Re: [Opensim-dev] Diva Groups and Offline IM
Really cool, thank you Diva! :) I wonder, per chance could this commit be ported over to 0.7.5-postfixes? On 02/16/2013 03:18 PM, André Verwijs wrote: NICE..!! thanks :) On Sat, Feb 16, 2013 at 9:07 PM, Diva Canto d...@metaverseink.com mailto:d...@metaverseink.com wrote: Heads up: after discussion within core devs, the Groups and Offline IM addons I did for D2 are going to be donated to core opensim. This means that core opensim will have support for these out of the box without having to install any further components (i.e. apache). They work both for standalones and grids. Diva Groups has the optional HG service for supporting groups with foreign users. It also implements notices with attachments, something that is missing from Flotsam groups. It does not implement the voting and accounting features of the viewer (patches welcome). For those using Flotsam or Simian Groups, they're all mutually exclusive -- the config decides which one to use. The code will be added within the next week. If there are any issues or questions, let us know. Cheers Diva ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de mailto:Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev -- Thank you for reading my email. I hope to receive a replay from you soon... _ *My Opensim/Second Life Blog * http://verwijs.wordpress.com http://verwijs.wordpress.com%20 *(Dutch, basic hardware/software help windows, Mac, Linux) * http://verwijs-pc.nl *My Twitter Page: * http://twitter.com/OpenSimFan http://twitter.com/OpenSimFan *My Facebook page (Be my friend, please ) * http://www.facebook.com/andre.verwijs http://www.facebook.com/andre.verwijs%20 ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Diva Groups and Offline IM
I don't have much of a user base as of yet, but I'll be glad to do whatever I can to help test. On 02/16/2013 06:12 PM, Diva Canto wrote: Potentially yes. But it's probably worth a new version number. Diva Groups works well within the limited testing I had time to do by myself; it needs a lot more testing, especially the grid configuration. On 2/16/2013 12:29 PM, John Sheridan wrote: Really cool, thank you Diva! :) I wonder, per chance could this commit be ported over to 0.7.5-postfixes? On 02/16/2013 03:18 PM, André Verwijs wrote: NICE..!! thanks :) On Sat, Feb 16, 2013 at 9:07 PM, Diva Canto d...@metaverseink.com mailto:d...@metaverseink.com wrote: Heads up: after discussion within core devs, the Groups and Offline IM addons I did for D2 are going to be donated to core opensim. This means that core opensim will have support for these out of the box without having to install any further components (i.e. apache). They work both for standalones and grids. Diva Groups has the optional HG service for supporting groups with foreign users. It also implements notices with attachments, something that is missing from Flotsam groups. It does not implement the voting and accounting features of the viewer (patches welcome). For those using Flotsam or Simian Groups, they're all mutually exclusive -- the config decides which one to use. The code will be added within the next week. If there are any issues or questions, let us know. Cheers Diva ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de mailto:Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev -- Thank you for reading my email. I hope to receive a replay from you soon... _ *My Opensim/Second Life Blog * http://verwijs.wordpress.com http://verwijs.wordpress.com%20 *(Dutch, basic hardware/software help windows, Mac, Linux) * http://verwijs-pc.nl *My Twitter Page: * http://twitter.com/OpenSimFan http://twitter.com/OpenSimFan *My Facebook page (Be my friend, please ) * http://www.facebook.com/andre.verwijs http://www.facebook.com/andre.verwijs%20 ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Diva Groups and Offline IM
I'm just theorizing here, but I would think that common components required for interoperability between grids would need to go into the core. Just recently I finished writing my own offline IM handler in c# only to realize it didn't save messages that came in from over the hypergrid. I've yet to dig in to see why, but after pouring over different examples on how to get profiles working I'm quickly realizing that there is a need for a common way for all of these services to communicate. Its one thing to have your own proprietary communications formats if you're running a closed grid, but in the case where your components need to talk to other people's - that seems to be a different story. For example, with my profiles module I wanted to create a set of data objects (psProfile, psProfilePick, psProfileClassifieds, etc.), serialize them on the sim end, then ship them off to the profiles server and vice versa. After seeing how other projects out there communicate with their server counterparts, it became blatantly apparent that my way would work just fine if it were used on a closed grid but not in a Hypergridded environment where it would need to talk to countless other profile servers that all likely communicate in different ways. I'm pretty sure the same would hold true for offline messaging, groups, and so on. On 02/16/2013 07:32 PM, R.Gunther wrote: Opensim core ? I hope it's going to be a module, and not in the core of opensim. I have the idea there's already to much in core and not enough in modules. On 2013-02-16 21:07, Diva Canto wrote: Heads up: after discussion within core devs, the Groups and Offline IM addons I did for D2 are going to be donated to core opensim. This means that core opensim will have support for these out of the box without having to install any further components (i.e. apache). They work both for standalones and grids. Diva Groups has the optional HG service for supporting groups with foreign users. It also implements notices with attachments, something that is missing from Flotsam groups. It does not implement the voting and accounting features of the viewer (patches welcome). For those using Flotsam or Simian Groups, they're all mutually exclusive -- the config decides which one to use. The code will be added within the next week. If there are any issues or questions, let us know. Cheers Diva ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Setting a single child prim to Phantom
I'm not sure if this works on OS yet as I've not tried it and I'm about to go to bed. But... If you're on viewer2+ try setting the physics type from prim to none in the build window. On SL that would set that part to no physics - or the equivalent of phantom. On 02/03/2013 02:16 AM, Oren Hurvitz wrote: I have recently become aware that there's an entire cottage industry of scripts whose purpose is to make a single prim in a linkset phantom. These scripts are needed because: a) Usually the phantom flag is set for an entire linkset, so it's necessary to use a script to change the flag for a single prim b) The LSL functions that change the Phantom flag (llSetStatus() and llSetPrimitiveParams()) change it for the entire linkset at once, so they can't be used to change a single prim. These scripts use a very complicated workaround that involve changing the prim's shape in several steps, which is brittle and dangerous. You can see the scripts here: http://wiki.secondlife.com/wiki/Phantom_Child In Second Life there is now a simpler and safer way to make a single prim stop colliding: llSetLinkPrimitiveParamsFast(LINK_THIS, [PRIM_PHYSICS_SHAPE_TYPE, PRIM_PHYSICS_SHAPE_NONE]); However, this isn't implemented in OpenSim. And I don't know if it's exactly equivalent to making the prim phantom. I propose extending llSetStatus() to allow setting the phantom flag for a single prim. This involves adding the flag OS_STATUS_PHANTOM_PRIM. It works just like the existing flag STATUS_PHANTOM, but it operates on the prim that contains the script instead of the entire linkset. This behavior mirrors the way llSetStatus() works with the STATUS_BLOCK_GRAB flag: it also comes in two flavors, STATUS_BLOCK_GRAB (for a prim) and STATUS_BLOCK_GRAB_OBJECT (for the linkset). Any thoughts? Does anyone know of a different way to safely change the phantom flag for a single prim in a linkset? Oren -- View this message in context: http://opensim-dev.2196679.n2.nabble.com/Setting-a-single-child-prim-to-Phantom-tp7578513.html Sent from the opensim-dev mailing list archive at Nabble.com. ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] www.opensim.net
Going by: *121*days until *OpenSim.net Grid Launch Party * at the upper left of the page I'd guess this the homepage (or future homepage) of someone's new grid. I think someone may need to read the OpenSim license - maybe the part where it states: * * Neither the name of the OpenSimulator Project nor the * names of its contributors may be used to endorse or promote products * derived from this software without specific prior written permission. - Orion Pseudo ** On 05/03/2010 10:59 AM, Fish Kungfu wrote: All the entries in the News and Journals section are posted by someone named, Ismail Malik. On May 3, 2010 9:49 AM, Tedd Hansen t...@konge.net mailto:t...@konge.net wrote: Hi Anyone know what www.opensim.net http://www.opensim.net is all about? The short story; It looks to me like someone (anonymous owner of opensim.net http://opensim.net) has hijacked a name (opensim) he or she may or may not have legal rights to with the (criminal) intent of abusing a (free) product name for his or her own benefit. (Yeah, darn those feminists, can’t “he” just cover both sexes?) I really hope this doesn’t mean that core developers have to spend time on legal issues because some 14 year old wants to run his/her own site. This is valuable time for professional developers that can be spent on making OpenSim better instead of bitching to lawyers. [09:57:25] Tedd1 http://www.opensim.net/videos/opensim-tutorial-6-basic-avatar [09:57:28] Tedd1 whats this? :) [09:58:05] Tedd1 or rather this; http://www.opensim.net/ [10:00:26] sachaMagne_ no idea [10:00:52] Tedd1 it seems to be ripping data directly from other sites [10:00:59] phrearch hey [10:01:00] Tedd1 and pretending to be opensim [10:01:25] Tedd1 view source - Public Grid List Sorted a href=http://opensimulator.org/wiki/Grid_List/Sorted/Alphabetically; [10:02:25] sachaMagne_ the opensimulator list is wrong anywat [10:02:46] sachaMagne_ if you select different sorts, you will find different result for the same grid [10:03:15] Tedd1 page contains RSS feeds, steals lists from opensimulator.org http://opensimulator.org, copy pastes info and mixes it with lies and half baked truth without ever mentioning that it is *NOT* opensim [10:03:28] Tedd1 *RSS feeds from other pages (of course) [10:03:36] Tedd1 so no actual contribution, a lot of stealing :) [10:04:09] sachaMagne_ not a breaking news for RSS aggregator ... :/ [10:07:22] Tedd1 someone should probably contact the domain owner [10:07:41] Tedd1 would be really pittyful if core members had to spend time on legal issues to get some 14 year old to play nice [10:07:46] sachaMagne_ i made a whois and the real owner is hidden [10:08:07] Tedd1 time that could be spent on features, bugfixes, etc [10:08:51] sachaMagne_ no legal information on the site itself Br, Tedd ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de mailto:Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Mono minimum version bump
Justin, I believe this may be one of the issues at hand: http://opensimulator.org/mantis/view.php?id=4087 . With mono 2.0.1 I had been receiving this same IL code error whenever an inventory item was moved between folders. After updating to mono 2.4.2 the issue resolved itself. Melanie wrote: The bug causes the Invalid IL code, the compiler generates an IL sequence the JITer can't emit as machine code. Melanie Justin Clark-Casey wrote: Melanie wrote: Hello, after discovering a bug in the 2.0.1 version of mono, which causes frequent crashing of regions, we have taken the step to bump the minimum required mono version to 2.4.2+. This has become necessary because the bug is rather difficult to isolate and therefore impossible to work around reliably. Please could you explain what this bug is? Thanks. ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Deprecate OpenSim.Grid.InventoryServer and OpenSim.Grid.AssetServer?
Not to butt in but, I kind of like the name B.U.S.T. It has sort of a maternal ring to it which is suiting to a core set of servers. But then again, I'm also the guy that came up with such weird acronyms as QUADRES for Quick Usable and Dirty Report Execution System. :P Thanks, :) - John Toni Alatalo wrote: On Jul 8, 2009, at 7:51 PM, Sean Hennessee wrote: MW wrote: I also would rather a different name than BUST, and also before any How about BOSS? Basic Open Simulator Servers? nice acronym - perhaps too JBoss-y a name though, and it being also a server framework (the open source j2ee thing) is a little bit close. i don't mind BUST, but it's not a huge matter i think anyways. ~Sean ~Toni ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] [Opensim-commits] r9924 - in trunk: OpenSim/Region/ClientStack/LindenUDP bin
I'm sitting here clearing up a few scrap prims to make one of my builds
a bit more efficient (changing from 15 prim factory windows to 5 for
example) when it occurred to me. I've noticed that while opensim is
chugging along, updating the sim database and sending stuff back to my
inventory, movement in-world starts to lag and studder a bit. Once its
done things understandably go back to normal. I wonder a change similar
to Dr. Scofield's texture handler only for the database / prim storage
routines could make even further of a difference? I'd imagine if so,
the initial rez-in for example when an avi is wearing 300 some-odd prims
in attachments would have less of an impact. Just a thought... :)
Thanks,
- John / Orion Pseudo
drscofi...@opensimulator.org wrote:
Author: drscofield
Date: 2009-06-25 00:42:06 -0700 (Thu, 25 Jun 2009)
New Revision: 9924
Modified:
trunk/OpenSim/Region/ClientStack/LindenUDP/J2KImage.cs
trunk/OpenSim/Region/ClientStack/LindenUDP/LLClientView.cs
trunk/OpenSim/Region/ClientStack/LindenUDP/LLImageManager.cs
trunk/bin/OpenSim.ini.example
Log:
From: Alan Webb alan_w...@us.ibm.com
This change moves texture send processing out of the main
packet processing loop and moves it to a timer based
processing cycle.
Texture packets are sent to the client consistently over
time. The timer is discontinued whenever there are no
textures to transmit.
The behavior of the texture sending mechanism is controlled
by three variables in the LLCLient section of the config
file:
[1] TextureRequestRate (mS) determines how many times per second
texture send processing will occur. The default is 100mS.
[2] TextureSendLimit determines how many different textures
will be considered on each cycle. Textures are selected
by priority. The old mechanism specified a value of 10 for
this parameter and this is the default
[3] TextureDataLimit determines how many packets will be sent for
each of the selected textures. The old mechanism specified a
value of 5, so this is the default.
So the net effect is that TextureSendLimit*TextureDataLimit
packets will be sent every TextureRequestRate mS.
Once we have gotten a reasonable feeling for how these parameters
affect overall processing, it would be nice to autonmically manage
these values using information about the current status of the
region and network.
Note that this also resolves the pathologcal problem that
previously existed which was that a seated avatar generated very
few in-bound packets (theoretically) and would therefore be the
least able to retrieve the images being displayed by a
projector script.
Modified: trunk/OpenSim/Region/ClientStack/LindenUDP/J2KImage.cs
===
--- trunk/OpenSim/Region/ClientStack/LindenUDP/J2KImage.cs2009-06-25
07:39:48 UTC (rev 9923)
+++ trunk/OpenSim/Region/ClientStack/LindenUDP/J2KImage.cs2009-06-25
07:42:06 UTC (rev 9924)
@@ -260,7 +260,7 @@
return false;
}
}
-public bool SendPackets(LLClientView client)
+public bool SendPackets(LLClientView client, int maxpack)
{
if (!m_completedSendAtCurrentDiscardLevel)
@@ -284,7 +284,7 @@
}
int count = 0;
-while (SendMore count 5 m_packetNumber =
m_stopPacket)
+while (SendMore count maxpack m_packetNumber =
m_stopPacket)
{
count++;
SendMore = SendPacket(client);
@@ -391,6 +391,10 @@
}
}
}
+else
+{
+m_packetNumber = m_stopPacket;
+}
}
}
}
Modified: trunk/OpenSim/Region/ClientStack/LindenUDP/LLClientView.cs
===
--- trunk/OpenSim/Region/ClientStack/LindenUDP/LLClientView.cs
2009-06-25 07:39:48 UTC (rev 9923)
+++ trunk/OpenSim/Region/ClientStack/LindenUDP/LLClientView.cs
2009-06-25 07:42:06 UTC (rev 9924)
@@ -80,6 +80,8 @@
private ListObjectUpdatePacket.ObjectDataBlock m_primFullUpdates =
new ListObjectUpdatePacket.ObjectDataBlock();
+private Timer m_textureRequestTimer;
+
private bool m_clientBlocked;
private int m_probesWithNoIngressPackets;
@@ -140,6 +142,10 @@
protected int m_primTerseUpdateRate = 10;
protected int m_primFullUpdateRate = 14;
+protected int m_textureRequestRate = 100;
+protected int m_textureSendLimit = 10;
+protected int m_textureDataLimit = 5;
+
protected int
Re: [Opensim-dev] OpenSim's Acronyms
Dilatn = dilation SimFPS = simulator frames per second PhysFPS = physics frames per second AgntUp = agent uptime RootAg = root agent ChldAg = child agent The following are beyond me. :) PrimsATv PrmAtvScr ScrLPS1 FrmMSA André Filipe wrote: Dears, I'm a newbie in OpenSim and I'd like to know what the following acronyms (in bold) stands for, anybody can help me? Dilatn SimFPS PhysFPS AgntUp RootAg ChldAg PrimsATv PrmAtvScr ScrLPS1 FrmMSA Regards, ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] [Opensim-users] Difficulty build SVN 9480
Actually, I'm getting this same problem with the 0.6.5-rc1 checkout with Mono 1.9.1 and 2.0. Yet I've seen other using this release. Is there some trick to getting this one to build other then the normal instructions from the wiki? Thanks, :) - John Brent Seidel wrote: I am able to build SVN 9479, but when I try to build SVN 9480, I get the following results from nant: -- NAnt 0.86 (Build 0.86.2898.0; beta1; 12/8/2007) Copyright (C) 2001-2007 Gerry Shaw http://nant.sourceforge.net Buildfile: file:///home/OpenSim/svn/opensim-0.6.4.9561/OpenSim.build Target framework: Mono 2.0 Profile Target(s) specified: build [echo] Using 'mono-2.0' Framework init: Debug: [echo] Platform unix build: [nant] /home/OpenSim/svn/opensim-0.6.4.9561/OpenSim/Framework/ Servers/HttpServer/OpenSim.Framework.Servers.HttpServer.dll.build build Buildfile: file:///home/OpenSim/svn/opensim-0.6.4.9561/ OpenSim/Framework/Servers/HttpServer/ OpenSim.Framework.Servers.HttpServer.dll.build Target framework: Mono 2.0 Profile Target(s) specified: build build: [echo] Build Directory is /home/OpenSim/svn/ opensim-0.6.4.9561/OpenSim/Framework/Servers/HttpServer/bin/Debug [csc] Compiling 24 files to '/home/OpenSim/svn/ opensim-0.6.4.9561/OpenSim/Framework/Servers/HttpServer/bin/Debug/ OpenSim.Framework.Servers.HttpServer.dll'. [csc] /home/OpenSim/svn/opensim-0.6.4.9561/OpenSim/ Framework/Servers/HttpServer/AsynchronousRestObjectRequester.cs (95,41): error CS0246: The type or namespace name `TResponse' could not be found. Are you missing a using directive or an assembly reference? [csc] /home/OpenSim/svn/opensim-0.6.4.9561/OpenSim/ Framework/Servers/HttpServer/AsynchronousRestObjectRequester.cs (115,21): error CS0029: Cannot implicitly convert type `TResponse' to `TResponse' [csc] /home/OpenSim/svn/opensim-0.6.4.9561/OpenSim/ Framework/Servers/HttpServer/AsynchronousRestObjectRequester.cs (115,21): The generic parameter `TResponse' of `OpenSim.Framework.Servers.HttpServer.AsynchronousRestObjectRequester. c__CompilerGenerated0TRequest,TResponse' cannot be converted to the generic parameter `TResponse' of `OpenSim.Framework.Servers.HttpServer.AsynchronousRestObjectRequester.Ma keRequestTRequest,TResponse(string, string, TRequest, System.ActionTResponse)' (in the previous error) [csc] Compilation failed: 2 error(s), 0 warnings BUILD FAILED - 0 non-fatal error(s), 3 warning(s) /home/OpenSim/svn/opensim-0.6.4.9561/OpenSim/Framework/ Servers/HttpServer/OpenSim.Framework.Servers.HttpServer.dll.build(14,6): External Program Failed: /usr/lib/mono/2.0/gmcs.exe (return code was 1) Total time: 0.8 seconds. BUILD FAILED Nested build failed. Refer to build log for exact reason. Total time: 0.8 seconds. -- I used the same process for each version - checkout into a new director, runprebuild, and nant. I'm using mono 2.4 on CentOS 5.2. If anyone can give me some pointers, I'd appreciate it. thanks, brent ___ Opensim-users mailing list opensim-us...@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-users ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Migration SQL invalid causing server to fail to start
Teravus, It looks to me like that patch would fix it. Going by the last parameter missing in each line, I can only guess it was a copy/paste error by whoever wrote that section. To be safe perhaps someone who is running sqlite could make a test copy of their opensim folder and test it first? Thanks, ;) - John / Orion Teravus Ovares wrote: Hey A database guy needs to look at a migration that someone added in SQLite that causes the server to fail during a migration with the following message: 21:46:38 - [SQLAssetServer]: AssetStorage: Attempting to load OpenSim.Data.SQLite.dll 21:46:38 - [MIGRATIONS]: Upgrading AssetStore to latest revision. 21:46:38 - [MIGRATIONS]: NOTE: this may take a while, don't interupt this proces s! 21:46:38 - [OPENSIMBASE]: Asset server SQL not loaded (wrong number of arguments to function substr()) 21:46:38 - [APPLICATION]: Null Reference Exception Supposedly TommReady has posted a patch: http://opensimulator.org/mantis/view.php?id=3439 However, since I'm not a database guy.. I'm simply raising awareness to someone who is a database guy. Best Regards Teravus ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] RFC: Ways of creating profiles for creators who will never log in
I agree Brianna. In the meantime though, I wonder if it would be possible to section waterways off as different plots with build / script enabled along with an autoreturn? Although I'm not quite sure if autoreturn works yet, as in our sandbox we've already had large heaps of space junk lay there for days on end without poofing after 12 hours as per the parcel settings. :) Thanks, :) - Orion Brianna wrote: In the process of making this short HyperGrid YouTube... http://www.youtube.com/watch?v=Lz-x7jn2hwsfeature=channel_page I notice more HG nodes understandability shutting the doors to scripted objects for security reasons. It would be great to have a 'passport' scheme that would allow travel with boats and ballooning in the future days from known Grid citizens. - Original Message - From: Diva Canto d...@metaverseink.com To: opensim-dev@lists.berlios.de Sent: Thursday, March 26, 2009 3:09 PM Subject: Re: [Opensim-dev] RFC: Ways of creating profiles for creators who will never log in Whatever you decided to do (option 3 seems the cleanest), this will probably also be used in the Hypergrid, to keep track of foreign users. I need that! :-) Justin Clark-Casey wrote: Hello, For Inventory Archives I plan to preserve item creator information. When the archive is loaded I would like to recreate these profiles where possible/necessary (grid operators can choose not to allow this and that will be the default, I expect). However, unless an item creator has an account on the OpenSim to which the archive is loaded, they shouldn't be able to login to that instance. So far I've thought of 3 ways to create a profile without automatically allowing login. (1) Create a normal user account but set the password to something random. PROS * Doesn't require any changes to what we have today CONS * Creates user accounts which are never intended to be used for login * No way to distinguish archive created accounts from legitimate accounts ~ (2) Add a 'ProfileOnly' flag to the Users table PROS * Minimal changes to what we have today * Makes it clear that an entries has been created for its profile only, which can be used as a flag to disallow logins CONS * Creates user accounts where many details will be irrelevant unless item creators then get accounts on the instance. * Complicates administration tasks (e.g. create user). ~ (3) Separate the current 'users' table into 'userprofiles' and 'users' tables. 'userprofiles' will largely contain all the metadata about a user that you can see in the profile on the Linden Labs Second Life client today (name, about, interests, 1st life, etc.). 'users' will contain the data associated with a particular account (passwordHash, passwordSalt, homeRegion, homeLocationX, etc.) PROS * Makes it possible to create user profiles without creating user accounts. * Makes it possible to have somewhat separate profile and authentication plugins allow mix match. However, the reuse of avatar name as the login identifier makes things a bit awkward. * Simplifies database understandability - the only people in the 'users' table are those with actual accounts, though on the other hand this does create 2 tables instead of 1. CONS * Short term adjustment pain for systems accessing OpenSim's databases directly * Complicates administration tasks (e.g. create user). I suspect that archiving isn't the only potential use for this functionality. For instance, the Hypergrid may also find it useful to preserve user information when a user rezzes an object on a foreign system. Of the above approaches, I prefer (3) over (2) since it seems to me to be the better long term approach even if there is some short term pain. I'm don't think that (1) is a good option. I've reproduced most of text at http://opensimulator.org/wiki/Creating_profiles_not_used_for_login for reference. Comments? ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Phantom prims and memory usage?
John, So far I've not noticed much of a change in cpu use, but within my configuration its rather difficult to judge as we currently have a mess of scripts that it seems I'm forever having to clean up. So far it seems that the phantom tests have made an impact on memory usage though, as I stated in my last post the initial test using a clean reboot of the server saw a decrease of around 220 megs of ram used. I've not had the chance to continue any further with this as I'm currently under the wire to wrap up my research paper (oddly enough on ethics within virtual worlds), but so far within just general use I've noticed things seem to run a bit smoother. As for cpu, I'd imagine with less things for the avatar to collide with it would potentially be a bit less. I should hopefully have some more data by the end of the week. Thanks, :) - John John Hopkin wrote: John Sheridan wrote: Ok, I'm trying a little bit of an experiment here and I'm wondering if anyone else has tried this and had similar results? We're running ODE on Linux Ubuntu using a vps setup hosted on quad core Xeons. I'm going on the concept that if an object does not need any sort of physical interaction that if its made phantom the server will in effect use less resources. In this case I have one server thats running six sims with around 30,000 prims spread between them all. After going around and setting as many objects as possible to phantom (telephone poles, windows, tables, chairs, misc props and anything thats not needed to keep the avatar from falling out of the buildings) it appears as though that server's memory usage has gone down from about 70% (740 megs) down to 45% (520 megs). I would estimate that in total approximately 15,000 to 20,000 prims were set to phantom. Very interesting - thanks for looking into this. It seems that in actual builds, those we see in use in SL, for example, most prims don't need to be non-phantom but are, and it's not surprising to find that your exercise reduces the cost. I've sometimes wondered if there's a case for saying that new prims should be phantom by default, though I'm not necessarily suggesting that as a course of action for OpenSim. It certainly helps avatar movement when a cleanup like you describe is done, especially with furniture, poseballs etc, and it's well worth the time doing that even if only for the sake of smooth operation. Is there any indication of a corresponding drop in CPU usage, especially when avatars are moving around in there? ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] Phantom prims and memory usage?
Kyle, I agree... If this does turn out to have an impact on making builds more efficient then perhaps it should go in as a default setting? Although I wonder if that would be a server side function or if it would need to be client side. It certainly will be interesting to see. ... - And before I forget as my brains are about to turn into jelly, thank you Dahlia for your explanation. That really does help to clarify just how all of this works. :) Thanks, :) - John / Orion Kyle wrote: I just made most prims phantom on a new build that had been some trouble and it seems smoother now. I will try and get more definitive data but this seems promising and possibly should be considered to have prims phantom by default. Though this would throw off many new builders. -Original Message- From: opensim-dev-boun...@lists.berlios.de [mailto:opensim-dev-boun...@lists.berlios.de] On Behalf Of John Hopkin Sent: Wednesday, February 25, 2009 11:43 AM To: opensim-dev@lists.berlios.de Cc: opensim-us...@lists.berlios.de Subject: Re: [Opensim-dev] Phantom prims and memory usage? John Sheridan wrote: Ok, I'm trying a little bit of an experiment here and I'm wondering if anyone else has tried this and had similar results? We're running ODE on Linux Ubuntu using a vps setup hosted on quad core Xeons. I'm going on the concept that if an object does not need any sort of physical interaction that if its made phantom the server will in effect use less resources. In this case I have one server thats running six sims with around 30,000 prims spread between them all. After going around and setting as many objects as possible to phantom (telephone poles, windows, tables, chairs, misc props and anything thats not needed to keep the avatar from falling out of the buildings) it appears as though that server's memory usage has gone down from about 70% (740 megs) down to 45% (520 megs). I would estimate that in total approximately 15,000 to 20,000 prims were set to phantom. Very interesting - thanks for looking into this. It seems that in actual builds, those we see in use in SL, for example, most prims don't need to be non-phantom but are, and it's not surprising to find that your exercise reduces the cost. I've sometimes wondered if there's a case for saying that new prims should be phantom by default, though I'm not necessarily suggesting that as a course of action for OpenSim. It certainly helps avatar movement when a cleanup like you describe is done, especially with furniture, poseballs etc, and it's well worth the time doing that even if only for the sake of smooth operation. Is there any indication of a corresponding drop in CPU usage, especially when avatars are moving around in there? ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
[Opensim-dev] Phantom prims and memory usage?
Ok, I'm trying a little bit of an experiment here and I'm wondering if anyone else has tried this and had similar results? We're running ODE on Linux Ubuntu using a vps setup hosted on quad core Xeons. I'm going on the concept that if an object does not need any sort of physical interaction that if its made phantom the server will in effect use less resources. In this case I have one server thats running six sims with around 30,000 prims spread between them all. After going around and setting as many objects as possible to phantom (telephone poles, windows, tables, chairs, misc props and anything thats not needed to keep the avatar from falling out of the buildings) it appears as though that server's memory usage has gone down from about 70% (740 megs) down to 45% (520 megs). I would estimate that in total approximately 15,000 to 20,000 prims were set to phantom. Thanks, :) - John / Orion Pseudo ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] [Opensim-users] hypergrid teleports and non-hypergrid simulators
Just a bit of a day dream here, but for the sake of preventing users from becoming stranded on foreign grids / sims I wonder if it would be a good idea to make non-hypergrid sims accept and hold a hypergrid user's profile cache. That way as a user explores and pops around on a foreign grid / sim their home grid and other data would still follow them. Non-hypergrid sims would simply hold and pass this data on to the next teleport while doing nothing else with it. For example this is sort of the data flow I'm thinking of: 1 - Orion logs into Pseudospace at Ellis Island then teleports over to the UCI Gateway on OSgrid * - Ellis Island on Pseudospace passes his HG profile over to UCI Gateway via HG allowing him to return home and access inventory 2 - Orion pops over from UCI Gateway over to Wright Plaza (non-HG). * - UCI Gateway sends Orion's normal and HG profile over to Wright Plaza using normal TP - WP simply holds both profiles and does nothing with them - no access to home servers. 3 - Orion then pops over from WP to Bade Plaza (non-HG) * - Wp sends Orion's HG and normal profile to Bade using normal TP, Bade holds onto it but does nothing with it - still no access to home servers. 4 - Orion realizes he left his oven on at his loft back on Ravenport (Pseudospace) then decides to go home. Teleports from Bade back to UCI Gateway. * - Bade sends Orion's HG and normal profile back to UCI using normal TP, UCI recognizes him as an HG user and allows him inventory and tp home access. 5 - Orion poofs from UCI Gateway back to Ellis Island (Pseudospace's HG gateway) * - UCI Gateway sends Orion's HG and normal profile back to Ellis using HG - Ellis picks him up as a home user then continues normal operation. 6 - Orion pops over to Ravenport (non-HG) from Ellis to turn his oven off before his loft burns down. * - Ellis sends Orion and his profiles over using the normal teleport mechanism - HG profile is stored but ignored. Thanks, :) - John / Orion Pseudo Cristina Videira Lopes wrote: Mic Bowman wrote: we have two simulators with one region each running with hypergrid turned on. we want foreign users to go through one of the gateways to get into our grid. i know that the users can freely teleport around our grid once they have hypergrid teleported to a gateway region. what are the *expected* capabilities for that kind of users once the leave the hypergrid enabled region and move around the rest of our grid? [Sorry, moving this again to -dev, because I can't access my outgoing server on the other account while I'm traveling] Here's what happens. Foreign users carry around information about their servers (user, inventory, home region, etc) that is stored on their run-time profile cached in the regions. When they go from an HG region to another HG region, that information is passed on via the message expect_hg_user, which only HG regions understand. When they go from an HG region to a non-HG region, the receiving region doesn't understand that message, so it knows nothing about those servers for the foreign user. If the TP is far enough away, the profile is removed from the departing HG region (as normal for all TPs), so the user's servers info is permanently lost for the rest of the session. Even if the user comes back to the HG region, the information has been lost. Without that information the foreign user cannot: access inventory (except for the items that are cached in the viewer), go back home, etc... So jumping from an HG region to a non-HG region on the same grid is ok, in the sense that it doesn't crash anything, and that the user's agent is is able to interact with the world, but that user becomes impaired in many ways. The user is still able to chat, build move things (if allowed), TP around, etc. It's just the interaction with his/her home servers that is impaired. ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev ___ Opensim-dev mailing list Opensim-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/opensim-dev
Re: [Opensim-dev] weird idea #2: inworld applications - access to client's widget set?
While on the topic of weird ideas and in world apps... I posted this idea to the Lindens about a year ago back when I was first trying to figure out LSL, but it likely went off to the noobie duh bin as at the time I pretty much asked them to include a copy of Visual Basic in world :P Anywho, as it is we already have the LSL language with our own additions via the os functions. What I'm thinking would likely require client modifications which merely makes it something to think about for the future, but why not cobble together something that gives lsl access to the client's widget set? Optimally something like a Mono Winforms type of addition to lsl that would let a scripter actually use a real gui as an interface for their scripts instead of hacking one out with prims or a dialog box? Thanks, :) John / Orion Pseudo Dirk Krause wrote: Hi, this thing came up when I was thinking about what to do for OpenSims 2nd birthday. I thought it would be really funny to reconstruct the Sony Home Arcades in OpenSim, basically for giggles. I unfortunately don't have access to Sony Home for now so I don't know exactly what effort it means to model this, being not a good builder myself (for reference - http://tinyurl.com/def8fn ) The interesting point would be the ability to play either MAME or C64 games on the machines in these 'OpenSim Home (tm) Arcades'. So I looked up a C# c64 emulator on the web ( http://tinyurl.com/bobw9y ) but then came to think where such an emulator would run. (the following holds probably true for all kinds of applications running in the OpenSim context, namely: - graphic-heavy c# or c++ applications - flash/silverlight/moonlight applications - 'co-browsing', works in Rex with this nice trick: http://therexfiles.cybertechnews.org/?p=183 ) So, to stick with the arcade example, the good question is - where does the process run? I think there are these possibilities in general 1) SERVER - the application totally runs on the server side. One av takes over the game machine and his key strokes are transmitted to the server (via HUD?) and the emulator creates the graphic output. This would be a series of textures (not really good) or a video stream of sorts. 2) CLIENT - the applications totally runs on the client. This is possibly the easiest way to implement it (and out of scope for opensim-dev) since it needs hacking the client. But just for the record: as soon as the client detects arcade.jp2 as the texture, it fires up ye old space invaders and renders2texture the graphic output to the client. Other people would see either a) nothing but the standard texture as long as they are not playing it or b) a screenshot every 5 secs or so, since the client sends every 5 secs or so a screenshot to the server, updating the view for the cheering bystanders c) the real game, since their clients also fire up the emulator, receive the key strokes from the current player (while they are near him) which must be sent from the server of course. 3) BOTH- the application runs on both server and client with synchronicity calls every N secs with some prediction by the client side when the calls don't get through fast enough (basically like networked physics in professional games works) All in all you are in synchronicity hell the more 'real' the output for everyone gets because there can be no real simultaneousness. So sorted by applications: - Physics: either only server sided (like it is now) which is sufficient for most use cases, or both when the physics is fast and heavy like in games. - Video: Number 2c is used to play video in SL right now - one av activates the script that start the media playing on all clients in the vicinity. if they didn't activate media support then they see nothing. If they did the video starts on all clients, probably 1 to N secs off each, depending on their network, also slowly drifting into asynchronicity the longer the video runs. If it should be more synchronous then a streaming server is mandatory. - Turn based games could be implemented completely on server side. So a simple text adventure (Zork, anyone) or even a MUD could be implemented even on a different server with a gateway of sort. Come to think of it this could even be a tty terminal. Same goes for - co-browsing web pages, powerpoint projectors Could be either server sided (like it is now via the php render trick) or client sided (via the Rex trick) So the interesting part stays where to implement, say, a moonlight application? Let's say people want to create micro/casual games or small apps,then it would be interesting to see whether there would be an infrastructure to hook these things into? I would even go so far that there could be a mechanism that handles LL or OS scripts in a way that it either runs on the client (libomv Test.exe with some script) or on the server side (the existing
Re: [Opensim-dev] weird idea #2: inworld applications - access to client's widget set?
Well, I'm glad someone's at least taking a swipe at the client from a usability standpoint although it would be nice if they like a few of the other client projects I've seen would consider starting with a ground up rebuild of the code base. At least not only to solve the gpl vs bsd vs insert favorite license here problem but as well to make it so one can actually read and understand the code without loosing all grip on reality (virtual or not) then curling up into a fetal position under their desk while the world around them crumbles into some sort of bad rendering bug ala The Matrix. And yes, for those who actually can read and understand the Linden's coding style I am that much of a wimp. :) Thanks for the info Dahlia, :) - John / Orion Pseudo Dahlia Trimble wrote: That would most likely require modifications to the sl viewer. I think the Imprudence project is looking at providing a viewer with similar capabilities: http://imprudenceviewer.org/ On Tue, Jan 27, 2009 at 12:18 PM, John Sheridan j...@pseudospace.net mailto:j...@pseudospace.net wrote: While on the topic of weird ideas and in world apps... I posted this idea to the Lindens about a year ago back when I was first trying to figure out LSL, but it likely went off to the noobie duh bin as at the time I pretty much asked them to include a copy of Visual Basic in world :P Anywho, as it is we already have the LSL language with our own additions via the os functions. What I'm thinking would likely require client modifications which merely makes it something to think about for the future, but why not cobble together something that gives lsl access to the client's widget set? Optimally something like a Mono Winforms type of addition to lsl that would let a scripter actually use a real gui as an interface for their scripts instead of hacking one out with prims or a dialog box? Thanks, :) John / Orion Pseudo Dirk Krause wrote: Hi, this thing came up when I was thinking about what to do for OpenSims 2nd birthday. I thought it would be really funny to reconstruct the Sony Home Arcades in OpenSim, basically for giggles. I unfortunately don't have access to Sony Home for now so I don't know exactly what effort it means to model this, being not a good builder myself (for reference - http://tinyurl.com/def8fn ) The interesting point would be the ability to play either MAME or C64 games on the machines in these 'OpenSim Home (tm) Arcades'. So I looked up a C# c64 emulator on the web ( http://tinyurl.com/bobw9y ) but then came to think where such an emulator would run. (the following holds probably true for all kinds of applications running in the OpenSim context, namely: - graphic-heavy c# or c++ applications - flash/silverlight/moonlight applications - 'co-browsing', works in Rex with this nice trick: http://therexfiles.cybertechnews.org/?p=183 ) So, to stick with the arcade example, the good question is - where does the process run? I think there are these possibilities in general 1) SERVER - the application totally runs on the server side. One av takes over the game machine and his key strokes are transmitted to the server (via HUD?) and the emulator creates the graphic output. This would be a series of textures (not really good) or a video stream of sorts. 2) CLIENT - the applications totally runs on the client. This is possibly the easiest way to implement it (and out of scope for opensim-dev) since it needs hacking the client. But just for the record: as soon as the client detects arcade.jp2 as the texture, it fires up ye old space invaders and renders2texture the graphic output to the client. Other people would see either a) nothing but the standard texture as long as they are not playing it or b) a screenshot every 5 secs or so, since the client sends every 5 secs or so a screenshot to the server, updating the view for the cheering bystanders c) the real game, since their clients also fire up the emulator, receive the key strokes from the current player (while they are near him) which must be sent from the server of course. 3) BOTH- the application runs on both server and client with synchronicity calls every N secs with some prediction by the client side when the calls don't get through fast enough (basically like networked physics in professional games works) All in all you are in synchronicity hell the more 'real' the output for everyone gets because there can be no real simultaneousness. So sorted by applications: - Physics: either only server sided (like it is now) which is sufficient for most use cases, or both when the physics is fast and heavy like in games. - Video
