
> >notifyListenersOnReplication="true"/>
> >
> >   
> >
> >> className="org.apache.catalina.tribes.membership.McastService"
> >   address="228.0.0.4"
> >   port="45564"
> >   frequency="500"
> >   dropTime="3000"/>
> >> className="org.apache.catalina.tribes.transport.nio.NioReceiver"
> > address="auto"
> > port="4000"
> > autoBind="100"
> > selectorTimeout="5000"
> > maxThreads="6"/>
> >
> >> className="org.apache.catalina.tribes.transport.ReplicationTransmitter">
> >  > className="org.apache.catalina.tribes.transport.nio.
> > PooledParallelSender"/>
> >   
> >
> >> className="org.apache.catalina.tribes.group.interceptors.
> > TcpFailureDetector">
> >
> >  > className="org.apache.catalina.tribes.group.interceptors.
> > StaticMembershipInterceptor">
> > > className="org.apache.catalina.tribes.membership.StaticMember"
> >   port="5678"
> >   securePort="-1"
> >   host="tomee-soujava1"
> >   domain="staging-cluster"
> >
> > uniqueId="{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1}"/>
> >> className="org.apache.catalina.tribes.membership.StaticMember"
> >   port="5678"
> >   securePort="-1"
> >   host="tomee-soujava2"
> >
>  domain="staging-cluster"
> >
> > uniqueId="{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2}"/>
> >> className="org.apache.catalina.tribes.membership.StaticMember"
> >   port="5678"
> >   securePort="-1"
> >   host="tomee-soujava3"
> >
>  domain="staging-cluster"
> >
> > uniqueId="{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,3}"/>
> >  
> >
> >   
> >> className="org.apache.catalina.tribes.group.interceptors.
> > MessageDispatchInterceptor"/>
> >
> >   
> >
> >>  filter=""/>
> >> className="org.apache.catalina.ha.session.JvmRouteBinderValve"/>
> >
> >> tempDir="/tmp/war-temp/"
> > deployDir="/tmp/war-deploy/"
> > watchDir="/tmp/war-listen/"
> > watchEnabled="false"/>
> >
> >> className="org.apache.catalina.ha.session.ClusterSessionListener"/>
> > 
> >
> >
> >
> > Best regards,
> >
> > Elder
> >
> > Twitter: @elderjava <https://twitter.com/elderjava>
> > Blog: http://eldermoraes.com
> >
>


Building a Static Discovery Cluster

2017-07-05 Thread Elder Moraes
Hi everyone!

Could you help me? I've been struggling to build a cluster using Docker and
Tomee based on the static discovery membership.

Docker image: tomee:8-jre-7.0.3-plume

Please refer to the "Cluster" node described bellow. Could you point me
what is wrong?



  

  

  
  

  

  

  


   
  
  
 

  
  

  

  
  

  

  




Best regards,

Elder

Twitter: @elderjava 
Blog: http://eldermoraes.com