Are you using the latest release (CR2)?
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3946854#3946854
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3946854
---
All
Yes, it happens with CR2, too!
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3947294#3947294
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3947294
---
All the advan
According to your stack trace, it can't find messaging-users.properties:
anonymous wrote :
| Failed to load users/passwords/role files
| java.io.IOException: No properties file: messaging-users.properties or
defaults:
|
Can you verify this file is in the deploy/jboss-messaging.sar direc
Both files (messaging-users.properties and messaging-roles.properties) are
there and readable.
The point is: this happens after several thousands of messages have been
sucessfully delivered and suddenly its broken. There is no special event in our
application and jboss is not running out of mem
anonymous wrote : The point is: this happens after several thousands of
messages have been sucessfully delivered and suddenly its broken. There is no
special event in our application and jboss is not running out of memory. So the
exceptions seem to be the symptom of some more obscure cause...
Yes. With CR2 it happens always after about 30 minutes.
As I mentioned earlier, the traffic is at about 200 messages per second, there
are topics, queues and ejb3 mdb involved, but no clustering. OS is Linux and
java is Bea jrockit.
View the original post :
http://www.jboss.com/index.html?modu
Can you replicate in a test case (or similar) and send to me or Ovidiu so we
can investigate further?
Thanks
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3947342#3947342
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=
It seems to be a memory leak. When we try it with sun jdk 1.5.0_06 instead of
bea jrockit, jboss runs clearly out of memory.
Our applications uses lots of ObjectMessages in both queues and topics.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3947389#3947389
Ok, a few more questions:
You say the "traffic" is 200 msgs/sec. Do you mean you are *sending* 200
msgs/sec, or *consuming* 200 msgs/sec, or perhaps a *throughput* of 200
msgs/sec.
If you're sending messages to a queue, but not consuming them, then at some
point you are going to run out of mem
The throughput is about 200 msgs/s (please don't nail me down on this ;-)
On every queue sits an ejb3 mdb, so the messages should get consumed.
Furthermore we don't detect any "lost" messages.
The size of the messages is several KB.
They are all non-persistent.
DB is mysql.
View the original p
Just because the messages should get consumed doesn't mean they are being
consumed.
And even if they are being consumed it doesn't mean they are being consumed at
the same rate as they are being sent.
The most likely cause here is you are sending messages fasting than you are
consuming them, c
timfox wrote:
anonymous wrote :
| The most likely cause here is you are sending messages fasting than you are
consuming them, consequently they will build up on the queue or subscription,
until the system runs out of storage for them.
|
Oops, jboss-messaging just let the VM run out of me
"mwelss" wrote : timfox wrote:
|
| Oops, jboss-messaging just let the VM run out of memory?? Is that your idea
of good desgin? If that is not a bug, but a feature as you put it, we obviously
must rethink our plan to use jboss-messaging!
|
|
If configured properly JBoss Messaging will
I have a testcase that replicates it. The cause seems to be a topic selector
whith an "OR" operator.
I will send the exmaple by email to you, it needs a messaging config based on
ejb3.
Thanks.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3948095#3948095
R
Great, thanks.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3948098#3948098
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3948098
---
All the advantages of Linux Ma
I haven't started looking at this yet, but just to get me started, you're
saying that if you change the selector so it no longer has an "OR" expression,
the leak disappears?
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3948130#3948130
Reply to the post :
h
Yes, exactly! The "OR" makes the difference.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3948179#3948179
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3948179
---
http://jira.jboss.org/jira/browse/JBMESSAGING-405
Clebert is taking a look at it.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3948245#3948245
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3948245
--
I have added a testcase on jms, go to jboss-head/jms/test and call ant
memory-leak-tests
(On windows only for now, using a JDK 5).
At the end of the test, I have these following unexpected objects. It doesn't
seem a memory leak, just something that I didn't expect at this point.
| [junit]
Clebert,
did you try my testcase? Can you reproduce the problem?
Best regards,
Martin
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3949276#3949276
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3949276
Hi Martin-
Hopefully we can investigate this soon.
Many of us are travelling this week and next week is JBoss World in Vegas so
time (as usual) is a precious commodity right now.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3949661#3949661
Reply to t
I tried to use the expression part only in a memory-leak testcase, and didn't
have any luck. (it's not leaking)
I will expand to everything you have in your testcase. At this point I don't
think the problem is the expressions.
take a look on MemoryLeakTest.java on jms project at jboss-head.
Vi
I'am sorry I didn't put a readme in the zip-file, but it's very easy to
reproduce the leak with my testcase, just take a jboss messaging configuration
based on ejb3. Adjust the settings at the beginning of my build.xml to your
needs and run ant (without parameters). The leak will be obvious with
Hi Martin-
Sorry for taking so long to look at this, it's been a very busy few weeks.
I have run your test you sent to me using a fresh copy of JBoss 4.0.4GA, JDK5,
and EJB3-RC8-FD.
I installed EJB3 fresh as per the EJB3 install instructions into the all
configuration of the app server, and ra
We're also experiencing this leak (or at-least something related, and something
we didn't see before upgrading from JBossMQ), but for us it takes a long time
to become a real problem.
We run 4.0.4.GA with 1.0.1.CR2 on Sun JDK 1.5.0_07, with about 5 m/s. With
1.5GB max heap it takes about 4 - 6
Ok, I am going to run Martin's test program on a different machine and let it
run for days to see what happens.
bos - if you can reproduce your issue with a small test program that would be
great too - do you only see your problem when you use selectors?
View the original post :
http://www.jbo
Tim,
the memory leak becomes visible very quickly with my testcase, it will eat 1-2
MB every minute.
BUT: I didn't try your configuration yet, I can reproduce this reliably on
jboss 4.0.4.GA-patch1 with ejb3 config and jboss-messaging 1.0.1CR2. I will try
with EJB3-RC8 this afternoon and keep y
Sorry, have not been able to test without selectors yet. I'll try out some
different senarios as soon as I can, and produce some test code. But
unfortunately I won't be able to do anything before friday at the earliest.
View the original post :
http://www.jboss.com/index.html?module=bb&op=view
"bos" wrote : Sorry, have not been able to test without selectors yet. I'll try
out some different senarios as soon as I can, and produce some test code. But
unfortunately I won't be able to do anything before friday at the earliest.
Np, thanks for your help anyway.
View the original post :
h
"mwelss" wrote : Tim,
|
| the memory leak becomes visible very quickly with my testcase, it will eat
1-2 MB every minute.
| BUT: I didn't try your configuration yet, I can reproduce this reliably on
jboss 4.0.4.GA-patch1 with ejb3 config and jboss-messaging 1.0.1CR2. I will try
with EJB3-
Ok, a few things to report.
The call to rt.gc() in ClientServlet really slowed things down, so I removed it.
I then noticed that memory became exhausted (500MB) in a few minutes.
I also managed to replicate this using a simple self contained jms program
(using no EJBs) which simply sends messag
So, with the rt.gc, the system would also have run out of memory eventually,
but I guess it was slowing things down so much on my system that I didn't run
it long enough to notice
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3952334#3952334
Reply to the pos
anonymous wrote : Martin - which version of EJB3 are you running?
the one that comes with jboss 4.0.4.GA-patch1. Rumors say it could be RC6...
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3952345#3952345
Reply to the post :
http://www.jboss.com/index.html?
anonymous wrote : in PlexMDB, when the message is received it sends 4 more
messages to the topic:
Yes, that is all on purpose: this is where the selectors enter the game!
If you look at the client, it creates consumers (using the selctor) for targets
10 or 20
But the MDB sends one message each
No, that's not quite correct.
You are creating *two* consumers, each with the same selector expression
""target='10' OR target='20'"
For each message sent to the queue you are posting 4 messages, with target =10,
20, 30, 40
Therefore *both* of the two consumers will *each* receive the first tw
Yes, you are absolutely right! Shame on me.
The bad news is, that it seems like I did not 'extract' the testcase
sucessfully or it takes longer for the leak to appear...
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3952371#3952371
Reply to the post :
htt
Martin - I have tried out your latest test case, and I confirm that I observe a
memory leak.
After a lot of grief I have tracked it down to a race condition in message
acknowledgement that only occurs when used with EJB3.
I won't bore you with the details, but I am running your test case again
Martin-I ran your test program with 50 messages - it took a few hours and
after my fix I didn't notice any memory leak.
This fix will be in the 1.0.1GA release which should be out fairly soon once
we've cleared up this latest few issues we are having.
View the original post :
http://www.jb
Sounds great. Thanks!
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3952584#3952584
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3952584
All the advantages of Linux Managed Hosting--Without the Cost and Risk!
Fully tr
39 matches
Mail list logo
