You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@avalon.apache.org by Carsten Ziegeler <cz...@s-und-n.de> on 2003/11/06 13:30:31 UTC
[Vote] Removing ComponentManagerProxy
Volker found recently a hugh performance problem in the
current DefaultComponentFactory in Excalibur.
You can find more information here:
http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
I propose to remove the ComponentManagerProxy as this is
a) a performance killer and b) not soo useful as it might
look.
In some situations the ComponentManagerProxy is able to
release components that the developer has forgotten to
release - but only in some cases. As the general rule
when using excalibur is, to always release components,
removing the ComponentManagerProxy doesn't do any
harm, but provides greater performance.
Please cast your votes:
Carsten
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
Re: [Vote] Removing ComponentManagerProxy
Posted by Vadim Gritsenko <va...@verizon.net>.
Carsten Ziegeler wrote:
>Volker found recently a hugh performance problem in the
>current DefaultComponentFactory in Excalibur.
>
>You can find more information here:
>http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
>http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
>I propose to remove the ComponentManagerProxy as this is
>a) a performance killer and b) not soo useful as it might
>look.
>
>In some situations the ComponentManagerProxy is able to
>release components that the developer has forgotten to
>release - but only in some cases. As the general rule
>when using excalibur is, to always release components,
>removing the ComponentManagerProxy doesn't do any
>harm, but provides greater performance.
>
>Please cast your votes:
>
>
<kidding> What is ComponentManagerProxy? </kidding>
+1
Vadim
Re: [Vote] Removing ComponentManagerProxy
Posted by Gianugo Rabellino <gi...@apache.org>.
Carsten Ziegeler wrote:
> Volker found recently a hugh performance problem in the
> current DefaultComponentFactory in Excalibur.
>
> You can find more information here:
> http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
> http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
>
> In some situations the ComponentManagerProxy is able to
> release components that the developer has forgotten to
> release - but only in some cases. As the general rule
> when using excalibur is, to always release components,
> removing the ComponentManagerProxy doesn't do any
> harm, but provides greater performance.
>
> Please cast your votes:
Makes sense. +1.
Ciao,
--
Gianugo Rabellino
Pro-netics s.r.l. - http://www.pro-netics.com
Orixo, the XML business alliance - http://www.orixo.com
(Now blogging at: http://blogs.cocoondev.org/gianugo/)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
Re: [Vote] Removing ComponentManagerProxy
Posted by Torsten Curdt <tc...@vafer.org>.
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
+1
--
Torsten
Re: [Vote] Removing ComponentManagerProxy
Posted by Giacomo Pati <gi...@apache.org>.
Carsten Ziegeler wrote:
> Volker found recently a hugh performance problem in the
> current DefaultComponentFactory in Excalibur.
>
> You can find more information here:
> http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
> http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
>
> In some situations the ComponentManagerProxy is able to
> release components that the developer has forgotten to
> release - but only in some cases. As the general rule
> when using excalibur is, to always release components,
> removing the ComponentManagerProxy doesn't do any
> harm, but provides greater performance.
>
> Please cast your votes:
+1
--
Giacomo Pati
Otego AG, Switzerland - http://www.otego.com
Orixo, the XML business alliance - http://www.orixo.com
RE: [Vote] Removing ComponentManagerProxy
Posted by Carsten Ziegeler <cz...@s-und-n.de>.
Joerg Heinicke wrote:
>
> What happens with non-released components? Garbage Collection?
>
No, with non-released components you might get a memory-leak.
Therefore it is required to release a component fetch via a lookup
at some point.
Carsten
Re: [Vote] Removing ComponentManagerProxy
Posted by Joerg Heinicke <jh...@virbus.de>.
Carsten Ziegeler wrote:
> Volker found recently a hugh performance problem in the
> current DefaultComponentFactory in Excalibur.
>
> You can find more information here:
> http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
> http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
>
> In some situations the ComponentManagerProxy is able to
> release components that the developer has forgotten to
> release - but only in some cases. As the general rule
> when using excalibur is, to always release components,
> removing the ComponentManagerProxy doesn't do any
> harm, but provides greater performance.
>
> Please cast your votes:
>
> Carsten
+1
What happens with non-released components? Garbage Collection?
Joerg
--
System Development
VIRBUS AG
Fon +49(0)341-979-7419
Fax +49(0)341-979-7409
joerg.heinicke@virbus.de
www.virbus.de
Re: [Vote] Removing ComponentManagerProxy
Posted by Gianugo Rabellino <gi...@apache.org>.
Carsten Ziegeler wrote:
> Volker found recently a hugh performance problem in the
> current DefaultComponentFactory in Excalibur.
>
> You can find more information here:
> http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
> http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
>
> In some situations the ComponentManagerProxy is able to
> release components that the developer has forgotten to
> release - but only in some cases. As the general rule
> when using excalibur is, to always release components,
> removing the ComponentManagerProxy doesn't do any
> harm, but provides greater performance.
>
> Please cast your votes:
Makes sense. +1.
Ciao,
--
Gianugo Rabellino
Pro-netics s.r.l. - http://www.pro-netics.com
Orixo, the XML business alliance - http://www.orixo.com
(Now blogging at: http://blogs.cocoondev.org/gianugo/)
Re: [Vote] Removing ComponentManagerProxy
Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Carsten Ziegeler dijo:
> Volker found recently a hugh performance problem in the
> current DefaultComponentFactory in Excalibur.
>
> You can find more information here:
> http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
> http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
>
> In some situations the ComponentManagerProxy is able to
> release components that the developer has forgotten to
> release - but only in some cases. As the general rule
> when using excalibur is, to always release components,
> removing the ComponentManagerProxy doesn't do any
> harm, but provides greater performance.
>
> Please cast your votes:
+1. I am afraid of future memory leaks. But this will help us hunting for
bugs :-D
Best Regards,
Antonio Gallardo
Re: [Vote] Removing ComponentManagerProxy
Posted by Leo Simons <le...@apache.org>.
Carsten Ziegeler wrote:
> Ok, we have many +1 from cocoon committers, we have only
> one +1 from an avalon committer (myself :( ) and no other
> vote from other avalon committers.
I'm not voting, because I think I am not fully aware of all the ins and
outs of this issue; I'm behind on mail, I've yet to read Volkers post.
> So, this is situation is a little bit difficult.
>
> But as we don't have a -1, "lazy consensus" applies.
indeed; which is why there's no problem. After all, we opened up parts
of the avalon CVS to the cocoon committers precisely to facilitate this
kind of thing.
cheers,
- Leo
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
RE: [Vote] Removing ComponentManagerProxy
Posted by Carsten Ziegeler <cz...@s-und-n.de>.
A release would be great!
We still have the outstanding issue of releasing some components
from excalibur, like source resolver, xml etc. Last time I made
the attempt I failed - and currenlty the built system is
absolutely unusable :(
Carsten
> -----Original Message-----
> From: news [mailto:news@sea.gmane.org]On Behalf Of Leo Simons
> Sent: Saturday, November 08, 2003 1:27 AM
> To: dev@avalon.apache.org
> Subject: Re: [Vote] Removing ComponentManagerProxy
>
>
> Carsten Ziegeler wrote:
> > I removed the code some minutes ago - and we are already testing.
> > As we have a Cocoon release scheduled for next week, we hopefully
> > find all problems - if there are some.
>
> will you need an ECM release as well? Other avalon-hosted stuff that
> needs a release or update? (now could be as good a time as any to push
> for it :D)
>
> - LSD
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
> For additional commands, e-mail: dev-help@avalon.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
Re: [Vote] Removing ComponentManagerProxy
Posted by Leo Simons <le...@apache.org>.
Carsten Ziegeler wrote:
> I removed the code some minutes ago - and we are already testing.
> As we have a Cocoon release scheduled for next week, we hopefully
> find all problems - if there are some.
will you need an ECM release as well? Other avalon-hosted stuff that
needs a release or update? (now could be as good a time as any to push
for it :D)
- LSD
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
RE: [Vote] Removing ComponentManagerProxy
Posted by Carsten Ziegeler <cz...@s-und-n.de>.
Berin Loritsch wrote:
>
> Carsten Ziegeler wrote:
> > Ok, we have many +1 from cocoon committers, we have only
> > one +1 from an avalon committer (myself :( ) and no other
> > vote from other avalon committers.
> >
> > So, this is situation is a little bit difficult.
> >
> > But as we don't have a -1, "lazy consensus" applies.
> > So, if noone complains in the next hours I will remove
> > the proxy code.
>
>
> I think the Cocoon developers are the biggest remaining users
> of the code, so I think we trust them. It definitely is presenting
> more issues than it is solving, so +1 from me.
>
> Please test thoroughly to make sure that things are happening as
> expected.
>
I removed the code some minutes ago - and we are already testing.
As we have a Cocoon release scheduled for next week, we hopefully
find all problems - if there are some.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
Re: [Vote] Removing ComponentManagerProxy
Posted by Berin Loritsch <bl...@apache.org>.
Carsten Ziegeler wrote:
> Ok, we have many +1 from cocoon committers, we have only
> one +1 from an avalon committer (myself :( ) and no other
> vote from other avalon committers.
>
> So, this is situation is a little bit difficult.
>
> But as we don't have a -1, "lazy consensus" applies.
> So, if noone complains in the next hours I will remove
> the proxy code.
I think the Cocoon developers are the biggest remaining users
of the code, so I think we trust them. It definitely is presenting
more issues than it is solving, so +1 from me.
Please test thoroughly to make sure that things are happening as
expected.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
RE: [Vote] Removing ComponentManagerProxy
Posted by Carsten Ziegeler <cz...@s-und-n.de>.
peter royal wrote:
> On Nov 7, 2003, at 3:31 AM, Carsten Ziegeler wrote:
> > Ok, we have many +1 from cocoon committers, we have only
> > one +1 from an avalon committer (myself :( ) and no other
> > vote from other avalon committers.
> >
> > So, this is situation is a little bit difficult.
> >
> > But as we don't have a -1, "lazy consensus" applies.
> > So, if noone complains in the next hours I will remove
> > the proxy code.
>
> +1, whack it!
Done :)
Carsten
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
Re: [Vote] Removing ComponentManagerProxy
Posted by peter royal <pr...@apache.org>.
On Nov 7, 2003, at 3:31 AM, Carsten Ziegeler wrote:
> Ok, we have many +1 from cocoon committers, we have only
> one +1 from an avalon committer (myself :( ) and no other
> vote from other avalon committers.
>
> So, this is situation is a little bit difficult.
>
> But as we don't have a -1, "lazy consensus" applies.
> So, if noone complains in the next hours I will remove
> the proxy code.
+1, whack it!
-pete
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
Re: [Vote] Removing ComponentManagerProxy
Posted by Niclas Hedhman <ni...@hedhman.org>.
On Friday 07 November 2003 16:31, Carsten Ziegeler wrote:
> Ok, we have many +1 from cocoon committers, we have only
> one +1 from an avalon committer (myself :( ) and no other
> vote from other avalon committers.
>
> So, this is situation is a little bit difficult.
>
> But as we don't have a -1, "lazy consensus" applies.
> So, if noone complains in the next hours I will remove
> the proxy code.
I think the situation is basically, "Cocoon community knows what's best for
ECM."
Niclas
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org
RE: [Vote] Removing ComponentManagerProxy
Posted by Carsten Ziegeler <cz...@s-und-n.de>.
Ok, we have many +1 from cocoon committers, we have only
one +1 from an avalon committer (myself :( ) and no other
vote from other avalon committers.
So, this is situation is a little bit difficult.
But as we don't have a -1, "lazy consensus" applies.
So, if noone complains in the next hours I will remove
the proxy code.
Carsten
Carsten Ziegeler wrote:
>
> Volker found recently a hugh performance problem in the
> current DefaultComponentFactory in Excalibur.
>
> You can find more information here:
> http://marc.theaimsgroup.com/?t=106779872700001&r=1&w=2
>
> http://marc.theaimsgroup.com/?t=106802136600002&r=1&w=2
>
> I propose to remove the ComponentManagerProxy as this is
> a) a performance killer and b) not soo useful as it might
> look.
>
> In some situations the ComponentManagerProxy is able to
> release components that the developer has forgotten to
> release - but only in some cases. As the general rule
> when using excalibur is, to always release components,
> removing the ComponentManagerProxy doesn't do any
> harm, but provides greater performance.
>
> Please cast your votes:
>
> Carsten
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
> For additional commands, e-mail: dev-help@avalon.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org