Hi
Does anyone know if the max_agents/agent_limit in
estate_settings.xml/regionsettings table has been implemented? If not, is
there any other way of controlling the number of logins?
Thanks
Geetika Sharma
=-=-=
Notice: The information contained in this e-mail
message and/o
I also think that just because a texture comes off of HTTP doesn't mean
security can't be applied.
I can see scenarios to duplicate textures across multiple web servers to
improve downloading performance.
I kind of look forward to the day when I can apply URLs as textures and
manage all my
is issue..
Salahzar
PS: Also I noticed I have to restart all the regions almost once a day
otherwise the console becomes unresponsive and the regions appear not
working :(
-- next part --
An HTML attachment was scrubbed
Lake, Dan wrote:
> Justin,
> I filed a Mantis entry for the timer_dictionary.patch.
>
> http://opensimulator.org/mantis/view.php?id=3311
>
Thanks very much Dan, committed as r8812.
Regarding the localID SceneGraph dictionary, as you've seen in the comments
we've pretty much known that the l
Like I mentioned before, I spent some time adapting the SVN trunk of
libomv to OpenSim, so I think I can look into this release, if there is
really an interest to update it.
I will demand my ice cream after I'm done though
On Wed, 2009-03-18 at 19:55 +, Justin Clark-Casey wrote:
> Teravus Ova
+1
Patches welcome.
Melanie
Tommi Laukkanen wrote:
> Ouch I forgot that evil madness. It would be good to put in proper
> caches which drop items based on time. Eternal caches are well..
> antipattern. Namely intentionally build memory leak. Anti patterns are
> well known arch demons of software
Teravus Ovares wrote:
> Looks like LibOMV is just about to prepare a release. This would be
> their first release in quite a while. I'd be willing to put in
> some time to update it after their release.(though, it'll probably
> be two weeks from now that I have the time).
>
> Thoughts?
Chris Hart wrote:
> Perhaps start with baby steps - how about just a simple move to change
> opensim.ini.example and opensim.ini default locations out of the bin and
> put it in a config folder - I for one am tired of looking for config
> files in among all the rest. Can we do the same with the *.c
Tommi Laukkanen wrote:
> I apologise that I linked the latter chapter to this specific thread.
> I should have posted it as a separate message as it was a more general
> comment on our decission making process.
>
> I guess it sprung from my wonderment about how the decission turned
> out to be not
Looks like LibOMV is just about to prepare a release. This would be
their first release in quite a while. I'd be willing to put in
some time to update it after their release.(though, it'll probably
be two weeks from now that I have the time).
Thoughts?
Sincerely
Teravus
On 3/16/09, Ar
My professional opinion is that a protocol that serves the user needs does
just that. Asset security is a user need and new, experimental protocols
such as MXP which are trying to establish a user base have an opportunity to
make an advancement it the management of assets and such a technological
a
+1 from me too - I think there may well be uses of OpenSim where one may want
to do the opposite of securing assets and
actually make them as accessible as possible.
Configurability is key.
Stefan Andersson wrote:
> General +1's, cheers and yays.
>
> Best regards,
> Stefan Andersson
> Tribal
Justin,
I filed a Mantis entry for the timer_dictionary.patch.
http://opensimulator.org/mantis/view.php?id=3311
~Dan
-Original Message-
From: opensim-dev-boun...@lists.berlios.de
[mailto:opensim-dev-boun...@lists.berlios.de] On Behalf Of Justin Clark-Casey
Sent: Wednesday, March 18,
Hi Salahzar,
Also sprach Salahzar Stenvaag (salah...@gmail.com):
> PS: Also I noticed I have to restart all the regions almost once a day
> otherwise the console becomes unresponsive and the regions appear not
> working :(
I have that too usually, the grid becomes unresponsive, and when
pressing
There are memory leaks in OpenSim, so restarting every so often isn't an
option. And, you are into swap a little bit there. It seems that recent mono
likes some headroom in real memory and tends to crash when the system hits
virtual memory. You might see better results with mono 2.0.1 in that
situa
Perhaps start with baby steps - how about just a simple move to change
opensim.ini.example and opensim.ini default locations out of the bin and
put it in a config folder - I for one am tired of looking for config
files in among all the rest. Can we do the same with the *.config files
for the other
My professional opinion is that it is a wild goose chase to try to
protect assets. The unfortunate reality is that when opensimulator
becomes widely adopted there will be open clients and anyone can go
and use them to get someone elses assets. There is just no way to
protect yourself against it. Wo
Hello OSWorld. Just a question about current limits in OS. We are running 4
instances of OpenSIm rev 8810, holding 11 sims on ovh rps-2 (2GB RAM)
connected to an external grid. Top clearly indicates that free memory is
still available
Tasks: 109 total, 1 running, 108 sleeping, 0 stopped, 0 zombie
I apologise that I linked the latter chapter to this specific thread.
I should have posted it as a separate message as it was a more general
comment on our decission making process.
I guess it sprung from my wonderment about how the decission turned
out to be not to fix the overtly long config fil
General +1's, cheers and yays.
Best regards,
Stefan Andersson
Tribal Media AB
Date: Wed, 18 Mar 2009 11:38:41 -0700
From: d...@metaverseink.com
To: opensim-dev@lists.berlios.de
Subject: Re: [Opensim-dev] Accessing textures via HTTP [bayes]
The road to security is like an obstacle course i
The road to security is like an obstacle course in the Olympics. The
obstacles range from really easy (what's happening out there with
OpenID, and the current opensim access to inventory) to really hard
(have the server disconnected from the public network and require the
physical presence of u
I was under the impression that the asset server was going to require
authentication regardless. It would be possible to keep track, for
each agent, what textures have been 'authorized' (by being in view).
Remember, we can't protect content in the cache of the client, and the
client can be hostile
The current j2k assets transferred over via the xfer protocol provides a
rudimentary protection for texture assets. Opening the doors to jpeg
textures via http effectively makes all textures available for download by
any web browser. This is significantly reduced asset security and may
violate lice
If I may interfeere...
An asset is, at first, a binary blob. So, if you uploaded a jpeg or png, that
should be served fine. In that case, we might need a jpeg-or-png-to-j2000
conversion to serve the asset to an SL viewer. Which we kind of have in the
spanners at any rate.
The accept-hea
I didn't see anyone respond 'politically'. As far as I could tell,
everyone gave their opinion on it, sincerely. Can you describe what
you mean by "We are not building political machine to steer opensim
architecture are we?" this a bit more and demonstrate where you
thought it occurred, Tommi?
-
On the server-side, this can be made configurable, default to false. If
server admins want to serve jpgs, let's let them.
Since this work is being done within the Idealist viewer, a separate
issue is whether you want Idealist to assume the existence of jpgs on
the server or not. But that's a cli
It is a bit sad that this effort goes to waste. The opensim.ini is too
big to be efficient and everyone is wasting a lot of time browsing it
back and forth. I think something like splitting it up or moving
defaults to other file is really needed.
What is complex now is finding the properties you n
I would not go so far that I would suggest we should go and try to
protect our assets by using semi supported image format instead of
well supported format...
-Tommi
___
Opensim-dev mailing list
Opensim-dev@lists.berlios.de
https://lists.berlios.de/mailm
Ouch I forgot that evil madness. It would be good to put in proper
caches which drop items based on time. Eternal caches are well..
antipattern. Namely intentionally build memory leak. Anti patterns are
well known arch demons of software design and we should try to purge
them off the code base ,)
It's really the proliferation of eternal caches that causes issues,
IMHO.
Melanie
Tommi Laukkanen wrote:
> One could catch the objects which are causing the leak by using a
> profiler. If someone has .NET profiler and skills he could try to
> trace the source of the problem so it can be fixed..
I don't think it's a goal of OpenSim to serve texture assets to web
browsers, nor should it be. Such a move would open doors to texture piracy
and reduce any incentives for content creators to distribute their content
on any OpenSim based platform. Grid operators who would want to serve their
text
One could catch the objects which are causing the leak by using a
profiler. If someone has .NET profiler and skills he could try to
trace the source of the problem so it can be fixed...
-Tommi
___
Opensim-dev mailing list
Opensim-dev@lists.berlios.de
htt
Justin,
Thanks for taking a look at the wiki. The timer patch should be complete, as
long as there is no need for multiple timers per (localID,objectID) pair. I
could enter it into Mantis if it looks ready to go. The other patch may be
similar to existing code which is either commented out or
I think j2k is not really supported by any main stream web browser
software nor SDK APIs of different languages. Requiring opejpeg native
lib to be included in all clients is not a good design decission. It
would be much cleaner if you can operate with the nativate image
manipulation API like Syste
Lake, Dan wrote:
> I work at Intel for Mic Bowman. He mentioned a while back that we are
> developing some workloads for understanding the scalability bottlenecks
> in OpenSim. I recently created a test which generates 2000 scripted
> objects which rotate and change color at random intervals. Th
One of the nagging problems we have appears to be memory leaks. There are
consistent and frustrating reports of more heavily used regions using up all
the RAM, then starting swap, then lagging, then freezing every week.
In order to help bring a little clarity to this, Nebadon has put together:
Wow Dan thanks so much I just tweeted it here
http://twitter.com/Dr_Manhattan thank you and Mic/Intel for this help.
Kyle
www.reactiongrid.com
From: opensim-dev-boun...@lists.berlios.de
[mailto:opensim-dev-boun...@lists.berlios.de] On Behalf Of Lake, Dan
Sent: Wednesday, March 18, 2009 8:37
I work at Intel for Mic Bowman. He mentioned a while back that we are
developing some workloads for understanding the scalability bottlenecks in
OpenSim. I recently created a test which generates 2000 scripted objects which
rotate and change color at random intervals. The workload consumed multi
38 matches
Mail list logo