6:26 a.m.
AS ejb ha restart would fail
----------------------------
Key: JBAS-4574
URL: http://jira.jboss.com/jira/browse/JBAS-4574
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: Clustering
Affects Versions: JBossAS-4.0.5.GA
Reporter: Ben Wang
Assigned To: Ben Wang
This is an issue discovered during a pre-sales opportunity. Basically, it can be
reproduced like this:
1. Use a web tier (calling remote slsb ejb from jsp, e.g.) with 2 ejb tiers (for
clustering, say, ejb1, ejb2)
2. access it first to ejb1
3. shutdown ejb1, and access it again will failover to ejb2
4. shutdown ejb2 and bring up ejb1, should produce error traces (I don't have now).
Will attach it here later.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
6:40 a.m.
New subject: [JBoss JIRA] Closed: (JBAS-4574) AS ejb ha restart would fail
[ http://jira.jboss.com/jira/browse/JBAS-4574?page=all ]
Adrian Brock closed JBAS-4574.
------------------------------
Resolution: Duplicate Issue
Ben. Please discuss problems in the forums before opening bug reports.
Or at least use search first before duplicating bug reports.
This is a duplicate of JBAS-1476
When you shutdown the server EJB2 you shutdown the whole cluster,
it ceased to exist.
When you started the server EJB1 you created a new cluster.
All clients are orhpaned by the shutdown of the whole cluster.
You either need to reget the client stubs (including the naming stub)
or introduce the RetryInterceptor in the client side to do this automagically:
http://wiki.jboss.org/wiki/Wiki.jsp?page=RetryInterceptor
JBAS-1476 would allow the client to automatically know when a cluster
was shutdown completely (assuming you are using a transport that
can recover from its entire view becoming invalid - i.e. not RMI).
AS ejb ha restart would fail
----------------------------
Key: JBAS-4574
URL: http://jira.jboss.com/jira/browse/JBAS-4574
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: Clustering
Affects Versions: JBossAS-4.0.5.GA
Reporter: Ben Wang
Assigned To: Ben Wang
This is an issue discovered during a pre-sales opportunity. Basically, it can be
reproduced like this:
1. Use a web tier (calling remote slsb ejb from jsp, e.g.) with 2 ejb tiers (for
clustering, say, ejb1, ejb2)
2. access it first to ejb1
3. shutdown ejb1, and access it again will failover to ejb2
4. shutdown ejb2 and bring up ejb1, should produce error traces (I don't have now).
Will attach it here later.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
7:01 a.m.
New subject: [JBoss JIRA] Commented: (JBAS-4574) AS ejb ha restart would fail
[ http://jira.jboss.com/jira/browse/JBAS-4574?page=comments#action_12369712 ]
Ben Wang commented on JBAS-4574:
--------------------------------
Adrian, yes I do remeber 1475 issue. But I am not sure by including view id will solve the
problem completely here.
4.0.5 currently uses default SingleRetryInterceptor for HA SLSB but it doesn't work in
this case. Reason being that we cache the server info in the NamingContext implicitly so
unless we flush it out, retry won't work even when you do jndi look up for the new
naming stub.
I have thought about posting this to the forum. But I have discussed this with Brian
before and I am creating a branch first to look into the solution further (I have an ad
hoc one) before going to the forum, so that's the reason for this Jira.
AS ejb ha restart would fail
----------------------------
Key: JBAS-4574
URL: http://jira.jboss.com/jira/browse/JBAS-4574
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: Clustering
Affects Versions: JBossAS-4.0.5.GA
Reporter: Ben Wang
Assigned To: Ben Wang
This is an issue discovered during a pre-sales opportunity. Basically, it can be
reproduced like this:
1. Use a web tier (calling remote slsb ejb from jsp, e.g.) with 2 ejb tiers (for
clustering, say, ejb1, ejb2)
2. access it first to ejb1
3. shutdown ejb1, and access it again will failover to ejb2
4. shutdown ejb2 and bring up ejb1, should produce error traces (I don't have now).
Will attach it here later.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
7:31 a.m.
New subject: [JBoss JIRA] Commented: (JBAS-4574) AS ejb ha restart would fail
[ http://jira.jboss.com/jira/browse/JBAS-4574?page=comments#action_12369716 ]
Adrian Brock commented on JBAS-4574:
------------------------------------
Let's please not start another long speculative discussion about what the problem
might be
in the release notes.
The forums are there to decide what needs fixing and only once that work is done should
JIRA
issues be created.
Anything else just makes it impossible for users to understand
a) what the original was
b) what was changed to resolve it
Claiming you've had private discussions with somebody else isn't helping at all,
nobody else knows what those were or can validate the conclusion is correct.
Use the forums or go on work on a non open source project.
AS ejb ha restart would fail
----------------------------
Key: JBAS-4574
URL: http://jira.jboss.com/jira/browse/JBAS-4574
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: Clustering
Affects Versions: JBossAS-4.0.5.GA
Reporter: Ben Wang
Assigned To: Ben Wang
This is an issue discovered during a pre-sales opportunity. Basically, it can be
reproduced like this:
1. Use a web tier (calling remote slsb ejb from jsp, e.g.) with 2 ejb tiers (for
clustering, say, ejb1, ejb2)
2. access it first to ejb1
3. shutdown ejb1, and access it again will failover to ejb2
4. shutdown ejb2 and bring up ejb1, should produce error traces (I don't have now).
Will attach it here later.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
7:45 a.m.
New subject: [JBoss JIRA] Commented: (JBAS-4574) AS ejb ha restart would fail
[ http://jira.jboss.com/jira/browse/JBAS-4574?page=comments#action_12369718 ]
Adrian Brock commented on JBAS-4574:
------------------------------------
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4066629#...
AS ejb ha restart would fail
----------------------------
Key: JBAS-4574
URL: http://jira.jboss.com/jira/browse/JBAS-4574
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: Clustering
Affects Versions: JBossAS-4.0.5.GA
Reporter: Ben Wang
Assigned To: Ben Wang
This is an issue discovered during a pre-sales opportunity. Basically, it can be
reproduced like this:
1. Use a web tier (calling remote slsb ejb from jsp, e.g.) with 2 ejb tiers (for
clustering, say, ejb1, ejb2)
2. access it first to ejb1
3. shutdown ejb1, and access it again will failover to ejb2
4. shutdown ejb2 and bring up ejb1, should produce error traces (I don't have now).
Will attach it here later.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:03 a.m.
New subject: [JBoss JIRA] Commented: (JBAS-4574) AS ejb ha restart would fail
[ http://jira.jboss.com/jira/browse/JBAS-4574?page=comments#action_12369896 ]
Ben Wang commented on JBAS-4574:
--------------------------------
It is you that want to make this into long discussion! I don't mind to discuss it in
the forum. But Jira is not just about release note. It is about potential issue as well. I
created this jira to match my 4.0.5 branch so it is easier to track later. I don't
even assign it to any release to fix yet.
AS ejb ha restart would fail
----------------------------
Key: JBAS-4574
URL: http://jira.jboss.com/jira/browse/JBAS-4574
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: Clustering
Affects Versions: JBossAS-4.0.5.GA
Reporter: Ben Wang
Assigned To: Ben Wang
This is an issue discovered during a pre-sales opportunity. Basically, it can be
reproduced like this:
1. Use a web tier (calling remote slsb ejb from jsp, e.g.) with 2 ejb tiers (for
clustering, say, ejb1, ejb2)
2. access it first to ejb1
3. shutdown ejb1, and access it again will failover to ejb2
4. shutdown ejb2 and bring up ejb1, should produce error traces (I don't have now).
Will attach it here later.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
6877
days inactive
6878
days old
5 comments
2 participants
participants (2)
-
Adrian Brock (JIRA) -
Ben Wang (JIRA)