2:56 a.m.
Jiri Pechanec created JGRP-1511:
-----------------------------------
Summary: Relaying node does not connect bridge bus if another cluster exists
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to localhost,
second with mixed UDP stack bound to localhost and RELAY2 bridge to the other machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
2:58 a.m.
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Jiri Pechanec updated JGRP-1511:
--------------------------------
Attachment: relay.tgz
Relaying node does not connect bridge bus if another cluster exists
-------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:14 a.m.
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Jiri Pechanec updated JGRP-1511:
--------------------------------
Attachment: relay-logs.tgz
Logs attached.
The logs demonstrates another behaviour. If the UDP-only node is stopped, then the
bridging node starts the bridging stack as well.
Relaying node does not connect bridge bus if another cluster exists
-------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:32 a.m.
New subject: [JBoss JIRA] (JGRP-1511) RELAY2: Relaying node does not connect bridge bus if another cluster exists
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Jiri Pechanec updated JGRP-1511:
--------------------------------
Summary: RELAY2: Relaying node does not connect bridge bus if another cluster exists
(was: Relaying node does not connect bridge bus if another cluster exists)
RELAY2: Relaying node does not connect bridge bus if another cluster
exists
---------------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
9:48 a.m.
New subject: [JBoss JIRA] (JGRP-1511) RELAY2: Relaying node does not connect bridge bus if another cluster exists
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Bela Ban commented on JGRP-1511:
--------------------------------
Jiri, you need to be running udp-relay on *both* local channels. The first process started
will create the local and bridge channels, the second will only create the local channel.
When the first dies or leaves, the second node will then create the bridge channel.
Perhaps we should have a chat before you start creating JIRAs (I can show you everything
via elluminate on my local box) ?
RELAY2: Relaying node does not connect bridge bus if another cluster
exists
---------------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
9:48 a.m.
New subject: [JBoss JIRA] (JGRP-1511) RELAY2: Relaying node does not connect bridge bus if another cluster exists
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Bela Ban commented on JGRP-1511:
--------------------------------
In essence, only the current coordinator starts the bridge channel, but all members of a
given (local) site need to have the *same* config.
RELAY2: Relaying node does not connect bridge bus if another cluster
exists
---------------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:21 a.m.
New subject: [JBoss JIRA] (JGRP-1511) RELAY2: Relaying node does not connect bridge bus if another cluster exists
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Jiri Pechanec commented on JGRP-1511:
-------------------------------------
Hi,
I'll retry it with your tips. If that's the issue I'd recommend to
emphasize/re-phrase it in the documentation.
I have mislead myself by applying the future architecture of Infinispan to JGroups
configuration https://community.jboss.org/wiki/DesignForCrossSiteReplication
RELAY2: Relaying node does not connect bridge bus if another cluster
exists
---------------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:27 a.m.
New subject: [JBoss JIRA] (JGRP-1511) RELAY2: Relaying node does not connect bridge bus if another cluster exists
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Bela Ban commented on JGRP-1511:
--------------------------------
Have you seen http://www.jgroups.org/manual-3.x/html/user-advanced.html#Relay2Advanced ?
Jiri, you can always contact me on IRC (#jgroups @ irc.freenode.net) if you have
questions.
RELAY2: Relaying node does not connect bridge bus if another cluster
exists
---------------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:27 a.m.
New subject: [JBoss JIRA] (JGRP-1511) RELAY2: Relaying node does not connect bridge bus if another cluster exists
[
https://issues.jboss.org/browse/JGRP-1511?page=com.atlassian.jira.plugin....
]
Jiri Pechanec closed JGRP-1511.
-------------------------------
Resolution: Rejected
Misconfiguration
RELAY2: Relaying node does not connect bridge bus if another cluster
exists
---------------------------------------------------------------------------
Key: JGRP-1511
URL: https://issues.jboss.org/browse/JGRP-1511
Project: JGroups
Issue Type: Bug
Affects Versions: 3.2
Environment: jgroups-3.2.0.Alpha1.jar
Fedora 14
java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.10) (fedora-55.1.9.10.fc14-x86_64)
OpenJDK 64-Bit Server VM (build 19.0-b09, mixed mode)
Reporter: Jiri Pechanec
Assignee: Bela Ban
Attachments: relay-logs.tgz, relay.tgz
I tried to simulate geo cluster using RELAY2.
Configuration files are attached and were developed from vanilla udp.xml and tcp.xml.
The cluster setup
two machines
10.34.3.135
10.34.40.20
Each machines run two instances of Chat demo. One with plain UDP stack bound to
localhost, second with mixed UDP stack bound to localhost and RELAY2 bridge to the other
machine.
When relaying instance is started first it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-16703, cluster=ChatCluster, physical address=127.0.0.1:47331
-------------------------------------------------------------------
-------------------------------------------------------------------
GMS: address=_jpechane-16703:NTB, cluster=global, physical address=10.34.3.135:7800
-------------------------------------------------------------------
** view: [jpechane-16703|0] [jpechane-16703]
{noformat}
and the geo cluster is properly formed.
If a UDP-only instance is started first and then the relaying instace, it prints
{noformat}
-------------------------------------------------------------------
GMS: address=jpechane-36521, cluster=ChatCluster, physical address=127.0.0.1:52324
-------------------------------------------------------------------
** view: [jpechane-64562|1] [jpechane-64562, jpechane-36521]
>
{noformat}
As you can see only local part of the cluster and not the geo bus was started.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
4313
days inactive
4313
days old
8 comments
2 participants
participants (2)
-
Bela Ban (JIRA) -
Jiri Pechanec (JIRA)