This is the one peeve I have, and maybe I'm way off, but j2ee has a gaping
hole here without having a safe singleton like mechanism that's container
neutral.
The assumption seems to be that things like this are services, and services
are container specific, like the various jdbc drivers etc..
Mbe
server of
choice.
- Original Message -
From: "Guy Rouillier" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, March 03, 2003 5:32 PM
Subject: Re: [JBoss-user] Feature idea
>
> - Original Message -
> From: "Rick LaBanca" <[EMAI
Jetty is separate but the jboss folks seem to want to treat it as the
official jsp portion of the suite (I agree). I think of jboss as the
controller over all its components so specifying some kind of wait page or
other logic should be at this top level somehow I think.
- Original Message
Sounds like a case for some kind of statement pooling if there is such a
thing. I haven't used j2ee for any of this stuff at this level, but my jdbc
experience has shown me that reusing prepared statements can be the biggest
optimization in many cases (with oracle it was especially so).
rick
--
That's a good idea, think I will go that way. I still hope for more external
settings (like separating the standard xml files from say an override file
to help between releases).
- Original Message -
From: "Jason Westra" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, November
I thought along the same lines as you. I'm not doing anything myself,
whatever the jetty cmp is doing triggers this, and the site isn't running
yet, this is initialization. I am pretty sure it's using the postgres map,
but it sure seems likely that I have something wrong in this area if it
works fo
gt;
> >>Cheers,
> >>
> >>Jules
> >>
> >>
> >>Sacha Labourey wrote:
> >>
> >>>OK, now that we know what is your environment, can you explain a
> >>>
> >>little bit
> >>
> >>>more your
sion clustering synch?
>
> OS
> JVM
> JBoss version
> What have you done to enable httpsession clustering
> How are you load-balancing/failing-over between the http nodes
> Are you testing using servlets or JSPs
> etc
>
> Jules
>
> Rick LaBanca wrote:
>
Sorry, thought the clustering was abstracted, using jetty. I'm just at the
point where I'm not sure of the intended operation. As for more detail, I'm
not sure what would be useful. I have it rigged so I can switch between two
servers, the only two together. Going to server B won't get me the sessi
I did think of the backplane idea as well. But we also have memory
consumption to think about. If I had four servers evenly balanced, say 100
users on each, I think I would end up with 400 sessions per server. Our
sessions are currently pigs, and that is motivation to fix that!
Perhaps better woul
I am beginning to get httpclustering to work. Aside from the overhead of
each server having to have the capacity to hold the total sessions in all
servers, what's the wire traffic like? Yea, I'll sniff it if I have to, this
is just to see if anyone has a quick idea.
I'm guessing it sends out X byt
I had this same issue. I Think it may be correct
that the ejb doesn't have access to this absolute path you are trying, but not
positive.
Anyway, the utility class generated by xdoclet
makes two string constants, one of them is the short name you used succesfully.
My xdoclet was using t
s on one transaction can at
> best
> > reduce latency/request while doing nothing for throughput.
> >
> > I think this was one of the simplifications adopted by ejbs to make
> > distributed computing comprehensible...so far I still think its
> reasonable.
> >
> > Still, I&
iple threads on one transaction can at best
> reduce latency/request while doing nothing for throughput.
>
> I think this was one of the simplifications adopted by ejbs to make
> distributed computing comprehensible...so far I still think its
reasonable.
>
> Still, I'd like to se
ansaction
> may be difficult to predict, and I think it is appropriate to internally
> prevent more than one thread working within a transaction as well.
>
> thanks
> david jencks
>
> On 2002.09.22 09:30:42 -0400 Rick LaBanca wrote:
> > I think a session bean doing threads i
I think a session bean doing threads is not encouraged, but accessing
something else that happens to be threaded could be ok. In my testing it is
ok in jboss right now anyway, I do spawn off threads for multiple requests.
I really don't understand the restriction anyway. So long as your response
They are jboss specific but j2ee will have them in some way eventually.
Jboss does a lot of cool things, but you have to make sure they aren't
specific!
What I did to replace an rmi server for now is have session beans access a
singleton. It seems the safest way to set up a service right now and
s all required bug-fixes and more, and check it into the
> release.
>
> If all development took place in the same tree, what you ask would be
> reasonable - however
>
>
> Jules
>
>
>
> Rick LaBanca wrote:
> > Wondering what others think about this issue (m
Title: RE: [JBoss-user] JMX question
Wondering what others think about this issue (may
be minor to most of you!)
I use 3.0.0. If I want to run 3.0.1 or 3.0.2, I
can't do it just by plopping in my wars/config. This is because some of the
settingshave changed (jetty packaged differently, jett
what about java web start? seems simpler to me (I used it, pretty easy, but
can't deliver different versions to different users afaik)
rick
- Original Message -
From: Greg Turner
To: [EMAIL PROTECTED]
Sent: Friday, August 16, 2002 9:08 PM
Subject: Re: [JBoss-user] Launching Swing Applica
In my situation, yes multiple threads. What I have is a service that I've
bastardized by making a stateless session bean that uses a singleton. When I
connect to it, i end up with many of these beans all calling asynchronously,
so yep, make it thread safe! I don't really know if they would end up
21 matches
Mail list logo
