I posted questions regarding a performance issue we encountered when this
upgrade was performed:
http://www.jboss.com/index.html?module=bb&op=viewtopic&t=74692
The problem turned out to be the garbage collector and our heap space settings.
This upgrade resulted in an increased use of permanent
Since JIRA entry has been closed I'm posting back here with update. I have
profiled this issue and see the following with respect to garbage collection:
As the application is exercised heavily, heap utilization increases and
decreases due to GC activity. The heap space used is generally around
Scott,
The config change suggested did not solve this issue. Those waiting threads
are likely a side affect and not cause. Note that this issue (accumulation of
waiting threads, leading to full degradation of performance) can be duplicated
by running our link checker against all 12 web sites
Does Jboss provide a way to set this attribute in
4.0.3?
JMX throws the error described above when attempting to apply changes to this
Mbean. This boolean value does not appear to be managed in a config file. The
java class "HandlerRequest" sets this attribute to true, right in the code.
Here's the link to Jira issue:
http://jira.jboss.com/jira/browse/JBAS-1169
Is there a jk.properties file deployed with jboss4.0.3? I believe this is the
right place to make the suggested configuration. I'm not finding this file in
our deployment.
Allen
View the original post :
http://www.j
I can see this property in JMX, although when I submit the change I get the
following error:
Cannot find setter method setModelerType
I see a JIRA entry for this issue with no resolution at this point. It appears
this will be corrected in 4.0.4.
I grepped through config for this property with
After looking at the javax.management code, it appears that the thread
TP-Processor137 is constructing an Mbean. Is there any other reason these
methods would be called? Trying to get some traction on this
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&
When deploying an exploded war the following error is thrown at startup:
***
16:01:29,682 ERROR [main ] [web.localhost.Engine] StandardContext[]:
Resources start failed:
java.lang.IllegalArgumentException: Document base
C:\usr\local\jboss\server\default\.\usr\local\idexx\deploy
I get the following message in my log at deploy time when I use the
jboss-web.xml config below, but it still works correctly:
[Servlet] Registration is not done -> stop
Here's the jboss-web.xml config:
/
host1
host2
Can someone elaborate on why this message is posted
Have many web sites sharing common J2EE services. These are packaged in same
WAR, partitioned into folders as follows:
en_US
nl_NL
etc...
With the context for this WAR set to "mysites", can I set up virtual hosting to
map various domains to the corresponding sites?
For example:
www.mysites.c
Numerous posts about this. Even a bug report regarding memory leaks, fixed in 3.2.4.
I have upgrade from 3.2.2 to 3.2.5 Using the same config as davidmboon. CMP1.1 EJBs
with commit option C. Jboss continues to report "Unable to passivate due to ctx
lock". I have found little help in the do
Working on implementing commit option c for a few CMP1.1 entity beans shared by two
applications. These beans will be managed in two containers. I have copied and
modified, with commit option c, the "Standard CMP EntityBean" config and placed it in
jboss.xml. Now when accessing these beans t
Alexey,
Thanks for your reply. I must have misinterpreted your earlier post. To bad about
the 1.0 EJBs but more reason for us to loop back and clean these up.
View the original post :
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3840442#3840442
Reply to the post :
http://www.j
Regarding older EJBs using 1.0 deployment descriptors but deployed into 3.2.3
container, can the container manage optimistic locking for these?
Does documentation exist re: the config for Optimistic using the 1.0 EJBs?
Regarding last post, this seems to contradict what is printed in the Admin/de
We have two applications running on the same Jboss instance (3.2.2). We will be
splitting these off on to separate Jboss instances and physical servers soon. In
general, these applications are stand-alone. However, there are a few shared EJBs
such as "member". After the split both applicati
Is it possible to reset this variable in JMX? We did not set this explicitly at
start up so the default gethost() value set the codebase host. We're looking for a
way to rest this with out restarting the app server. Were running version 3.2.4
Thanks
View the original post :
http://www.j
We are attempting to monitor response times for some transactions in our application.
This monitoring is an attempt to capture the information necessary to identify
performance baselines and provide specifics for a service level agreement that
includes expected response times for some user tran
These topics often don't have an outcome so wanted to take the time to briefly
describe my findings. Until rencently we were on Tomcat 3.2.1. Apparently there was
a bug in the read() method regarding it's return value. There was some code in our
application to deal specifically with that bug.
Will the new Web Console features be available for deployment on Jboss 3.2.3? or
Can the latest web console package be deployed on 3.2.3? I'm interested in using the
"monitoring" and "snapshot" features described for the web console in release 3.2.4
Thanks
http://www.jboss.org/index.html?
An update on this: First, thanks for the feedback to date. I have associated a
specific incident with the increase in CPU usage by Jboss. Some of our HTTP requests
contain a file that is read into memory and then saved on the server's file system.
In some cases, it seems this file data is co
On occasion, rarely actually, the call to InputStream.read(ÃÂ) hangs and never
finishes when reading from the HTTPServletRequest on the server side. This is rare
and seems to be due to one client that may be sending something in the request that is
causing this to hang. Regardless of the cause
Need some additional help. I have combed through thread dumps from a time period when
the Jboss was using %80 of CPU and then some after I restarted, right after and a few
days after. I found one peculiar thread. Before describing further, there's an issue
one user is having that seems to hap
Thanks a good suggestion.
I have a thread dump using kill -QUIT. Is there any information on what I should be
seeing for Jboss? I have looked through this but don' t have experience with
intrepreting the data. Any suggestions for reading etc. to educate myself? Other
tips?
http://www.jboss
Thanks Bill. This is a production system, mission critical to the business. Also,
this occurs infrequently and have not been able to duplicate in a test environment.
Will a profiler affect performance? That's my concern with this. I have no
experience, though. If not, I'll give this a try.
We are experiencing a periodic issue where Jboss (3.2.3) CPU usage jumps significantly
(80 - 90%) in a short period of time. When this is encountered we have to restart to
clear things up. I have run some OS (Solaris) commands (ps ...) to get information
on the threads when this happens. I h
25 matches
Mail list logo
