Fixes Mantis #3979
Applied with changes. Changed ThreatLevel to High since all discovery
functions are a high threat. Overriding that is the responsibility
of the grid owner.
Date: Wed, 5 Aug 2009 12:45:56 +0200
Subject: [PATCH] A few minor tweaks to the MRM API's in order to make it possible for MRM's to run in a separate AppDomain without poluting the primary appdomain of OpenSim
Specifically:
Added an explicit method for getting the "globals" of the MRM, removing the need to have the MRM script code loaded into the primary domain, in order to set up proxies
Added a [Serializable] attribute to TouchEventArgs, again in order to remove the need to have MRM script code loaded into the primary domain.
---------
Applied with whitespace changes
* MRM Scripts should do appropriate cleanup within this event, to allow for clean shutdowns and script updates. This means unbinding from events you are listening to, and releasing any resources.
|Date: Wed, 5 Aug 2009 09:51:52 -0700
|Subject: [PATCH] Closed two major memory leaks for scripted objects
|
|Two major memory leaks for the scripted objects were fixed
|- One leak had to do with remoting acrossing app domains. When a script and
| its controlling agent communicate across an application boundary, it calls
| functions on a stub proxy object that then invokes the remote method on
| the object in the other app domain. These stub objects (two for each script)
| were setup to have infinate lifetimes and were never being garbage collected.
|- The second leak was the result of adding a scene object part instance method
| to a scene event and never removing it. This cause the event's delegate list
| to maintain a link to that object which is then never freed as the scene event
| object is never destroyed.
Patch applied, please direct feedback to me. Possible issue: Longtime idle
scripts like vendors may fail.
[1] Added a new OnAttach event to Scene/EventManager
[2] Hooked up existing attach event handler in XEngine.
[3] Modified SceneGraph and Scene.Inventory to trigger
attach events at the appropriate places. I was forced
to distribut the changes across two files because of
the way attach processing is distributed across the
two files.
[4] In the case of RezSingleAttachmentFromInventory it is
necessary to handle event scheduling in a special way.
There is no synchronous path available, so the fact
that this object is attached, and who it is attached to,
is cached when the ScriptInstance is created. When
the script is started, the attached handler is driven
after on_rez (but before changed, this should be reviewed).
Signed-off-by: dr scofield (aka dirk husemann) <drscofield@xyzzyxyzzy.net>
* These have been mostly replaced by the PriorityQueue sender in LLClientView (it was still using threads and poll processing!)
* Thread Reduction!
* Might have been sending more image packets then necessary previously, texture times 2. We'll see if this reduces the texture packet load.
* If this doesn't melt the Http Server, this will significantly reduce the number of threads in use on regions with many users.
* Adds AddPollServiceHTTPHandler, and RemovePollServiceHTTPHandler to BaseHttpServer
* Generic enough to be used for many long poll services, not only the EventQueue.
* Made various changes to BaseHttpServer to accommodate the new interfaces.
* This version has been significantly re-architected and may fail in unusual and insidious ways.
* Please pay attention to any errors you get and post a Mantis if you can reproduce an issue with the HTTPServer. I'm including the pdb and having the http server compiled in debug for a few weeks so that when an error occurs, it'll print the pertenant data.
* Once again, this is the full C# WebServer, not the lite branch that is included in LibOMV (we need SSL!)
* Test for prim obstructions between the avatar and camera. If there are obstructions, inform the client to move the camera closer. This makes it so that walls and objects don't obstruct your view while you're moving around. Try walking inside a hollowed tori. You'll see how much easier it is now because your camera automatically moves closer so you can still see.
* Created a way to know if the user's camera is alt + cammed or just following the avatar.
* Changes IClientAPI interface by adding SendCameraConstraint(Vector4 CameraConstraint)
This involved adding a new OnUpdatePrimSingleRotationPosition event to IClientAPI so that we can get the changed position from the client.
Btw adding new events to IClientAPI is really tedious where you have to copy the change across to at least 5 or 6 other files.
[Note this doesn't fix the bug where any rotation changes to the root prim (but not the whole linkset) cause rotation errors on the child prims.]
* Remove the draconic locking around adding an avatar to the Scene
* Handle an extreme error case when border crossing fails and user uses map to teleport to a different region on the same instance causing control commands to go to a child agent.
* Make the Set Appearance method use the proper 'remove from physics scene' method.
* It *may* help border crossings.
* It *may* help the 'on avatar rez' lag, that people have been seeing the past week.
* It may also cause physics to crash more often on failed teleports (though.. I think I got the cases covered).
This fix re-introduces a small tilt into the capsule to prevent
avatar falling through terrain. Re-introduction of the tilt means
that some direction-dependent behavior when walking over prims, but
I have tried to minimize this.
Additionally this commit allows the capsule to wobble slightly when
being pushed around the terrain. This should make walking over prims
easier, as the capsule can wobble and glide diagonally over the prim's
edge, instead of rigidly being stopped vertically against the prim's
face.
this commit finally adds the VivoxVoiceModule: it supports positional
as well as conference call type voice (currently only per region
server), region and parcel voice, speaker indication (LL client
family), direct avtar-to-avatar voice chat. NOTE: you need to obtain
an customer admin account from Vivox to be able to use this module ---
DON'T ask me about how to about an admin account, i've NO clue, we
just wrote this code.
The vector parameter to llSetPos() specifies an absolute position for
an unlinked prim or the root prim of a linkset; however, when the
function is used by a child prim, the vector specifies a relative
offset from the root prim. The changed introduced in llSetPos() treats
this value as an absolute position in all cases, which has the unintended
effect of mangling the position of child prims.
This add a configuration option to the MRM module called "hidden".
if MRM is marked as enabled, the module will additionally check for
the "Hidden" flag, before registering for client side scriping events.
When MRM is running hidden, it will not respond to client side
scripting events, giving serverside scripting modules, like MRMLoader
and MRMAddin the ability to leverage the MRM engine. This way, even
a possible clientside exploit will not be possible, while still
allowing the MRM engine to run.
The admin_close_region method removes a region from the simulator without deleting it.
The region can then be recreated by calling admin_create_region with the same UUID.
There is also a change to admin_create_region to facilitate this.The reason I want
to have this functionality is to make it possible to detach regions when they are
idle and recreate them on demand through a web interface. It's probably doable
using the existing methods by saving and loading oars, but it also doesn't seem
like that should be necessary.
Set av_capsule_tilted to false in opensim.ini. Default is true, so there is
no change in avatar behavior (and no breaking of existing content which
relies on the tilted capsule).
This commit straightens up the avatar capsule so it behaves consistently
(e.g. same collision behavior against prims regardless of which direction
the avatar is coming from; ability to fit through narrow doorways).
Please note this introduces other side effects which have not been fixed.
In particular:
* The avatar frequently falls through the terrain if it is not flat, though
the avatar behaves pretty well on flat terrain. This requires investigation
of the ode terrain collider.
* The apparent foot position of the avatar with respect to the ground
is changed. This requires investigation of the avatar height/capsule height.
Please consider this as work in progress.
The patch included updates the root and child prims' AttchedAvatar
with the right UUID. It also cleans the AttachedAvatar properties
for the root and child prims on Drop and Detach
balancer plugin work again. Create a new method, GetClientEP, to retrieve
only the EndPoint for script usage. Marked the purpose of the method
in IClientAPI.cs with a warning. Also restored the corresponding SetClientInfo
functionality.
to be constantly pushed and popped uselessly, if you are not debugging packets. This showed some really big difference in a mock test, let's see how it behaves here.
Attached is a patch that changes the oar file saving of creation date/time to an integer
instead of a string. I did this after justincc emailed me saying there is a problem
with internationalisation doing it the old way and I said I'd fix it. Its been
tested with MySQL and I've made the changes for MSSQL but that hasn't been well tested.
The new IRCd module causes an error when multiple instances of OpenSim
are run on the same machine; since the port number (6666) is hardcoded,
the second and subsequent instances crash upon startup because the port
is already in use. Attached is a patch which adds a Port specifier to the
[IRCd] section of the config file, which defaults to 6666 if not present.
By extracting this code, it should make it easier to experiment with different ways of managing the update process. [Next step to make this module based, could be to create a SceneViewerFactoryModule]
I also fixed the issue where the code also loops through the m_forcesList and copies each force to the ScenePresence's movementVector. Which resulted in only the last force in the list actually be acted on. As each copy overrode the last one. So now it only copies the last force in the list.
After noticing on several occasions that the thread counts
we see when running OpenSIm were bordering on the astronomical
I decided to seriously investigate.
After much poking I discovered that the problem is actually very
simple. The XEngine secition of the example ini says that the
timeout for an iden thread is in seconds, and an example value
of 60 is specified. In fact, this is actually resulting in a 60mS
idle timeout, which is not normally enough for a smart thread
to survive. I have added a multiplier to the XEngine constructor
so that the number now matches the published behavior.
bit of GetClientInfo that is actually used seems to be userEP as part of the
OSSL osGetAgentIP() script function. Now commented are the parts where
we serialize and copy out the *entire* packet queue of the client
(locking the packet handler in the process).
- uses Environment.TickCount for all timestamps (instead of more
costly Util.UnixTimeSinceEpoch()
- takes care of Environment.TickCount overflow (which will happens
after 24.8 days of system uptime)
- avoids instantiating List copies for each check
- gets rid of one lock() invocation
- moves calculation of loop invariant variable out of the loop itself
option for LLUDPServer. On windows .NET the default socket receive
buffer size is 8192 bytes, on recent linux systems it's about
111K. both value can be a bit small for an OpenSim instance serving
many clients. The socket receive buffer size can be configured via
an OpenSim.ini config option
- adds a general catch clause to LLUDPServer.OnReceivedData() to
prevent it submerging when an unexpected Exception occurs.
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.
out which client connection has closed. So, in multi-region sims, things
can get messy fast.
This introduces a second parameters, which is a Scene object ref. Minor
adjustments to custom modules may be required due to this change.
- Set OwnerID = GroupID for deeded objects.
- Close a security loophole that would have allowed a user with deed rights in a group to deed ANY object to that group, even if it's not owned by them and/or not set to that group
- Set LastOwnerID correctly. Handle objects vs. prims correctly.
(a) Implements the ICommandableModule interface to clean up the user interface
(b) Uses a specification for a 'copse' (collected group of trees) which permits via an xml file: Tree type; Tree Line (high and low), Seed point; Radius of Copse; Number of required trees; Initial size of seeded tree; maximum size of seeded tree; growth rate; freeze growth state
(c) Multiple Copse may be defined for a region
(d) Growth on individual copse may be frozen or restarted, or all growth disabled/enabled
(e) Copse definitions are persistant, they are reloaded from the trees present on a region restart
(f) All trees in a copse may be removed and the copse definition deleted in one command
* This is necessary to avoid a dependency problem where LLStandaloneLoginModule references IInventoryService before the LocalInventoryServiceConnector is registered
* Correct some copypasta log messages in LocalInventoryServiceConnector
Changing the names of these methods because they were being picked up by
nunit as tests even though they were marked private. Naming them Check*
after the original Test*.
Several improvements in the connectors themselves.
Several improvements in configurations.
Needed to add a hack in IUserService and UserManagerBase, to be removed when that service is refactored.
Eat collision errors --- NOTE: this fix might be naive, it seems to
have helped us getting to 81 avatars (whereas we'd crash with 20
before), but it sure would benefit from some check-over by a person
skilled in the art of ODE physics.
This change addresses two issues:
[1] It adds a flag field to the blendface call which allows the
caller to indicate whether or not the generated asset is
temporary, and whether or not the asset being replaced should
be explicitly retired fromt the memory cache. The decimal
values correspond to:
0 - Permanent asset, do not expire old asset
1 - Permanent asset, expire old asset
2 - Temporary asset, do not expire old asset
3 - Temporary asset, expire old asset
'3' corresponds to the default behavior seen today, and is
the continued behavior of the non-blendface calls.
[2] The dynamic texture routines are highly-asynchronous and can
be scheduled simultaneously on a multi-core machine. The nature
of the texture management interfaece is such that updates may
be lost, and the nature of asynchornous operation means that
they may be processed out of order. A lock has been added to
ensure that updates are at least atomic. No attempt has been
made to enforce ordering. The lock applies to the SceneObjectPart
being updated and is held for the lifetime of the TextureEntry
used to carry texture updates (the one instance carries all
faces supported by the prim).
Users of these services should remember that the dynamic texture
call is asynchronous and control will be returned *before* the
texture update has actually occurred. As a result, a isubsequent
GetTexture call may not return the expected asset id. A script
must wait for the corresponding TEXTURE_CHANGED event before
retrieving any texture information.
SL's LSL supports lone idents:
integer x;
x;
as well as lone idents in for-loop assignments:
for (x; x < 10; x++) { ... }
while those are errors in C# (MONO at least). This patch skips lone
idents in such places.
Fixes Mantis #3042.
For loops with no assignment are no longer syntax errors. For example,
this is now valid:
for ( ; i < 10; i++) { ... }
Corresponding changes to lsl.{lexer,parser} in r99 in opensim-libs.
Fixes Mantis #2501. Fixes Mantis #2884.
It seems overloading the method TestllAngleBetween confused nunit a bit
and caused it to try running the private TestllAngleBetween method. The
method's access modified prevented that from happening, and nunit
complained.
* For some reason, if a null was recieved (indicating a missing asset), the code had stopped passing that on to the waiting lock, resulting in a perpetual freeze
* This change passes the null on correctly
* Many thanks to thomax for being insistent in presenting his analysis of the problem :)
* Issue was that region server was silently dropping an XmlException caused by trying to deserialize the blank asset service response
* So make asset service return http status NOT FOUND rather than OK in accordance with REST
* and interpret this correctly in the async response so that a null object is sent back
* This means that this fix won't be active until both region simulator and server reach this revision
* If an oar save fails to get responses to all asset requests to the asset service then timeout after 60 seconds
* Timeout executes abort, since missing assets in an OAR seems bad
* This means that oar saves won't permanently hang and instead can be retried if something goes wrong with the asset service
* This is not a solution to mantis 3714. Hopefully a fix will be along shortly since I can now consistently reproduce that problem
This patch adds oar file date and time (UTC) meta data to an oar file
when it is created. It also adds a unique ID, though this id does not
in anyway identify the machine that the oar file was created on.
When an oar file with this meta data is loaded this extra information
is saved with the region settings and available via LSL through:
- osLoadedCreationDate()
- osLoadedCreationTime()
- osLoadedCreationID()
If there is no meta data these fields will be blank. Subsequent oar
file loads will erase the information for the previous oar file
load. Persistence has only been implemented for MySQL, the other
backends need updating.
Overall this allows us to much more easily identify the specific version of
software that clients are using. Its very straightforward to edit the oar file
to change the ID string to be something more human friendly.
Included in the patch is a new file OpenSim/Data/MySQL/Resources/030_RegionStore.sql
required for the MySQL DB migration.
btw I had a chat with justincc about this a few weeks ago since he
wrote the oar file import/export and he sounded happy to accept
something that included date/time information but didn't want anything
that would silently leak private information like machine names.
- fixes wild swings in memory usage related to usage of GetDrawStringSize()
We've been seeing wild swings in memory usage and a large chunk of
memory leak. From analysing this it's pretty clear that the mono
garbage collector is rather buggy! When exercised heavily it looks
like it frees more than its meant to resulting in crashes.
GetDrawStringSize() measures the size in pixels of text. To do this
memory for an image is allocated and used to call the GDI text
measure functions. Although no reference to the temporary memory
for the measuring is kept, it takes quite a while for the mono
garbage collector to clean up - so if lots calls to
GetDrawStringSize() are made at once there can be a spike in memory
usage. If the garbage collector is not fast enough then the GDI
layer runs out of memory. It also looks like the garbage collector
is not always reclaiming all of the memory.
I've attached an OpenSim patch which works around the garbage collector
issues. Instead of dynamically allocating memory for measuring
text sizes, it serialises (on a per region basis) access to a single
block of memory. The effect of this is to be nicer to the garbage
collector as it has a lot less work to do, at the cost of some
theoretical loss in performance (nothing noticeable with our tests
which hit it pretty hard).
OpenSim still does leak memory slowly, but it is a lot more stable
with this patch. I suspect that either the garbage collector misses
bits of freed memory or the GDI/cairo layer leaks a bit each time a
texture is created. Thats going to be a lot harder to hunt down, but
for reference if someone has OpenSim running on Windows it would be
interesting to see if it has the same problem as it would tell us if
its a mono/GDI problem or an OpenSim problem.
Modify dynamic texture handling so that an explicitly targetted
face is not scheduled for immediate expiration. The requirement
for precaching explicitly requires these assets to persist. They
do however remain temporary.
This approach leaves the legacy mode of operation (ALL_SIDES)
unchanged in this respect.
See the files: bin/config-include/GridCommon.ini.example and bin/config-include/StandaloneCommon.ini.example to configure and enable this caching method.
See the files: bin/config-include/GridCommon.ini.example and bin/config-include/StandaloneCommon.ini.example to configure and enable this caching method.
This appears to be due to the fact that no asset cache has been
configured, possibly as a result of the configuration changes that
have been made recently. I've attached a patch to display a message
to that effect rather than throwing an error.
The IRCStackModule used Nini.Config.ConfigCollection.Contains() to
determine whether the "IRCd" section was present in the config. This
ConfigCollection, however, stores an ArrayList of IConfig objects, not
strings, so calling Contains("IRCd") always returns false since "IRCd"
is a string, not an IConfig object.
* This includes methods such as PlaySound which take a Position as an argument, allowing you to trigger sounds arbitrarily across the scene without needing a parent object in the position.
The attached patch implements llPassTouches. It has been added
to the export/import XML along with the flag for AllowedInventoryDrop.
The MySQL backend has been updated as well, though I haven't
done one of those before so could do with a check. I added
the migration mysql file as well.
The other data backends need updating as well.
* unfortunately, while the client requires uuids and we want to be able to have arbitrary string ids, these cannot be kept in sync
* I think the problems last time were due to a serialization change
* So the major inteface version has been bumped to take care of any lingering issues here.
* This means that region servers beyond this revision can only connect to similarly uptodate grid services, and vice versa
The attached patch implements llPassTouches. It has been added
to the export/import XML along with the flag for AllowedInventoryDrop.
The MySQL backend has been updated as well, though I haven't
done one of those before so could do with a check. I added
the migration mysql file as well.
The other data backends need updating as well.
* If you have something using XmlRpc that isn't in core, change your method signature from:
(XmlRpcRequest request)
to:
(XmlRpcRequest request, IPEndPoint remoteClient)
Changes to support client-side image pre-caching in the region. This
commit adds an additional calling sequence to the DynamicTexture data
and URL calls. The new interface allows a dynamic image to be loaded
into a specific object face (rather than the mandatory ALL_SIDES
supported today.
This is in part fulfilment of ticket #458.
RequestUserInventory is supposed to drive a supplied callback
when it completes. In fact, it fails to do so if the user's
inventory does not exist (e.g. the inventory database is
unavailable for some reason), and the requestor is left
sleeping forever.
The code has been modified to return empty lists via the
callback as an accurate reflection of what is there: nothing.
treated as another ini source to load.
For example:
Include-Asset = AssetSetup.ini
will load AssetSetup.ini after all other ini files are done.
This works recursively, too
The image render module is returning everything twice. Once with
data, once with null. This change adds a return to stop this
behavior. This was not apparent until I added a message to the
catching routine which issued a warning message when no data was
returned.
We've encountered problems with textures never fully downloading and
objects not moving or being deleted (from the client's point of view)
even when the bandwidth settings on the client have been set very
low. This can happen over reasonably lossy links (eg you're on the
other side of the world from the server) as the server retries 3 times
and then gives up.
Whilst its possible to set ReliableIsImportant, this forces the server
to keep retrying no matter what which potentially could lead to
problems. This patch allows for the setting of MaxReliableResends
explicitly (is set to 3 normally) in OpenSim.ini so if you know you
will have clients connecting with poor connections you can set it a
bit higher (10-15 works quite well even for very poor connections).
This may break a lot of things, but it needs to go in. It was tested in standalone and the UCI grid, but it needs a lot more testing.
Known problems:
* HG asset transfers are borked for now
* missing texture is missing
* 3 unit tests commented out for now
Adding a jsonp wrapper to the user supplied status report
uri if the key "callback" exists. It will work with many
javascript toolkits to provide an ajax callback to allow
the browser to update stats reports without the
intervention of an intermediate server.
This is to handle the changes in the v1.23 viewer
of LL regarding the adult rating. With this patch
a region can be changed to the adult rating from
LL viewer v1.23 and above.
* User interface is ... primitive at best right now.
* Loads bans from bans.txt and region ban DB on startup, bans.txt is in the format of one per line. The following explains how they are read;
DNS bans are in the form "somewhere.com" will block ANY matching domain (including "betasomewhere.com", "beta.somewhere.com", "somewhere.com.beta") - make sure to be reasonably specific in DNS bans.
IP address bans match on first characters, so, "127.0.0.1" will ban only that address, "127.0.1" will ban "127.0.10.0" but "127.0.1." will ban only the "127.0.1.*" network
default but can be enabled in OpenSim.ini. If enabled, things can be sold
for $0. Other amounts will cause the buyer to see a message and the transaction
will fail.
further ACL stuff:
- adding StrictAccessControl variable: DON'T set this to false if you
want to enforce ACL, it will disable ACLs right now. Default is
true.
once we've got code added to allow child agents but prevent them
from becoming root agents when the ACL denies access to the avatar,
setting this to false will then allow avatars to see into a
neighboring region but not enter it (currently ACL prevent both,
seeing and entering).
- enhancing log statements
- AssetType isn't InventoryType. Those enums contain different numbers. Use AssetType for the asset type, InventoryType for the inventory type.
- The ToString method (or ToLower) of AssetType/InventoryType doesn't necessarily return the correct LLSD string.
- Replaced several magic numbers by their corresponding enum.
- Fixed the invType for gestures and animations in the library.
This should fix Mantis #3610 and the non-terminating inventory loading
* The OpenSimulator core developers have voted to remove all currency functionality from OpenSimulator leaving the 'IMoneyModule' interface in. This affects all systems that used the example money module. This effects All systems that used the XMLRPC External Money Module Hooks interface. If you previously used this interface, please consult with the OpenSimWi Redux folk who are keeping the old module with this interface up to date.
* A notice to the opensim-dev mailing list to come as well.. since this is likely a breaking change for some.
* Create the OpenSim.Region.ReplaceableModules namespace for modules that we intend to have people replace (see readme)
* Create the OpenSim.Region.ReplaceableModules.MoneyModule namespace
* Put our current Sample MoneyModule in this namespace. (more modifications here next commit)
OSP resolver, and the caching of the uuid seperate from the string that is a UUID.
Change this behavior back to something that ensures the data for the 2 is the same.
Put the 2 unit tests that depend on the new behavior into ignore state.
Adding more SL likeness for Email module in CORE.
I've added some SL likeness to the Email module so
that it looks more like emails going out in the
same standard as SL uses
Scene.NewUserConnection()
- adding reason reporting
this enforces estate bans very early on and prevents us from
circulating client objects that we'd then have to retract once we
realize that the client is not allowed into the region
console capable of processing commands. Create LocalConsole as a console
that uses cursor control and context help. Precursor to a distributed
console system for the new grid services. No functional change intended :)
Move json stats to non-published resource name
Remove well-known resource name for json stats,
creating dynamic uris with private keys and add
a user configurable resource name for region owner usage.
Different people using Hippo 0.5.1 report that trying
to send group instant messages crashes the viewer
(Hippo 0.5.1). This is the case even for empty groups
or if all group members are online.
Don't allow packets to be resent before they have actually been sent for the
first time. Switch from serializing a packet to get it's length to the LibOMV
provided Length property. Fix resend timing. Fix the use of dangling references
to Acked packets. Fix the packet handler to play nice with the packet pool.
Fix the packet pool. Add data block recycling to the packet pool. Packet pool
is now ENABLED by default. Add config option to disable packet and data block
reuse. Add ObjectUpdate and ImprovedTerseObjectUpdate to the packets being
recycled.
to recycley data blocs within a packet. Recycle the ObjectUpdate* data
blocks. Speeds up loading even more.
This may mean that the packet pool is now viable.
The attached patch provides the necessary infrastructure to support
security and authentication features of the xmlrpc server.
* Read/Write keys for accessing a Group's xmlrpc service.
* Requiring user session verification for write operations.
arbitrary combinations of root and child prims from one or multiple
link sets. Please test throughly and consider things UNSTABLE until
this is proven out.
If an avatar is sitting when the client disconnects, the avatar
is not disassociated from the SOG on which (s)he was sat. This
produces any, and varied, effects.
I have updated RemoveCLient in Scene, to check, and stand the
client up immediately prior to disconnect. This seems like the
most robust way to handle the situation. Though in this case
it might be worth factoring out the animations from other
standup processing. It does no harm, but in this case it is
entirely redundant.
Added support for access control lists.
Scene: Added test to AddNewClient for an entry in the access
list when connecting to a region with limited access.
EstateSettings: Added an HasAccess(UUID) property to test for
an entry in the estate's access list.
RemoteAdmin: Add RPC calls for admin_acl_list, clear, add,
and remove.
llSetPrimitiveParams in a large linkset can disrupt the
entire region. However, when the script is in a large
linkset, it appears to totally lag out the scene and
stops updates from being sent.
The attached patch fixes a few problems that people were
having with the Messaging provided by the XmlRpcGroups
optional module, namely:
* Fixes 2x echo in group messaging
* Fixes problems with cross instance, non-neighbor, messaging
Some other IRC timing wrinkles showed up:
[1] If connect processing blocked in socket activation, then
the watch dog saw the session as connected, and eventually
tried to ping, but because the socket create was still
blocked, it barfed on a null reference. This then drove
reconnect. Changed the watchdog handler so that it only
tries to ping connections that are connected and not pending.
[2] If the socket creation actually fails, then the connect and
pending flags were reset. This resulted in the connection
being retried at the earliest possible opportunity. The
longer login-timeout is preferrable, so the status flags
are not reset, and the failed login is eventually timed
out.
[3] The Inter-connection interval is primed so that the first
session can connect without delay.
* Make OGS1UserServices inherit from UserManagerBase
* This allows grid mode regions to use the same user data plugin infrastructure as grid servers and standalone OpenSims
Changes to enable script state persistence across non-restart
serialization situations (inventory/OAR/attachments)
Also fixing test cases for OAR and IAR so they don't barf with the new code.
An attachment with the physical checkbox checked will not allow the
phantom checkbox to be cleared. This interfers with scripting
functions such as llMoveToTarget(), which won't work while an object
is phantom. If the prim containing the script is rezzed to the ground,
it will then allow the phantom checlbox to be cleared, and the script
works as expected.
* The change shouldn't affect anyone who has it working currently and makes it a ton easier for everyone else to get it working.
* Handle a case when there's no Event-Calling-Function but it's obviously a REGISTER method
* Refactors the xmlrpc calls to a single location to
make it easier to debug and include alternative
xmlrpc call mechanisms
* Includes an alternative xmlrpc call mechanism that
sets HTTP Keep-Alive to false which solves nearly all
System.Net exceptions on some windows environments
Added support for loading bare asset binaries (as opposed to
xml encoded asset base) to both sandbox asset service and cable beach.
* Added support for enabling region asset service when mxp is enabled.
* Moved base http server content type defaulting before invocation of
request handle method to allow for variable content type in the response.
This patch adds few properties to ScenePresence and
thus allows region module or MRM script:
1. Force flying for avatar or,
2. Disable flying from avatar
Fixes IRC reconnect problem
When a session fails to establish, the login attempt eventually
times out and the login is retried. This should occur once every
25 seconds (to give the server plenty of time to respond). In fact
the interval was typically only 10 seconds, this was being caused
by a second reset that was being scheduled when the failed
listener thread was terminated. Because the second reset occurred
inside the ICC timeout, it eventually gets scheduled after only
10 seconds.
In addition to this, the connector was being added to the monitoring
twice. This was harmless, but entirely redundant.
Both of these problems have been fixed and tested. Each connector
now maintains a count of how often it has been reset. The listener
thread records this value on entry and checks for a change on exit.
If the counts are the same, then the listener is exiting and can
potentially reschedule the connection.
* Example in region module:
Scene.GetModuleInterface<IMRMModule>.RegisterExtension<IMyInterface>(this);
* In the MRM:
//@DEPENDS:MyExtensionModule.dll
...
Host.Extensions<IMyInterface>.DoStuff();
that region. I decided against sending the terrain on every call to osTerrainSetHeight
(which makes it abysmally slow), and added a osTerrainFlush instead, which should be
called after all the terrain-changes have been done. Changed some return types to
LSL types, too, and removed some end-of-line spaces.
This hooks up the LandManagementModule to handle the DeedParcelToGroup
packet. Now people can start testing land assigned to and owned by groups.
Also fixes a viewer crash issue when searching for and then joining a group
with an agent that is not already being tracked by groups server.
Added is a patch that adds a rough Groups implementation. This patch allows
the creation, adding and maintaining Groups, Roles and Members. Work has begun
on a very naive implementation of messaging, and minimal support for notifications
{no attachments yet}. Proposals are not yet supported, but are on the to-do list.
This implementation is not active by default, and must be configured in
OpenSim.ini to become active.
Following feedback from 0003440, i've made some changes to the new texture pipeline to optimise
performance. The changes are:
- Fixed a math issue where a small percentage of images with a certain size (on the packet boundary) would not have their final data delivered. This issue has been present since pre- 0003440
- It was suggested that a discardlevel of -1 and a prioriy of 0 meant to abandon the transfer, this is incorrect and caused some textures to clog.
- The texture throttle blocking queue is now only filled in relation to the actual throttle amount.. i.e, on a connection throttled to 300k, only twenty packets will be placed in the queue at a time, on a larger connection it will be much more. This is to balance responsiveness to requests and speed, and to minimise wasted packets.
- The engine now keeps track of the number of pending textures, and the stack will not be walked if there's no textures pending, saving CPU. Textures are only considered "pending" when they've already been decoded.
- As part of the above, some textures may receive twice as much data per cycle if the number of pending textures is below the cycle threshold, this should prevent loading from slowing down when there are fewer textures in the queue.
Adds a test to see if the first option on osDynamicTextureData is "AltDelim",
then picks up the first character after the whitespace and uses as a delimiter
instead of ;. If this string does not appear at the start of the data, the
default ; will be used, hence this should not break existing code.
WindModule plugins --- manifesting itself through:
2009-04-16 15:32:02,764 [Heartbeat for region sea 3] [Scene]: Failed with exception System.IndexOutOfRangeException: Array index is out of range.
at System.Random.Sample () [0x0003e] in /usr/local/src/mono/build/mono-2.0.1/mcs/class/corlib/System/Random.cs:91
at System.Random.NextDouble () [0x00000] in /usr/local/src/mono/build/mono-2.0.1/mcs/class/corlib/System/Random.cs:142
at OpenSim.Region.CoreModules.World.Wind.Plugins.SimpleRandomWind.WindUpdate (UInt32 frame) [0x00019] in /tmp/opensim-deploy-oTyFP12501/opensim-deploy/OpenSim/Region/CoreModules/World/Wind/Plugins/SimpleRandomWind.cs:92
derived classes
- make OpenSimBackground inherit from OpenSim instead of OpenSimBase
so that it will have a MainConsole instance and we can use console
commands, setting m_gui to false
* Tweak the LLSetStatus results in the ODEPlugin. Hopefully it's a little less unstable.
* ODEPlugin is using experimental math for LLSetStatus, use with caution! :)
* This may eliminate the occasional archive test freezes, since they appeared to occur when somehow the asset server didn't pick up on the presence of a request in the asset
quque
* WebStatsModule doesn't crash on restart. GodsModule doesn't crash when there is no Dialog Module. LLUDPServer doesn't crash when the Operation was Aborted.
* ODEPlugin does 'Almost NaN' sanity checks.
* ODEPlugin sacrifices NaN avatars to the NaN black hole to appease it and keep it from sucking the rest of the world in.
- Fixed some locking issues. Either lock, or don't (if you don't have to). Only locking access half of the time won't work reliably.
- Had to adapt test helpers that use the "old" IRegionModule. TerrainModule isn't one anymore.
* Introduced preprocess step in FetchAsset (Might revert this later)
* Some minor CCC
* Added actual implementation of GetUserProfile( uri ) and the corresponding handler to OGS1.
* Introduced non-functioning GetUserUri( userProfile) awaiting user server wireup (this might move elsewhere)
This is sent to all root prims of all attachments of an avatar when the
animation state changes. llGetAnimation() can thenbe used to find the
new movement animation. This eliminates the need for fast timers in AOs
* Interfaces now live in Interfaces subdirectory.
* Namespace does not yet reflect this change.
* Final namespace for MRMs will probably sit somewhere around OpenSim.Extend.MRM[?]
Fixes Mantis #3421
Change the agent/avatar events subscriptions to just OnNewClient. The data only needs to be sent once and keeping track of log ins/movements is not required. This will also send cloud data to child agents so that they can see clouds above neighbouring regions not just regions that they have visited.
of objects from the scene, and of scripts from objects. This facilitates
the development of modules that can register prims with externall servers
for inbound email and XMLRPC. Currently implemented in XEngine only.
Also applying cmickeyb's compiler locking patch, since it seems risk-free.
* Revamps the server side texture pipeline
* Textures should load faster, get clogged less, and be less blurry
* Minor tweak to ensure the outgoing texture throttle stays private.
* Fixes mantis 3440
Addresses Mantis #3381
The current implementation works as expected if the object has no rotation or
only rotation around the Z axis; you can spin the object left or right (around
the world Z axis).
It works a little unexpectedly if the object has a non-Z-axis rotation; in this
case the body is spun about its local Z axis, not the world Z-axis. (But SL
also behaves oddly with a spin on an arbitrarily rotated object.)
* Add cursory integration with script engine.
* LLMoveToTarget, LLSetBouyancy, LLSetStatus (Physical only), LLApplyImpulse, LLApplyTorque, LLPushObject.. etc.
* Still missing linked physical active and LLSetStatus with an axis lock.
* Enables border crossings when using the BulletDotNETPlugin
* Enabled variable time steps in BulletDotNETPlugin
* Still no 'linked physical objects' yet
* Still no script engine integration
* Now, the 144 unit tests takes roughly as long time to run (16s on my laptop) that the 10 long running takes. The database tests takes forever.
* Feel free to run the unit tests as you code, and the rest before commit.
The attached patch implements osKey2Name and osName2Key
which converts between a UUID key for an avatar and
an avatar name and vice-versa.
osKey2Name is similar to llKey2Name except that it will work
even if the avatar being looked up is not in the same region
as the script.
* This allows you to utilize System.Drawing tools on textures within the region.
* Example: use System.Drawing.Bitmap to make your texture, then use Host.Graphics.SaveBitmap to make an asset from it in JPEG2K. You can edit (but not overwrite) existing textures using Host.Graphics.LoadBitmap.
* This makes MRM scripting ever so slightly more secure. If you have enforced Object Permissions enabled, it may be acceptable to enable MRM within your regions.
* Security bug reports on this feature are much appreciated (eg: anyone finding ways around this to execute a MRM as a basic user).
* Corrected issue with MRMs where it would attempt to overwrite an already loaded DLL. (and thus fail with cryptic UnauthorizedAccessException.)
* Made DrunkenTextAppreciationModule.cs MRM not crash with StackOverflowException
* Added some temporary logging to MRM World.*
* Avatar and Object now inherit from IEntity.
* Avatar.Position is now Avatar.WorldPosition to match IObject property.
* Implements event World.OnChat += delegate(IWorld sender, ChatEventArgs e);
* Optimizes SceneGraph - fetches on primitives via "GetGroupByPrim" wont search the entire list if the primitive is infact the root. (Core)
* Updates Test MRM.
* Notes: this requires heavy testing, it may cause new issues where LL have recycled agent block data for non-security purposes. It can be disabled on Line 4421 of LLClientView.cs by changing m_checkPackets to false.
* This represents approx 1/8th of the packets being checked.
* Now validating input to the Physics scene and warning when something is awry.
* This should help nail down that Non Finite Avatar Position Detected issue.
- Change several classes to use the new plugin for handling of region-modules
(NOTE: No regionmodule is using this yet)
- Add necessary prebuild parts (don't forget to runprebuild)
Attention: Work in progress. This shouldn't break anything, but you never know...
* It's not ready to try. It doesn't do what you expect in many ways.
* It throws errors and prints jibberish on the console
* Test it out only if you're brave and you've backed up first.
* The opensim.ini line is physics = BulletDotNETPlugin
Added safemode config to Standalone Hypergrid, similar effect to AllowRegionAccessToInventory in Inventory Server. Everyone should have these vars set to their default values except me!
* Adds IObject.WorldPosition and IObject.OffsetPosition - this is equivilent to AbsolutePosition and OffsetPosition in SOP respectively.
* Adds IObject.WorldRotation and IObject.OffsetRotation - as above.
log4net config file other then bin/OpenSim.exe.config
- moving ArgvConfigSource initialization up to allow for configuration
of XmlConfigurator.Configure()
This Fixes Mantis #3331
Also corrected the capping calculation in llApplyImpulse and several integer "boolean" comparisons from "== 1" to "!= 0" for true
* Renamed 'Material' to PhysicsMaterial (Wood, Glass, Metal, etc.). May want to place in subclass with other physics specific properties. (We however need to support these features in ODE/etc first.)
* Renamed Faces to Materials. IObjectFace to IObjectMaterial - this is for clarity for those coming from a 3D Programming background (it also makes more sense if/when we support Meshes in core). Properties and members remain identical.
* Added XMLDoc comments to IObject to assist people writing MRMs in XMLDoc aware editors.
* Changes World.Objects from Array IObject[] to IObjectAccessor.
* Syntactically identical in most behaviour, however the indexer is now ranges not from 0..Count, but any valid internal LocalID. Additional indexers have been added for UUID.
* Example: for(int i=0;i<World.Objects.Count;i++) will not work any more, however foreach(World.Objects) will remain functional.
* This prevents us needing to create a list for each access to World.Objects which should [in theory] present a dramatic speed improvement to MRM scripts frequently accessing World.Objects.
Permissions now use proper events and not delegate lists, which makes
for much easier reading and much less work adding new methods.
I finally found a way to raise events with return values without it becoming
late bound.
With some support from HomerH, this patch adds support for Wind
Model plugins via the mono.Addin framework.
* Adds console & OSSL access to Wind Parameters
* Adds plug-in support for custom wind models
* Provides two example Wind Model plug-ins
Documentation for the wind module is temporarily located at http://code.google.com/p/flotsam/wiki/CoreWindModule [^]
-- will move this documentation to http://opensimulator.org [^]
after the patch has been committed.
* This may help one of the symptoms or mantis 3363 , however it probably won't solve the occasional NonFinite Avatar Position detected.. issues that some people see. That is probably an entirely different issue(NaN).
* This may help people on certain 64 bit systems where the end byte position of each layer data packet is incorrect but the start positions are correct.
* The console will still be extremely chatty with 'Inconsistent packet data in JPEG2000 stream:' messages, however.. if OpenSimulator was able to recover the data, it will say HURISTICS SUCCEEDED
* Important: HttpServer.dll was changed to HttpServer_OpenSim.dll so that the HttpServer references do not conflict if you've copied the OpenMetaverse.Http.dll and requirements to the OpenSimulator bin folder.
This means that if you reference HttpServer.dll in any projects, you will need to change the reference to HttpServer_OpenSim.dll. It still uses the Same HttpServer namespace though.
* This is to help identify which test is freezing, since all the tests in the previous dll (coremodules) succeed
* Unfortunately they are not executed in the same order in which the results are listed in Bamboo
number map. Instead, write the map to a file for later use. That is not
yet used, so currently runtime errors after a sim restart will have wrong
line numbers
This functionality will be upstreamed later.
** Fixed call of new AvatarAppearance without arguments, which caused bots look like clouds of gas
** Added a SendAvatarData in ScenePresence.SetAppearance, which is expected after SetAppearance is run
** Fixed AssetXferUploader: CallbackID wasn't being passed on on multiple packets asset uploads
** Set VisualParams in AvatarAppearance to stop the alien looking bot from spawning and now looks a little better.
*** TODO: Set better VisualParams value then 150 to everything
* This change starts the script immediately after an object is loaded, rather than waiting till they are all loaded
* This should be okay, but please report any new errors
* These have actually been removed from HGHyperLink.TryUnlinkRegion, because some parameters were parsed but never used.
* This might be a situation where the warnings have shown up an oversight
Attached is a patch which enabled through an OpenSim.ini option the ability
to read long notecard lines. Currently although the data is read from
the notecard it is truncated at 255 characters (same as for the LL servers.
This patch allows the setting of that limit to a different value.
; Maximum length of notecard line read
; Increasing this to large values potentially opens
; up the system to malicious scripters
; NotecardLineReadCharsMax = 255
this allows for save/restore functionality using notecards without
having to worry about very short line length limits.
Attached is a patch which provides osAvatarPlayAnimation() the ability to
also trigger animations contained within the same prim as the script, as
llStartAnimation() does. (It also modifies osAvatarStopAnimation(),
otherwise the script wouldn't be able to stop animations it had started.)
* Sometimes, null is a valid return value to indicate 'none found'. doh.
* Sometimes, the Grid server does not send simURI - this you need to reconstruct yourself. Euw.
(I believe) this solves mantis issue #3287
- move a couple constants from InventoryArchiveConstants to
ArchiveConstants, now only one of these is needed
- change InventoryArchiveConstants references to ArchiveConstants
- remove InventoryArchive AssetInventoryServer plugin dependency on
OpenSim.Region.CodeModules
- trim trailing whitespace
* In most cases, the setting in OpenSim.ini.example is taken as the canonical one since this is the file virtually everyone ends up using
* OpenSim will start up with a blank OpenSim.ini, in which case sqlite is the default database (as before)
It's good practice to isolate unit tests so their outcome (pass/fail)
does not depend on whether another test has been run/passed/failed. A
method is used to populate the DB independently for each test, and a
TearDown method cleans up the database after each test.
Also adding extra comment in C-style comment test.
* A saved archive now immediately expires the data in the asset cache that it used, rather than retaining all the assets (esp textures) in the cache.
* This is an imperfect solution. Ideally we would only expire the assets newly requested for the archive (not existing ones). But doing that would require a lot more
restructuring.
* I don't believe there are any locking issues due to the locking performed by the underlying memory cache, but please report any issues.
* Add some comments to the Wind Module
* Add the BinBVH decoder/encoder as a scene object (to encode/decode animations programmatically).
* Add m_sitState for upcoming code to improve sit results.
Also truncate messages that may exceed the limit set by the packet size. The limit in OpenMetaverse is 1100 bytes including a zero byte terminator.
Fixes Mantis #3244
* I'm concerned however that the 'minimum fly height' should really be implemented in ScenePresence and not in the specific physics plugin so that all of the physics plugins can take advantage of it and if desired, a person could swap out the 'minimum fly height' functionality with other functionality.
* Uses 'mouselook' or left mouse button down, to determine when to use the camera's UP axis to determine the direction of movement.
* We crouch-slide no more.
llSetHoverHeight() should not clamp the x/y position of an object the way MoveTo does,
and it should recalculate the absolute height to hover at as an object moves to reflect
the current ground/water height under it.
Correctly implementing required adjusting the Physics interfaces and implementing at
the physics plug-in level. The attached is a patch that correctly implements
llSetHoverHeight() including updates to the ODE physics plug-in.
* This should make saving large OARs a somewhat better experience
* However, the problem where saving an archive pulls large numbers of assets into the asset cache isn't yet resolved
* This patch also removes lots of archive writing spam that crept in
Changes to IWindModule interface: Change from assuming a single array of
256 Vector2 values to a lookup function that takes region x, y, z and returns a Vector3
* Changed llWind() to use new lookup method of IWindModule
* Moved logic for determining the wind at a given point in the data array from
llWind() to the Wind Module itself.
* Renames MiniRegionModule to MRMModule to make it more distinct from the actual Mini Region Module[s] executed in Scene.
* Renames MiniRegionModuleBase to MRMBase for convenience. MRM's need to be adjusted to inherit from MRMBase.
Changed OpenSim.Framework.Communications.Tests.LoginServiceTests to use the LLStandaloneLoginService (from the LLStandaloneLoginModule) rather than LocalLoginService. Really these login tests should most likely be somewhere else as they are testing specific implementations of login services.
Commented out the old LocalLoginService as its no longer used, but want to check there are no problems before it gets deleted.
Added a flag (bool m_autoCreateLindenStack = true) which says if the ClientStack will be autocreated and initialised when creating regions. This helps with moving ClientStacks to Region modules.
Currently this flag is hardcoded to true, as it is only for testing at the moment, so you need to change the value in the code if you want to turn off auto creating.
The ini files in this folder are loaded after the masterini file (if that is set) and before opensim.ini.
The default folder it looks for and searches is "bin\config", but that can be set by using the command arg "-inidirectory=<path>" (path is local to bin\) when starting up opensim.exe.
isn't tested, but it doesn't break anything else. The reason for this
function is to let in world tools be used to coordiante out of world
network services that need access to client ip addresses.
Also in that plugin it registers the IUserService with all the Scenes (as they are created). So now we can start changing over all uses of IUserService, that currently access it from the CommunicationsManager to accessing it from the Scene.RequestModuleInterface call.
Once that is done we can move the UserService creation out to its own plugin and remove all references to it from the CommunicationsManager. Then we can take the next CommunicationsManager interface and repeat.
Added a ApplicationRegistry to OpenSimBase.
Changed LoadRegionsPlugin so it registers itself to that application registry.
Added a event to LoadRegionsPlugin, that is triggered when it creates a new scene ,although maybe this event should actually be in opensimBase incase other plugins are creating regions (like the RemoteAdminPlugin).
* This commit also does a first pass at creating folders for an inventory archive (previously everything was dumped in the same destiantion folder).
* This code might not work yet and nobody else should be using it yet anyway :)
- handle GetUser request for nonexistent user gracefully
- include throttle levels in ClientInfo
- code to save/restore throttles in client stack
- only update/send updates to active clients
- make animation classes serializable
Added OpenSim.Grid.UserServer.Modules project/dll which now contains the components of the userserver. With the OpenSim.Grid.UserServer being the setup and initiate exe.
If -background=true is specified on the command line, a null pointer exception crashes the server in OpenSim/Region/Application/OpenSimBase.cs in method StartupSpecific. Its trying to dereference m_console which is null, presumably because we're in background mode.
This patch fixes a bug where if a script in a child prim has taken
control of an avatar when they sit, although permission for camera
control is revoked when they stand, free camera control is not
restored. Currently it is only restored if the script is in the root
prim (though its not clear to me where this happens!).
* Moves Entity Updates into a seperate thread, allowing for OpenSim to utilize a computers CPU more effectively in return for potentially greater user and prim capacity.
* Removes an expensive Sqrt call performed during Update on each object. This should lower CPU requirements for high-prim regions with physics enabled.
* MXP Changes: Centers the region around 0,0 for primitives instead of 128,128. Prim display should now look more correct for MXP viewers.
Thanks mirceakitsune for a patch that prevents the avatar from trying to
walk along the Z-axis in mouselook mode (or left-click the avatar and
walk) while looking up or down.
Fixes Mantis #946.
* Added log4net dependency to physxplugin in prebuild.xml.
* Added missing m_log fields to classes.
* Replaced Console.WriteLine with appropriate m_log.Xxxx
* Tested that nant test target runs succesfully.
* Tested that local opensim sandbox starts up without errors.
* Added estate ban table to migration scripts of all supported databases.
* Added nhibernate mapping for EstateBans property of EstateSettings
* Refactored property accessors for EstateBan object.
* Added comments for EstateBan properties.
* Ensured that NHibernate tests pass with NUnitGUI.
* Ensured that nant test target passes.
This fixes mantis #3210. Thank you, tlaukkan!
The built-in default is OpenSim, unless a user server url is given, then
that is used, unless "shard" is also goven, then shard takes precedence.
The defult in OpenSim.ini is "OpenSim" for compatibility.
* Not locking causes enumeration exceptions as described in this matis
* part.TaskInventory needs to be locked for every access as it's a dictionary
* Extra locking will hopefully not cause any major issues - in places where the enumeration of the dictionary performs other lock or long running operations, the dictionary is
cloned instead
required file system objects are not present in the test harness.
This makes the main code ignore the failure, therefore the test succeeds.
Not elegant and maybe a unit test guru has a better way. Marked as a TODO
I've changed the extension point name, and the internal references that
used the same string. I also fixed up the messaging around the asset
loader so that it is more explicit.
The attached patch implements osGetDrawStringSize that looks like:
vector osGetDrawStringSize(string contentType, string text, string
fontName, int fontSize)
in LSL. It is meant to be used in conjunction with the osDraw*
functions. It returns accurate information on the size that a given
string will be rendered given the specified font and font size.
This allows for nicely formatted and positioned text on the generated
image.
Known issue: after TP, the self client doesn't see the animations going, but others can see them. So there's a bug there (TPs only, crossings seem to be all fine).
Untested: did not test animation overriders; only tested playing animations from the viewer.
* Also remove a duplicate write archive call in the unit test which might be causing test failures for people using mono 2.2 (though not 1.9.1, it would seem)
* Moves llEmail() delay to after e-mail send rather than before, in line with SL
* Thanks DoranZemlja
* Last build failure looks like a glitch, but one that has already happened twice recently which I need to look at
* This slightly extends a lock in WorldCommModule so that it covers the GetNewHandle method which states in it's doc that it assumes locking has happened before the method is
called
* Thanks DoranZemlja
and dropped nonths ago, because it is required to get smooth region
crossings with AO running. Without it, in some corner cases, anims will
continue to run in an unstoppable state.
The change makes two principal implementation changes:
[1] It removes the hard coded set of possible asset server client
implementations, allowing any arbitrary implementation that has been
identified to the PluginLoader as an appropriate extension. The
extension point for asset server client extension
is /OpenSim/AssetServerClient. All of the old configuration rules have
been preserved, and any of the legacy configuration values will still
work as they did before, except the implementation is now loaded as a
plug-in, rather than as a hard-coded instantiation of a specific class.
The re-hashing of IAssetServer as an extension of IPlugin made upgrading
of the implementation classes a necessity.
Caveat: I have not been able to meaningfully test the crypto-grid
clients. I believe they should work correctly, but the refactoring
necessary to handle plug-in based initialization (vs constructor-based
initialisation) admits the possibility of a problem.
[2] The asset cache implementation, previously introduce as a hard-code
class instantiation is now implemented as an IPlugin. Once again the
previous (configurationless) behavior has been preserved. But now it is
possible for those interested in experimenting with cache technologies
to do so simply by introducing a new extension for the asset cache
extension point (/OpenSim/AssetCache).
I've tested all of the configuration settings, after applying the patch
to a newly extracted tree, and they seem to work OK.
The following patch fixes http://opensimulator.org/mantis/view.php?id=3107
[^] by changing call from setRot to llSetRot, the later handles
child prim being rotated relative to root prim in linked set.
This changeset add the RegionReady module code. The module sends a
message on a configurable channel when an oar file has finished
loading or if the script engine has emptied its queue for the first
time (eg server startup). Config is something like this:
[RegionReady]
enabled = true
channel_notify = -800
The module also knows if there was an error with startup.
interface. Shared modules will now only get added once, so the command
handler is called once per module, not once per scene. Removal of scenes
has no adverse effects. Nonshared modules will be called for each scene.
In SL if llAbs() is called with the minimum integer value of -2147483648
it will return that value untouched without error.
this patch replicates the SL functionality.
OpenSim currently throws an overflow exception: number too small under
mono or a "System.OverflowException: Negating the minimum value of a
twos complement number is invalid. " under .NET
NOTHING has been deleted or moved off to forge at this point. what
has happened is that OpenSim.Region.Environment.Modules has been split
in two:
- OpenSim.Region.CoreModules: all those modules that are either
directly or indirectly referenced from other OpenSim packages, or
that provide functionality that the OpenSim developer community
considers core functionality:
CoreModules/Agent/AssetTransaction
CoreModules/Agent/Capabilities
CoreModules/Agent/TextureDownload
CoreModules/Agent/TextureSender
CoreModules/Agent/TextureSender/Tests
CoreModules/Agent/Xfer
CoreModules/Avatar/AvatarFactory
CoreModules/Avatar/Chat/ChatModule
CoreModules/Avatar/Combat
CoreModules/Avatar/Currency/SampleMoney
CoreModules/Avatar/Dialog
CoreModules/Avatar/Friends
CoreModules/Avatar/Gestures
CoreModules/Avatar/Groups
CoreModules/Avatar/InstantMessage
CoreModules/Avatar/Inventory
CoreModules/Avatar/Inventory/Archiver
CoreModules/Avatar/Inventory/Transfer
CoreModules/Avatar/Lure
CoreModules/Avatar/ObjectCaps
CoreModules/Avatar/Profiles
CoreModules/Communications/Local
CoreModules/Communications/REST
CoreModules/Framework/EventQueue
CoreModules/Framework/InterfaceCommander
CoreModules/Hypergrid
CoreModules/InterGrid
CoreModules/Scripting/DynamicTexture
CoreModules/Scripting/EMailModules
CoreModules/Scripting/HttpRequest
CoreModules/Scripting/LoadImageURL
CoreModules/Scripting/VectorRender
CoreModules/Scripting/WorldComm
CoreModules/Scripting/XMLRPC
CoreModules/World/Archiver
CoreModules/World/Archiver/Tests
CoreModules/World/Estate
CoreModules/World/Land
CoreModules/World/Permissions
CoreModules/World/Serialiser
CoreModules/World/Sound
CoreModules/World/Sun
CoreModules/World/Terrain
CoreModules/World/Terrain/DefaultEffects
CoreModules/World/Terrain/DefaultEffects/bin
CoreModules/World/Terrain/DefaultEffects/bin/Debug
CoreModules/World/Terrain/Effects
CoreModules/World/Terrain/FileLoaders
CoreModules/World/Terrain/FloodBrushes
CoreModules/World/Terrain/PaintBrushes
CoreModules/World/Terrain/Tests
CoreModules/World/Vegetation
CoreModules/World/Wind
CoreModules/World/WorldMap
- OpenSim.Region.OptionalModules: all those modules that are not core
modules:
OptionalModules/Avatar/Chat/IRC-stuff
OptionalModules/Avatar/Concierge
OptionalModules/Avatar/Voice/AsterixVoice
OptionalModules/Avatar/Voice/SIPVoice
OptionalModules/ContentManagementSystem
OptionalModules/Grid/Interregion
OptionalModules/Python
OptionalModules/SvnSerialiser
OptionalModules/World/NPC
OptionalModules/World/TreePopulator
The two most notable changes in the crossing process were:
* Object gets passed in only one message, not two as done before.
* Local object crossings do not get serialized, as done before.
These changes replace all direct references to the AssetCache with
IAssetCache. There is no change to functionality. Everything works as
before.
This is laying the groundwork for making it possible to register
alternative asset caching mechanisms without disrupting other parts of
OpenSim or their dependencies upon AssetCache functionality.
* This fixes crashing on large sets of physical prims because of stack collisions (assuming you follow the directions on linux for starting ode with ulimit). After the maximum joints are created, objects will start to fall through the ground and be disabled. Not the best solution, but it's better then a crash caused by a stack collision with the process exceeding the maximum available memory/recursions per thread.
* Make a clean region, make a stack of 5000 prim, 20 layers high. Make them physical, *SLOW*, but no crash.
* The actual AABB of the heightfield on the Z is now determined by the minimum and maximum heightfield value in the terrain array (assuming it's a reasonable number). This might optimize collisions in simulators that have a small difference between minimum and maximum heightfield values.
* Moved MoveAgentIntoRegion further down in the CompleteMovement method.
* changed a couple of methods from protected to public in SceneCommunicationService
Created nunit test for LSL API and example test for
llAngleBetween which was marked untested in wiki.
Run new test succesfully with NUnitGUI and nant build.
line editing, context sensitive help (press ? at any time), command line
history, a new plugin command system and new appender features thet let you
type while the console is scrolling. Seamlessly integrates the ICommander
interfaces.
* This adds more explanation for the new proxy settings in OpenSim.ini.example
* Also does some formatting correction
* I did some additional reformatting on top of that
* Such failures are now fatal to grab the user's attention.
* However, they could be made non-fatal (just with a loud error warning) if this proves too inconvenient
* So at the moment once can type 'help terrain fill' as well as 'terrain fill help'
* Current implementation is a transient hack that should be tidied up soon
OpenSim.Region.Environment into a "framework" part and a modules only
part. This first changeset refactors OpenSim.Region.Environment.Scenes,
OpenSim.Region.Environment.Interfaces, and OpenSim.Region.Interfaces
into OpenSim.Region.Framework.{Interfaces,Scenes} leaving only region
modules in OpenSim.Region.Environment.
The next step will be to move region modules up from
OpenSim.Region.Environment.Modules to OpenSim.Region.CoreModules and
then sort out which modules are really core modules and which should
move out to forge.
I've been very careful to NOT BREAK anything. i hope i've
succeeded. as this is the work of a whole week i hope i managed to
keep track with the applied patches of the last week --- could any of
you that did check in stuff have a look at whether it survived? thx!
* Typing help will now give a list of these topics at the top (as well as the rest of the current help stuff)
* Typing help <topic> will give information about commands specific to that topic
This patch fixes the problem where if an object containing a script is
deleted at the same time as an object containing the same script is
rezzed, it can result in the assembly file being deleted after the
second object script initialisation has found it but not started using
it yet, resulting in the script not starting up.
* This option makes OpenSim use the usual db based asset service in standalone, and the grid based one in grid mode
* The other options can (local, grid, etc) can still be used explicitly as before
* Also change OpenSim.ini.example and the surrounding explanative text
This is now called at the beginning of DeRezObject, so we know the database is upto date before we attempt to delete a object.
Fix Mantis #1004
Which happened because Database backups don't happen if a object is still selected, so when you select a part in a link-set and then unlink it and then delete it, all without unselecting the prim at all. The unlink changes never get updated to the database. So then when the call to delete the prim from the database happens, which is called with the SceneObjectId. That SceneObjectId is never found, as the database still has that prim as part of another link set.
It is possible that these changes might have to be reverted and for us to find a different method of fixing the problem. If the performance overhead is too high or it causes any other problems.
* This matches the existing OnOarFileLoaded event
* This brings up the question of how these things can be made generic so that they don't have to be tied into EventManager, but that's a topic for another day
previous commit message]
This patch reimplements the Draw method in the VectorRenderModule which is
used to create dynamic textures.
The previous version was limited to creating square dynamic textures, it
also didnt allow for dynamically loading an image containing transparency
except at 256x256.
The extraParams string in such functions as osSetDynamicTextureData can
now be passed a comma seperated string of name value pairs which set the
width,height and alpha value of dynamic textures. e.g.
"height:512,width:2048,alpha:255"
Backward compatibility is still preserved so passing the old params of
either a string integer "256" "512" will still work in the same fashion as
will passing "setAlpha" on its own
* Adding an option to use the tree module to manage the trees in the simulator (grow/reproduce/die)
* Setting it to off by default in an effort to reduce the number of threads in use by default
* You can also turn it on in a 'one off' way with 'tree active true' on the console. To 'one off' turn it off, it's 'tree active false'. The permanent way to do that, however is in the opensim.ini.
* If you log-in while the simulator is starting up, you won't be able to move and the sim stats will say 0 FPS, and 0 Physics Frames and you may see only terrain. Once the sim finishes starting up, it'll all resume as normal.
This changeset fixes a race condition where a script (XEngine run) can
startup before a reference is added to it in all of the required
places in the XEngine class. The effect of this is that a script can
sometimes on startup miss script events. For example a script which
starts up and initialises itself from a notecard may never receive the
dataserver event containing the notecard information.
The patch isn't as clean as I'd like - I've split the constructor of
ScriptInstance up so it does everything it did before except
call Startup and post events like state_entry and on_rez. An Init
function has been added which is called after the ScriptInstance
object has been added to the necessary data structures in XEngine.
Happy to rework it if someone suggests a better way of doing it.
Adding Oarfileloaded and EmptyScriptCompileQueue event support which
allows (with a module) for programmatic notification of when a region
objects and scripts are up and running after a server start or
load-oar.
* Console command help should be output to the console, not to the log (as "help" does it already). That allows getting help/answers even if you only log into a file.
Fixes Mantis#2916.
* Increases the number of ImageDataPackets we send per PriorityQueue pop and tweak it so that the number of packets is ( (2 * decode level) + 1 ) * 2, and (((2 * (5-decode level)) + 1) * 2). The first one sends more data for low quality textures, the second one sends more data for high quality textures.
* No functional change
* It strikes me that there may be caps problems if double registration is presented if cleanup failed for a previous agent (so a caps handler will remain in memory for that
agent but with a different seed). This needs investigation
this patch makes load-oar a bit more tolerant to irrelevant
differences in the oar file format. Directory entries are now ignored
rather than trying to interpret them as files they hold which results
in the load-oar failing. This change makes it easier to manually
modify oar files.
* Often times now the only reason OpenJpeg doesn't work is because it requires Glibc 2.4 The error messages reflect that.
* In J2kDecoder module, It stops trying to decode modules if it encounters a dllnotfound exception and instead sends a full resolution layer that causes the texture sender to only send the full resolution image. (big decrease in texture download speed, but it's better then nasty repeating error messages)
sceneB
* However, I'm not convinced that the actual process in the test completely reflects reality, and a lot of stuff had to be rigged up (which should get resolved over time)
Console Command: link-mapping <StartXloc> <StartYloc>.
This results in only the last two digits of any of the locations defined in the xml file being used, and those 0-99 values being added to the StartXloc and StartYloc figures.
For more infomation, see the email on the mailing list and the soon to be added instructions on the wiki's hypergrid page.
Also made the Secion Names in xml file case insensitive.
* Added an additional lock in GetScenePresences()
* Changed ForEachClient to use GetScenePresences() instead of the main ScenePresences dictionary, so that there is no need to lock.
TODO: Add a initial startup option for setting the uri and making it autoload it.
Add support for scanning the xml file to check that its own region(s) aren't in the list, and if they are, ignoring them. This would allow setting up "virtual link/grid lists" on webservers, that people can add their own regions to and also point those regions to those same lists, so they load the other region's data.
Add support for automapping of those region/link lists.
* There appears to be a bug on mono 1.9.1 (and maybe later), where sometimes the async wait will be signalled even though that async callback has not executed
* This change may make it slightly better but it's difficult to tell (it definitely still occurs)
* Also this patch closes the wait handle explicitly, as recommended in the MSDN docs. This doesn't have any impact on the bug though
* The position will be reset if the connection turns out to be a root one, and since the client is already authenticated another authentication will not occur anyway
* In fact, it would only be activated if there was no [Communications] section at all (which would be the case for most people with existing region setups unless they
specifically added it in
* This fix means that enabling inter-region rest comms is now the default
The attached patch adds the automatic granting of PERMISSION_CONTROL_CAMERA
on request to a script when an avatar is sitting on the object that the script
resides in.
It also automatically removes PERMISSION_TAKE_CONTROLS and PERMISSION_CONTROL_CAMERA
from the scripts when an avatar stands up. It doesn't remove the other automatically
granted permissions but this follows LL server behavior. Removing these two
permissions avoids some potential race conditions (accidentally taking
control after the avatar has stood up) which may be why LL put
them in the first place.
Fixed all NHibernate unit tests by implementing missing persistency
methods, tables, columns and fixing bugs in the existing implementation.
Two minor changes to classes outside NHibernate module: Added Scene
instantiation for SceneObjectGroup in OpenSim.Data.Tests.BasicRegionTest
as this was required by the NHibernate persistency. In the process added
also mock constructor to Scene which only populates RegionInfo in the scene
which is used by ScenePart.RegionUUID. NHibernate module is still in
experimental state and has not been tested at opensim region or ugaim runtime
configuration. Adding unit tests to build is not yet advisable nor using
NHibernate module in any production setup.
* Must compare user region with previous calling of the same method to determine if we should look the user up.
* The first time, the previous region handle will be 0, so therefore it'll use the cached version.
* The second time, previousRegionHandle will be what we looked up on the first round
* The third time, it'll say.. we tried to deliver it once with or without lookup, we tried to deliver it a second time with lookup, and failed both times so end!
** Update limiter logic reversed
* Resolve a few bugs in the Viewer Stats collector
** Catch the logoff ViewerStats post (darn those caps deregistrations)
** Check the type of the correct map when checking
* Resolve a leak in Prototype_distributor
** .Close, .Dispose()!
* It still isn't quite ready to be used mainstream.
* A couple of things to note, it doesn't keep track of the logs if nobody is looking at the stats.
* It doesn't read the whole log file. Just the last 10 lines of the stream. Tested to 1GB+ logfiles with no noticeable performance issues.
* Adds a prototype web stats module which is disabled by default. It's functional with one report right now, however, the database structure may change, so I don't recommend enabling this to keep actual stats right now. I'll let you know when it's safe.
* Adds Prototype for ajaxy web content
* removed a warning or two.
That improves performance drastically, at least for Mono, as the
(de-)serializers can then be optimized (and won't use reflection anymore).
On my system, before this change de-/serialization took ~9s/9s, whereas
after the change it takes ~.5/.2s.
Committing the above patch with changes. Thank you, Gerhard. Removed
unneccessary verbosity in case of no error, added more output for errors.
Added support for HTTP Uri fetching which I had added in my last commit.
Converted the Console.WriteLines to log4net.
- Introducing synchronous Teleports. Now the receiving region calls back the sending region after the client has been made a root agent there, that is, after client sends CompleteMovement to the destination.
- SendCloseAgent moved from OGS1 Remoting to RESTComms.
- Adapted code to match the corrected signatures
- Fixes Mantis #2934. Hopefully.
Note: Physics on linked objects still don't work correctly:
It doesn't crash the region anymore, but the example object in
the mentioned mantis now falls through the ground.
* This means that you can now save terrain textures, water height, etc.
* Estate settings are not supported
* Older OAR files without these settings can still be loaded
* This allows configuration of the assetset and library control file paths to be other than ./inventory/Libraries.xml and ./assets/AssetSets.xml
* This is controlled via the LibrariesXMLFile and AssetSetsXMLFile configuration settings in [StandAlone] in OpenSim.ini (in standalone)
and via the user and asset config xml files for grid mode
* Thanks to SirKimba for the patch
* Not out of the weeds yet. Rotating the root part with 'edit linked parts' still borks all child part rotations and rotating a selection of child parts with 'edit linked parts' still borks the position/rotation of the child parts selected
On a call of llVolumeDetect(1) (or any other number !=0) volume
detection is enabled. Together with VD, the phantom flag is set to the GUI.
On a call of llVolumeDetect(0), vd detection is switched of again,
also the phantom state is removed. On a call to llSetState(STATE_PHANTOM,
false) while VD is active, also VD is switched off. The same is true for
unchecking the phantom flag via GUI. This allows to take back VD without
the need to script just by removing the phantom flag.
Things missing in this patch: persistance of the volume-detection flag.
This needs more discussion and will be included in another patch soon.
missing from when this is run in grid mode. But, worse, turns out that this whole feature of kicking
off the user after a crash was entirely missing from standalone mode.
* This facility allows you to save changes to an object that you've rezzed into a region back into their original inventory item without having to take a copy of the rezzed
object.
aggressive at nixing the user out of the cache. We're now relying on NeedSceneCacheClear to decide
whether to nix it or not. All other mods in other files are for better debugging messages.
* This pushes an identifier for the OpenSim scene to the physics scene. This allows log messages from the physics scene to identify which OpenSim scene they relate to.
* Thanks Gerhard
We need to update all child agents whenever the root agent crosses regions. The update
now includes child agents in common neighbours. This is so that those get updated with the
seeds of the new child agents that are spawned from the receiving region.
This also fixes some timing issues. We need to close child agents from the originating region
before we update child agents in the receiving region.
It is not documented to be applied to llGetTimeOfDay at all and is not appropriate for the other two because dilation does not directly affect script speed in OS.
Fixes Mantis #2783 and #2162
some of our DNS/dotted decimal issues that plague teleporting.
We are increasing the MajorInterfaceVersion from 1 to 2 in the
gridserver, so sims will need to be updated on various grids.
the SimpleModule doesn't prevent the start of the region server, if used.
Note: This module doesn't do anything sensible; it only serves as an example.
* This means that UserProfileCacheService no longer needs to know about IClientAPI and can leave it to callers to do their own error logging
* This is also more consistent with the way that item inventory manipulation is handled
* I don't really think Scene.PacketHandlers.cs should be a permanent home for these handlers - this is just for convenience
* This means that LL RAW terrains (one source being the upload/download buttons on the estate dialog in the viewer) are now imported the 'right' way around rather than being
flipped on the y axis
* Existing RAW terrains before this patch will need to be flipped with the "terrain flip y" command from the console after import
* More details on the mailing lists soon.
-This line, and those below, will be ignored--
M OpenSim/Region/Environment/Modules/World/Terrain/FileLoaders/LLRAW.cs
M OpenSim/Framework/Communications/Cache/CachedUserInfo.cs
M OpenSim/Framework/Communications/Cache/UserProfileCacheService.cs
* This allows terrain to be flipped on the x or y axis with the command "terrain flip x" (or y)
* See terrain help from the command prompt
* This is in anticipation of change the way around in which terrain raw files are imported to match that of Second Life (to reduce user confusion and improve useability)
* Thanks jonc!
The attached patch ads an OpenSim.ini option (AutomaticLinkPermission)
which when enabled makes PERMISSION_CHANGE_LINKS to be granted to
scripts by default. When enabled llGetPermissions will always return it
as granted and llCreateLink//llBreakLink will succeed without
doing llRequestPermissions. ONLY ENABLE THIS IN TRUSTED ENVIRONMENTS.
The patch also fixes a minor bug in llCreateLink
related to a potential dereference of a null client object.
* Warning! Physics API change. This means that the NBodySimulation needs to be updated!
* PhysicsActor -> void SetVolumeDetect(int) needs to go into classes that use PhysicsActor as their base class.
* Disabled by default (see OpenSim.ini.example for how to enable)
* Saves exceptions to a folder on disk (default "crashes") when enabled.
* These reports can then be uploaded or posted to help debug an error.
* Fixed a bug that caused physics proxies to be scattered when you link an object.
* Single physical prim work exactly the same as before, just linked physical prim will have changed.
For a script to successfully cross, both source and destination region must
enable the feature. WARNING: Trusting binaries from other sims allows
ARBITRARY REMOTE CODE EXECUTION for ANYONE! Please do not use except
in ultimate trust scenarios!
* Initializes ODE only when a scene is grabbed rather than on plugin load. This means we don't initialize ode if that physics engine is not used, and it allows other ode use
plugins to be used instead.
* However, this is not currently activated since it's not absolutely fully tested and there's a bug lurking in there to do with the sending of the BulkInventoryUpdate packets
The attached patch fixes the bug where llCreateLink will cause a null dereference
if the avatar who gave link permission for the object is no longer logged in.
to fix the A -> B -> A issue. Because the code is very reuse friendly,
it shouldn't cause much additional memory bloat in the sims.
I'm sure this isn't the right fix, but it is progress, and confirms that
the close path while leaving a region is part of the issue.
before" issue. This involves the EventQueueGet module so I'm not sure
this is the right fix but it appears to work.
When the ScenePresence leaves force a close on it. The reuse of the
queues doesn't seem to be working, so the forced close causes a new one
the next time in. This does appear to work.
Use height calculation in Basic Physics and Physics of Simplicity so that avatars larger than the default walk with straight legs and shorter walk on the ground.
prevent adjacent sims from using identical Local IDs for the attachment
Thanks to Mana Janus (Hippo Viewer) for providing the crucial bit of
information, namely that, due to a bug in the viewer, adjacent sims can't
use the same local ids.
Migrate sculpt meshing code to primMesher version. This should result in more accurate physical sculpted prim proxies.
Remove much obsolete code from Region/Physics/Meshing
Adds a new method to IClientAPI to allow adding message handlers
for GenericMessages (of which "autopilot" is one). Part 2 adds a
specific autopilot handler in ScenePresence.cs.
2) Removing unused variables and functions.
3) Simplifying the navigation logic in ScenePresence.cs. The
original patch was somewhat complex because it included orientation
logic for a future enhancement of orienting the avatar to point
towards the direction being walked. Currently this isn't working,
though, so I removed the orientation code, which leaves just the
smaller and hopefully simpler-to-understand movement code.
mechanism to prevent prim loss. Preserve link order on sim restart
and drag copy. Fix drag-copied prims' inventories. Fix persistence
of child prim inventories.
* This is Diva's hypergrid patch, as perviously discussed on the opensim-dev mailing list
* Applied some minor prebuild.xml jiggling to resolve a dependency issue
* Thanks Diva!
* Entities should now in theory be lock-free externally.
* Other properties may cause blocking however[?].
* ScenePresence maintains separate locks so isn't fixed by this commit.
* Important Changes: Scene.Entities is now IEnumerable directly. You do not need to use Entities.Values, you can Enumerate on .Entities directly. (So 'foreach Scene.Entities' vs 'foreach Scene.Entities.Values').
* Locks: Entities maintains it's own internal locking states. This means you do not need to lock entities anymore. I'll be going through and removing locks on it systematically.
* Not referenced anywhere yet.
* Designed to replace Dictionary<UUID,EntityBase> within SceneGraph.Entities
* Allows indexed access and better handling of locks.
* Someone needs to implement IEnumerable on this before we can switch it over.
from the presence module to the message server, through the user server
and on into the database. This should fix the "Already logged in" issue
that grids see after a sim crashes, or a user crashes out of a sim.
Not yet a 100% solution for friends, but getting there.
This patch adds the option of adding the email when you create
a new user. This works in Gridmode as none Gridmode. This
option is also added to RemoteAdminPlugin. With a new handler
you can create a user with a email.
- I first did lock on m_scenes for that access, but as m_scenes (currently) is
only changed in Initialise (and locked there), and PostInitialise is called
after the calls to Initialise, there shouldn't be a need to lock here.
- Trying out git-svn dcommit. Hope this works.
* SceneObjectPartInventory.cs isn't a particularly good name but it's probably not got a long life
* A proper inventory interface to follow
* Parallel changes for other inventory partial classes to follow at a later date
* Deleting the object before the async to inventory is bad since if OpenSim crashes (or is shutdown!) in the time inbetween, then the object is lost
* Also now delete the object from the scene only after it has gone to inventory - in the old way there was still a race condition
* This is still not ideal since the 'deleted' object remains in the scene and probably could be manipulated. But this is better than the alternative
* Last time I accidentally left the existing code in place that merely wiped the group from the client (but not the scene)
* Put in an incomplete 'delete object o inventory' test to assert that this happens
* This is primarily to see if not loading ODE in the unit tests will allow them to proceed, though the option of separate loading is probably a good thing in itself
Attached is a patch which fixes mantis bug 2645 where scripts in child
prims
freeze after being unlinked from the root prim. This got introduced
some time after the 0.6 freeze. Have checked test suites
pass even after nant clean ;-)
http://opensimulator.org/mantis/bug_update_page.php?bug_id=2645
Also removes a redundant ResetIDs call
Regards,
Chris
release a direct descendant of the stable 7364, with all the features and
none of the issues.
This omits the following patch chain: r7383 r7382 r7381 r7377 r7375 r7373
r7372 r7370 r7369 r7368 r7367 r7366
- The initial priority wasn't passed in, so a discardLevel of -1 in the first
request automatically cancelled texture-download.
- One-packet textures weren't handled, as CurrentPacket starts at 1 and
StopPacket == 0 in that case.
- Increased MAX_ALLOWED_TEXTURE_REQUESTS to 15; incremental texture downloads
will cause more requests.
* This allows one to override normal OpenSim permissions and prevent non-gods from editing any scripts. This allows edit ability
to be rescinded after it has been given, and prevents the security hole where a single script with liberal perms would allow code changes.
* The default setting remains the existing one of never overruling normal edit permissions.
* These two settings may be enough to stop non-gods entering artbirary script code in a closed grid/standalone configuration.
* This renders RootPart == null checks useless - the replacement is to check SOG.IsDeleted. However, in many cases this will not be necessary since updates to deleted parts
will not be sent to the client
* This should remove any remaining race conditions where an object is deleted while another thread is yet to obtain the root part to perform some operation
* Doing this is probably a necessary prerequisite to moving to a model without a separate SOG and SOP
* Unfortunately it's not possible to eliminate all RootPart == null checks since in some contexts it is currently used to check whether an object was created successfully
causes defautlt behavior. "true" now means usable unconditionally, "false"
means disabled, and a list of UUIDs restricts it.
This changes SECURITY! If you used "true" here before, you shoudl review
your setup!
the IM module and makes it into a module of it's own, which can be used by
all other modules. Removes some ugly hacks. Refer to the IM module to see
how it's used. Also fixes the persistence issue (Mantis #2598)
Objects will be persisted now MinimumTimeBeforePersistenceConsidered seconds
after the last change, but latest MaximumTimeBeforePersistenceConsidered after
the first change (both are configurable in OpenSim.ini.example and are set to
60 and 600 as default).
Contains a migration. May contain nuts.
Please back up your inventory data store. This revision changes the interface
version!! No older regions can connect to these new UGAIM, and the new regions
can't connect to the old UGAIM. Fixes a long-standing issue of permissions loss
Currently persisted on MySQL only.
* This doesn't allow complete script lockdown of a sim, many avenues (copying, editing) are still uncloseable at the moment
* Default remains to allow all users to create scripts (subject to existing permissions if enabled)
* Switch default for serverside_object_permissions from false to true - it seems more natural that we enforce permissions by default rather than not!
* Add some explanation in OpenSim.ini.example for serverside_object_permissions
This patch makes llAllowInventoryDrop work with the permissions module
enabled. Changes include:
- Enabled PropagatePermissions when permissions module serverside perms
is on
- change ownership of item when item is dropped into an object.
Ownership changes to the owner of the object the item is dropped into
- propagation of permissions if the permissions module enabled (eg
next-owner mask applied)
- CHANGED_ALLOWED_DROP is now passed to the change script event if an
item was allowed to be dropped into the object only because
llAllowInventoryDrop is enabled (instead of CHANGED_INVENTORY being
passed).
- Sets object flags correctly when llAllowInventoryDrop is called so
clients are notified immediately of the change in state. Am not
sure that calling aggregateScriptEvents is the right way to do it,
but it works and seems to be the only way without making further
changes to update LocalFlags
physical center of an avatar, for display purposes. This should keep the
avatar feet above ground visually in most cases. Tweaked for both height
extremes and various leg lengths. Improvements welcome
public bool ExternalChecksCanCreateAvatarInventory(int invType, UUID userID)
public bool ExternalChecksCanCopyAvatarInventory(UUID itemID, UUID userID)
public bool ExternalChecksCanCopyAvatarInventory(UUID itemID, UUID userID)
public bool ExternalChecksCanDeleteAvatarInventory(UUID itemID, UUID userID)
to ExternalChecks to handle avatar inventory checks (as opposed to object inv checks).
* opensim-dev e-mail to follow concerning this shortly
* This may alleviate a little the freezing experienced by existing avatars when a new client logs in
* Race condition risks look minimal since one wouldn't expect another thread to start fiddling with that presence
destination user is offline/out of range. No more eternal cache is needed
for tracking IDs. Code cleanup. Removed some casts from IScene to Scene.
Decline now properly places item in trash rather than deleting it outright.
* Code Cleanliness Fixes in LLClientView
* Using field instead of local variable for handlerUpdatePrimGroupRotation (if you notice any new oddities with prim group rotation after this patch, please mantis)
* This is done by sending a 'major interface version' number on sim registration. Developers must increment this every time they make a change that would make the previous
OpenSim revision failure incompatible with the new one (non-fatal incompatibilities are fine).
* This number resides in OpenSim.Framework.Servers.VersionInfo.MajorInterfaceVersion
* This allows the grid service to stop older, incompatible regions from connecting
This patch changes a couple of methods in Scene.Inventory to virtual,
so they can be overridden in subclasses. DeleteToInventory now returns
the UUID of the newly created asset, so that further actions on it can
be pursued in subclasses. This will make my life easier for making
inventory accessible in the hypergrid.
will allow people who don't want megaprims in their sim to prevent them
from being created. Any prim rezzed or pulled across the border will be
clamped to the size specified in OpenSim.ini if this option is set.
* Implements the SendInitiateDownload method in IClientAPI
* Uses the ITerrainModule Interface to write a terrain file to disk then uses a FileStream to read the binary file from the disk and put it in a byte array. and save to the xFer list.
* It then tells the client to download the file and the client initiates an Xfer request.
* Decouple sog and sop by removing the need to pass the sog to the sop when it is created - most of the code was doing this operation (and hence duplicating it) anyway
* Remove unused constructors
This patch addresses mantis bug 2576.
http://opensimulator.org/mantis/view.php?id=2576
Briefly, if you call llDie from many scripts at the same time (say a
build is cleaning up excess objects) then OpenSim deadlocks. Avatars
are unable to move, and whilst the console is active you can't do much
without it also locking up. This only occurs with the XEngine script
engine enabled.
I have attached a patch which works, but I'm not sure its the right way
to address the problem. The fundamental problem is that a lock on a
SceneObjectGroup's m_parts is taken when the object is deleted, a
callback to the script engine occurs and a fair way down the callchain,
potentially there are locks taken on several other SceneObjectGroup's
m_parts. Deadlock then occurs if you get unlucky enough
to get in the situation where with several llDie's are called and
SceneObjectGroups
have taken a lock on their own m_parts, and end up waiting on each
other's
locks to become available.
The patch adds a lock at a high level so that that the removal of script
instances
from an object only occurs once per scene at a time. This avoids the
potential
of deadlock. Theoretically there could be some performance hit but
AFAICT
the path taken is not a common occurrence.
Would welcome any suggestions for a better solution, otherwise feel free
to apply :-)
Note this patch was built against the 0.6.0 freeze as trunk was
rather broken for me this morning (creating a script killed the client
connection).
* Items will now be locally cached for only 24 hours from last access. (Rather than until restart)
* Caveat: Implementing the new caching mechanism means statistics gathering on AssetCache is no longer functional. (Justin - you might want to take a look and see if you can somehow get that back and running if you still need it)
PRIM_TEMP_ON_REZ and PRIM_MATERIAL are not implemented in
llSetPrimitiveParams so support for these is in the patch.
Also two deprecated functions throw errors. They are changed
to behave as in SL: llSetPrimURL - Does nothing except the sleep
(currently commented out) & llRefreshPrimURL shouts
"llRefreshPrimURL - not yet supported" on the error channel
* This means the recent memory fix should now be working correctly - so the current largest memory leak should be fixed. AssetCache still needs to be addressed however.
* This patch is highly experimental and may cause clients to not be able to connect, if this is the case, it will be rolled back in approximately 5 minutes.
* Introducing IClientCore - this will be the key replacement for IClientAPI in the long run, it has a very minimal set of methods designed to allow you to access specialist API's.
* See https://lists.berlios.de/pipermail/opensim-dev/2008-September/003049.html for the early discussion on this.
* Was caused by the lack of a local id. Local ids are now given from the same sequence as prims, rather than a separate one
* I don't believe this will cause any problems, but please revert to a separate sequence if it does
Included patch fixes error: Z and W terms in the quaternion were
swapped (ZERO_ROTATION is <0,0,0,1>, it was checking for <0,0,1,0>).
There is an issue with older prims: it seems their default sit
target was not always set to ZERO_ROTATION;
potential (asset bloat, asset server DOS due to no enforced delay)
Formatting cleanup. Change default permissions on the notecard to
not include "anyone can copy" and "anyone can move", as they are
meaningless on non-prim items.
- fixes IRCBridgeModule's XmlRpc method really paying attention to
region parameter
- cleans up indentation in IRCBridge code
- fixes ConciergeModule exception on client logout
* This should (hopefully) allow TestClient and stuff built on top of it to work again
* Will probably come back later and change variable names to stop this happening again
* Really this should be 1, but I think that this would be too slow compared to a Second Life server until we improve our ability to send textures of variable quality
* This may improve one aspect of sim performance where there are many avatars. However, there are still other performance problems that are unrelated to this change
* Value may be further tuned
* Removed temporary decals since the multipler setting will stick around now
* Inserts proper animation state names into data/avataranimations.xml file so that llGetAnimation() works as one would expect.
* Thanks StrawberryFride!
* This is a partial implementation of llGetAnimation that returns the name of the animation as stored in data/avataranimations.xml but not its state name (since we don't yet
have these).
* Thanks StrawberryFride
This patch allows an object to directly message another object given its
key using the method osMessageObject(key objectUUID,string message).
to communicate with an object it must implement the dataserver listener
method. The dataserver method is passed the key of the calling object
and a string message.
* Now I've had time to analyze this, the Second Life grid doesn't appear to send this to the client (which in principle should know what wearables it has already)
So basically simplifies what a lsl script that detects a avatar sitting on a prim, then stopping the sit animation and playing a custom animation, does.
Also added another ScenePresence.HandleAgentRequestSit() method , that accepts the name of the sit animation. So that modules can override the animation used, when they are doing a server controlled sit.
Started some work on making the stand pose be played as soon as a user logs into a region. Rather than them starting with their arms stretched. This still needs more work
* This should probably be 1, but currently by default it is 8, to reflect what was being eon3 in OpenSim before this revision. So if the client requested a maximum throttle
of 1500 kilobits per second, we would actually send out 1500 kilobytes per second
* Adjusting this multiplier down towards 1 may improve your OpenSim experience, though in other situations it may degrade (e.g. if you're using a standalone over high bandwidth
links)
* This is currently a user setting because adjusting it down may currently reveal other OpenSim bugs.
* The attached patch changes the LSLInteger operator overrides for == and != to return LSLIntegers 1 or 0 instead of a bool and adds similar operator overrides for >, <, >= and
<=
* Thanks idb!
on-/offline updates, calling cards for friends.
This adds methods in the DB layer and changes the MessagingServer, so a full
update (incl. UGAIM) is necessary to get it working. Older regions shouldn't
break, nor should older UGAIM break newer regions, but friends/presence will
only work with all concerned parts (UGAIM, source region and destination
region) at this revision (or later).
I added the DB code for MSSQL, too, but couldn't test that.
BEWARE: May contain bugs.
* This is to avoid repetitive null checks - I'm beginning to think that blasting away the root part on object deletion is actually a bad move. Perhaps we should leave it around
and let the client ignore any superfluous packets (which it may well do anyway), since we're constantly exposing a race condition
with the old Grid Instant Message over OGS1. Refactor the EventManager
to be independent of the rigid module structure design imposed by
the current implementation. Message routing is now done in the destination
module rather than in the event manager. This way, more or less granular
solutions are possible without core changes.
* This may help us detect if mysterious UDP disconnects are happening because of this.
* Shouldn't be any functional change but I would appreciate a buddy check from Teravus if he has time (as for all client stack changes)
* Since the client sets the appearance in the first place (in response to receiving wearables information originally) this seems a little redundant
* But I've realized I didn't actually test this assumption, so I'm going to reinstate it for now
* This is to make things less confusing to users (and maybe a little easier to do future version filtering for grid connections)
* If there's disagreement about this, then please say so (preferably in the opensim-dev mailing list)
* Whilst this does automatically get the client to rebake, on crossing a region border the 'local' assets are left behind
* There may be a cunning solution (such as squirting the assets on region crossing, or having them fetched from the original region) but
instead I'm going to opt for the easy solution of keeping them in the asset database, for now
* And hopefully rebaking all the time should no longer be necessary now
* It turns out that when the client baked the texture, the uploaded asset had the Temporary flag to true (Temporary is actually deprecated).
* It also had the StoreLocal flag set to true, which signifies that the asset should be stored locally. If it disappears we should reply to the asset request with
ImageNotInDatabasePacket
* However, last time this was enabled some clients started crashing. This may well no longer be the case and needs to be tested, but in the mean time we will store
the asset instead.
* This needs to be resolved in a better way, possibly by starting to send the ImageNotInDatabase packet again instead
When the avatar stops flying in mid air, it stays still
while it falls to the ground. Normally, the avatar would
use the falling animation (when it spins its arms and legs
while falling) until it reaches the ground.
The attached patch fixes the bug where when linking in a new set of
prims to an already linked set of objects the prims were placed at the
end of the list rather than just after the root prim. ie. link prim
order result was different on OpenSim compared to an LL server. This
causes a few issues with respect to compatibility of scripts,
especially when using llCreateLink.
* If a change is going to affect a unit test, then please could we change the unit test at the same time? Otherwise this will never get done
* It also seems a bad idea to disable tests which start failing unless there's a very good reason
The following should silently fail when attached,
llBreakAllLinks, llBreakLink, llCreateLink.
The following should be restricted to avatars in the same sim,
llGetAgentInfo, llSameGroup, llGetAgentSize, llGetAnimationList.
Comment added to the following unimplemented functions,
llGetAgentLanguage, llGetAnimation.
llGetBoundingBox is completely unimplemented. The attached
patch contains a partial implementation for single prim
objects and standing, flying and ground sitting avatars.
In the case of an avatar sat on an object or multi-prim
objects only the bounding box of the root prim is returned.
llRezObject and llRezAtRoot ignored any velocity parameter.
The attached patch makes use of it if it is not a zero vector.
Support of strength-slider in latest sl-client (1.21.6)
Added a patch, which includes the support of strength-slider
in latest sl-client (1.21.6) for Raise- and LowerSphere.
* This moves authentication from the client thread (where failure was difficult to detect) to the particular thread handling that packet
* I've kept the authentication outside of the crucial clientCircuits lock (though any delay here is probably swamped by the other delays associated with login)
* Also added more to the unit test to ensure this doesn't regress
* This checks that a client circuit is established when the udp server is given a use client circuit code packet
* And checks that other circuit codes do not exist
* Not sure why things still worked in the presence of this bug - possibly the problem is compensated for later on. If you are having udp session problems this bug fix may help
(though no guarantees).
logic to ConciergeModule. adding XmlRpc method to allow update of
welcome template for a region. documenting [Concierge] "password"
option in OpenSim.ini.example.
Attached patch protects against a NullReferenceException
in SceneObjectGroup.stopMoveToTarget (used by LSL function
llStopMoveToTarget). Thanks to M. Igarashi.
welcomes = /path/to/welcome/files/directory
a directory in which you can place welcome templates for concierged
regions (those regions that match the "regions" regexp). you can use
format substitution:
0: will be replaced by avatar name of the avatar entering the region
1: will be replaced by region name
2: will be replaced the name of the concierge
MemberwiseClone() also clones the "already backed up" flag, preventing prims
created by drag-copying from being persisted. If such a prim is made the root
prims of a link set, the entire set will not be persisted. Fixed now.
chat relaying via private channels, and old IRCBridgeModule
behaviour. also cleaning up IRCBridgeModule's OpenSim.ini
configuration variable names (still supporting "old" variable
names). refactored IRCChatModule into IRCConnector and incorporating
watchdog from IRCBridgeModule into IRCConnector.
enabling ChatModule to be used as a super-class and utilizing it in
ConciergeModule.
* Guys, there's an endless loop there *ON PURPOSE*. Please don't try to *fix* it. We must continue to process the UDP stream buffer on clients that disconnected nastily until it ends or the UDP server accept thread will die a horrible death.
* Unix epoch starts at midnight, not at 8:00am
* All date/time handling should be done in UTC in the server, not in
the local timezone.
* Refactor out repeated computation of a constant value
- Added setting of CreationTime to some places where inventoryitems
are created
This fixes Mantis#2390.
Add rezzing time to objects. Add Object return and traffic fields to land
database. Add plumbing for auto return. Implement auto return.
Contains a migration. May contain nuts.
* Implement the linear impulse portion of llPushObject. We should have a lsl compatible implementation of that portion of the push. Angular.. well. still have yet to implement a torque accumulator.
* llPushObject respects the region and parcel settings for Restrict Push, it also respects GodMode as is defined in the LSL spec.
* Regarding an earlier change, I think it would be possible to eliminate the creation of new IPEndPoints on every end receive if we did the client circuit lookup before starting
the next receive. However, this would be a performance trade off and hence not worth trying without performance testing
* This widened what I think is an existing race condition where asynchronous recieves could potentially stomp on each other's end points (though this must occur very rarely, if at
all, in reality)
* This allows multiple user profile providers to be specified in OpenSim.ini separated by commas
* If multiple providers are specified then a request for a user profile will query each in turn until the profile is either found or all have been queried
* Unfortunately I don't believe this order can currently be specified, which if true is something that will need to be fixed.
* Thanks to smeans for the original patch.
* Additionally, you can probably get more physical prim now together.. though, I think this puts us back on par with where we were in the beginning of the year on number of physical objects. Experiment. Make videos. Send Feedback. Enjoy.
- Added two missing caps (UpdateScriptAgent, UpdateScriptTask)
- Added one missing EventQueue event (ScriptRunningReply)
- Changed DNE and XEngine to use this new event
As we only use the mono engine anyway, the "Mono" checkbox is set by
default but doesn't have any function.
* If you are hosting many regions on a single instance, you will probably notice a decrease in region startup time and maybe a slight increase in performance.
* Single regions won't notice anything different
* I believe this is reasonable since code outside the Linden client stack shouldn't be aware of the packet format being used
* I would love to have made the method protected, but the LoadBalancerPlugin is still calling it and resolving that would require more work
* This should fix a long standing issue where you often wouldn't see other people simply turn around without moving at all
* Arguably lastPhysRot (to mirror lastPhysPos) is not a good name, may change variable names later
* This may help http://opensimulator.org/mantis/view.php?id=2377 where large linksets do not always correctly delete - since a lost kill packet to the client could result in
the symptoms described
* Re-enabled the native ODE prim types when possible
* Fixed several invalid assumptions in the prim recycle process.
* Added better message for 'reused a disposed physicsactor'
* Added a way to recover from errors during collision_optimized
* Added a way to recover from an error condition where prim_geom wasn't reset properly
* Calculate an agents height in LLGetAgentSize() from apperance parameters rather than physics avatar numbers
* Another good looking patch from idb - thanks!
* If you're experiencing knee bendiness try the windows settings, as the *nix settings seem to now be incorrect. (this update does that, but you may have your own opensim.ini settings active.
* Mac users, pray to chi11ken to make you a .dylib version
* This is semi-tuned and post teravus hack. (Though I didn't apply the terrain pitting fix hack. I'm still deciding if it's necessary as there was a lot of work over the past several months on the heightfield collider.
* Please use '--enable-shared --disable-demos --disable-asserts' if you are building your own libode in the configure step. Asserts are pretty much useless for use with .NET
* This also updates ODE.NET as, there were some API changes in May that were just added to ODE.NET today.
* Experimenting with the PacketPool mechanism.
* It's still disabled in the code, however there's now a flag to enable it.
* Converted to use Generic Collections vs Hashtables, also now uses a list of 'OK to pool' packets, starting with the high volume PacketAck packet.
* Implemented a proper update thread
* Removed the UpdateLock Mutex as it's no longer needed because updates can only happen one at a time now.
* This should actually improve performance significantly.. But, see the warning on the next line!
* Warning: If there are deadlocks that the threadpool timer method was hiding, this will expose them for all the nastiness they are.
See mantis #2379 for some useful scripts that work with this.
Note that I added another configuration parameter in addition to the patch to allow for legacy behavior in the absence of new configuration settings.
fix for the LandManagementModule. I changed LandSnapshot to use
CachedUserInfo to take a load off the user server. Also, missing owner
profiles are legit, warning removed.
XmlIgnored the SceneObjectPart.SitTargetAvatar property, as (afaik) this is only used to indicate that a avatar is sitting (or about to sit) on the prim.
And there are situations where it might not be cleared, like if a client crashes while sitting on the prim. Which results in that avatar's id getting stored in any xml backup or taking the prim into inventory. And then its always classed as a avatar is sitting on that prim.
* This was not a problem with objects consisting of less than 30 prims, since the extra schedules would be ignored
* However, above approximately 30 prims extra schedules would actually occur.
* For instance, a 140 prim object would end up triggering approximately 2500 ObjectUpdates to every avatar in range rather than 140
* Hopefully, this change will improve client responsiveness on deselect and was one of the reasons that the AgentThrottle restriction started causing problems yesterday.
The average-value of modify.ModifyBlock.Height in LLClientView.cs:4170
seem to be incorrect or it isn't the average? Mhhh...
So the terrain build -> Flaten Sphere is unuseable.
I have put in a patch that contains a workaround while
the main problem is not solved.
* Puts remote requests in a single worker thread
* Worker thread only starts when there are agents to serve
* When there are no agents to serve, it shuts down
* A good example of how to deal with threads in non-shared modules so they don't end up consuming threads per regions
If prim is part of SOG, then ask the SOG to update the
position, rather than asking the part itself.
Ghosted child prims should no longer result from llSetPos.
Not sure if this is the right approach for all cases ,
would appreciate feedback on the patch.
The attached patch fixes mantis bug 2312 (llGetPos() returns incorrect
values for child prims where the root prim is rotated). Regression
tests still pass.
Incidentally AbsolutePosition which was used before looks a little
suspicious to me as its always going to return the wrong value if the
root prim is rotated. GetWorldPosition does take the rotation into
account, but AbsolutePosition is used in a lot of places. Though i
don't understand why there is both GetWorldPosition as well as
AbsolutePosition so I've left the latter alone.
[i also cleaned up some indent problems, --- dr scofield]
Attached is a patch for LLGround which was just plain broken and could
cause a runtime error. It now returns valid data with valid input (ie
the offset does not take the position off the edge of the sim), but a
runtime error will occur if invalid data is given.
On invalid data the LL servers return the ground height based on a
valid point closest to the effective position calculated using the
supplied offset. Is the OpenSim convention to replicate the LL servers
as closely as possible? If so I can submit an additional patch to
replicate the LL behaviour.
If you use load-oar to transfer region data from one sim to another
then currently inventory items can be left with unknown owner
permission which results in them being no-mod/no-copy for
everyone. The attached patch fixes things up so if the owner uuid does
not exist on the destination system then it assigns ownership (and the
creator for completeness) to the master avatar id. This will make it
much more practical to share copies of regions :-)
* I believe this was the cause of the remaining packet_out_of_order messages in the Linden client logs
* There were race conditions where multiple clientstacks would overwrite each other's sequence numbers
* Changing network bandwidth in the preferences will now have a much more noticeable effect - a user may want to increase this if data is being slow to download from opensim
llSetLinkApha is not fully implemented and has not been updated
to use the recently added GetLinkParts and associated implementation
pattern as per llSetLinkColor and llSetLinkPrimitiveParams.
* Fixes 1 too large count in region you're in (Region no longer reports the green dot about yourself to you)
* For all local requests, the region will check itself instead of going to it's web service. (optimization)
Prevent EventAbortException inner exceptions of
TargetInvocationException being re-thrown from
OpenSim.Region.ScriptEngine.Shared.ScriptBase.
Executor.ExecuteEvent
Implementation of llModifyLand() and There is a bug on
permission-check of land-terraforming: x an y-coordinates
are interchanged on function-call ExternalChecksCanTerraformLand.
Correct: x is west, and y is north. 2) Missing check of
"Other allow to terraform-flag" (Parcel.ParcelFlags.AllowTerraform)
Introduce a resend counter on the ack queue. The header "Resent" field is
now obsolete. Implement 3 resends on reliable packets, variable.
Increase default resend timeout to 3000ms and default silence threshold
to 350ms.
* This revision also makes the exception catcher of last resort print to the log instead of just the console
* This means logs should now capture sim crashing exceptions, though not those which are caused by a crash of the virtual machine itself
* Resolve redundancy by removing the OpenSimBase check - if modules are recieve nini config directly they should probably be the ones to do the checking
Addresses llDie issues. The attached patch catches run time
exceptions that occur during method invocation (of type
TargetInvocationException) and exposes the internal exception.
This makes it possible to pass out the SelfDeleteException.
Also added handlers in a couple places to make sure that
exception was being passed out far enough to be handled
correctly. Tested on DNE.
* This patch aims to store look at data when an avatar logs off in grid mode
* However, in my short test it doesn't appear to be working yet - numbers are being stored but they don't look correct
* But this doesn't appear to cause any login problems
* Thanks tyre
instead of the 130s timeout somewhere. Additionally, mark the map-tile as
offline. This partly fixes the TP problems of Mantis 2332; the rest is a viewer
problem (just relogin).
- Added lookup in the data-layer
- MySQL works
- SQLite doesn't have a grid-db, so it won't work there
- I added MSSQL-code to the best of my knowledge; but I don't know MSSQL :-)
- Added the plumbing up to OGS1GridServices. This speaks with the grid-server
via XMLRPC.
- Modified MapSearchModule to use the new data. It's backward compatible; if
used with an old grid-server, it just returns one found region instead of a
list.
- Refactored a bit.
Note: This updates data, grid-server and region code. No new files.
Fix: objects being duplicated inventory on detach. Fix: Mad jumping around of
attachments while editing. Fix: Attachments being persisted to database on
login. Fix: Attachments being persisted when changed by a script like
invisprim refresh. Fix: Attachpoint set, but not reset correctly. Fix: prevent
spurious full updates while editing attachments. Several other fixes
* we appear to both setup a default configuration and provude identical defaults when we later try to read the values - this is probably not really necessary
old way: each region module interested in chat from client had to
- subscribe to scene.EventManager.OnNewClient
- then in its OnNewClient delegate it would subscribe to
client.OnChatFromViewer to capture chat messages coming
new way: ChatModule is the only region module that uses the "old
way" approach but is now forwarding all client chat via
scene.EventManager.OnChatFromClient
- each region module interested in chat from client now only
subscribes to scene.EventManager.OnChatFromClient
this not only simplifies code, but also allows us to substitute
ChatModule with derived classes (ConciergeModule is going to be one
example).
Also, this changeset changes ChatFromViewer to ChatFromClient as it
doesn't necessarily have to be a viewer that is a chat source.
i've taken great care to only comment out those OnNewClient delegates
that were only used for getting at the client chat --- hope it's not
breaking anything.
* Switched it on by default
* Updated OpenSim.ini.example to reflect this
* Caught a UDP Server issue that occurs when the network pipe is saturated
* Still experimental :D
In Second Life's implementation of llParseString2List(), all spacer
strings which occur within the source string are included in the
resulting list. In OpenSim's implementation, any spacers occurring
at the beginning of the string are discarded; furthermore, if multiple
spacers occur between non-spacer elements, or at the end of the source
string, only the first spacer is added to the resulting list;
the remainder are discarded.
Thank you, fusspawn, for an implementation of osGetSimulatorVersion.
This patch removes the llRequestSimulatorData(..., 128) hack and
implements the functionality as an osFunction.
regions in the in-world map. It currently only returns the first region that
matches the search string; in a future version it will return more search
results.
Note: File added; run runprebuild.
* This actually probably doesn't make a huge difference. We still end up resending a massive number of packets in quite a few situations, but it's not easy to consistently
reproduce this.
* Might really need some throttling of packet send to the client on the OpenSim end.
* This is a HUGE update.. and should be considered fraut with peril.
* SSL Mode isn't available *yet* but I'll work on that next.
* DrScofld is still working on a radical new thread pump scheme for this which will be implemented soon.
* This could break the Build! This could break your Grid!
A number of the deprecated functions had missing parameters
causing compile errors. These included:
llSound, llMakeExplosion, llMakeFountain, llMakeSmoke,
llMakeFire, llPointAt, llXorBase64Strings, llSetPrimURL
OS/SL script conformance, implement llSoundPreload and
llRemoteLoadScript to behave as in SL. In SL
llSoundPreload appears to do nothing. In OS
llRemoteLoadScript shouts an error when invoked
"Deprecated. Please use llRemoteLoadScriptPin instead."
* This is an initial basic experimental code for inventory import and export from the region server
* Probably not yet ready for general use
* Thanks Kayne!
Estate/ Manager Owner Uses the Region/Estate Menu Region TAB,
and uses "Teleport Home All Users..." (Action Button), the
action will complete but no one will be teleported and all
users still function in the region ok.
- Added TeleportFinish event to the event-queue. It works better than
before (you can teleport), but it doesn't work quite right yet (you
start TPing after the TP again).
Beware: Still experimental and non-working.
The current llGetLinkName returns an empty string if a prim
with the given link number is not found in the link set. In
SL an empty string is only returned if the prim name is actually
blank, a NULL_KEY is returned instead with a couple of exceptions.
* If EventQueueGet is enabled, use that instead of a few UDP packets, if it's disabled use the UDP packets like normal.
* We *really* suggest users do not turn on EventQueueGet yet.. as it's highly experimental, broken, and resource intensive
* Updated OpenSim.ini.example
* Testers of EventQueueGet.. in OpenSim.ini, it's in the [Startup] Section.. change EventQueue = false, to EventQueue = true.
* returns FAKEEVENT instead of the connection returning a 502. It doesn't like our 502's for some reason.. so, in leau of this.. send it a fake event.
* Once again, this is still 'really early' code, so please don't blame us if you have no more threads left.
* This won't function yet as far as the client can tell.. because it doesn't respond to the first query with a 200 message.
* We have to figure out how to encode those binary values in the example code in the module...
* Committing this so we have a start point. Will continue to work on this more today.
* This means that saving notecards in prim inventories should now work.
* Not the nicest code in the world - the transactions system is pretty fugly right now
* PLEASE NOTE: Currently, the prim will not repersist until up to 15 seconds after it is unselected.
* What we really need is a proper mechanism so that any prim updates still waiting when the simulator is quit are performed before exit.
DNE and move all of DNE into the DotNetEngine directory. Remove references
that would cause the script runtime to load the entire engine + scene into
each script appdomain. This might help DNE memory consumption.
* It seems kind of silly to be building a 256x256 array just to use two 16 float blocks.. but for now the layerdata routine requires it so we'll go along with that.
* We only fill a 32x16 area of the 256x256 float array with data.
* We use patches 0,0 and 0,1 for the first and second patch to determine the direction and magnitude of the wind.
The binaries are still different, but that is only a small step away now.
The OSSLPrim has been removed. This commit will breal all scripts
using Prim.Scale(), etc, syntax. It was not secure and will have to
be brought back in another form.
Scene. Make the script engines check that the engine name in the
//Engine:language comment is a valid engine and treat it as a normal
comment if it's not.
//DotNetEngine: needs to be written as //ScriptEngine.DotNetEngine: now, since
that is it's real internal name. //XEngine: still works
Check the client dialog box (from top menu) WORLD / REGION ESTATE
/ REGION tab. The client dialog box seems to have a hard limit of
about 32 characters per line available for displaying the region
version number. Our regions are sending a string which is greater
than the limit, causing the client to wrap the text and look ugly.
* A little wind wouldn't hurt anyone, right? This is the 'slightly breezy' setting.. hopefully you won't notice 'much' of a difference.
* It turns out the terrain patch routine is similar enough to the wind version that it can be used to hack together a breeze generator with a few mods.
* Not much configuration.. yet. You only get breeze updates in the general vicinity of your camera now to keep bandwidth usage down.. and we're not talking about 'much' movement at the moment.
* initial version... could use improvement I'm sure.
The entire LSL API is now in the single, shared file
OpenSim/Region/ScriptEngine/Shared/Api/Implementation/LSL_Api_Base.cs
This is for both engines. The OSSL function are still separate.
and makes it use a common set of types in both engine. Fixes the issues with
running both engines and HTTP requests / listens / timers etc..
Also fixes a couple of minor Scene issues and a CTB by nullref.
Active gestures are sent as part of the login-response. Added
fetchActiveGestures to SQLite and MySQL; added an empty one for MSSQL and
NHibernate. Using the empty ones won't cause errors, but doesn't provide
persistence either, of course.
all methods needed outside the API ststic. Async command processing
is now wholly internal to the API. This sets the stage for the next
convergence step.
of the types, located in OpenSim/Region/ScriptEngines/Shared/LSL_Tyoes.cs
Also changes the compiler in DotNetEngine to use that. You _will_ need to
let your region recompile all your scripts!
Thank you, tyre, for a patch that refactors LSL to use a unified set of
method signatures and type names, reorders methods and removes unused
and adds new method stubs.
for inventory REST calls for the time being, as firefox, curl, and
also python's urllib2 cannot authenticate using digest auth.
fix permission checking for prim inventory to be the same as for
normal edit ops.
ordinary region modules and are able to coexist in one instance.
See http://opensimulator.org/wiki/ScriptEngines for details. There were
changes to OpenSim.ini.example, please note DefaultScriptEngine.
Also see the User docs and FAQ on the Wiki. Default is DotNetEngine.
* if a packet was really null it would be caught by the general exception handler at the top of the client thread, which would also provide more information and attempt a clean
shutdown
- Conformance fix: If the notecard can't be found, shout on DEBUG_CHANNEL as the docs state
- Bug fix: Don't let llGetNotecardLine throw an exception on freshly created notecards
* you'd think from all this that I was one of those fanatically tidy people. I can assure you from looking around my bedroom/study that this is not the case :)
* On script rez, XEngine was taking an m_scripts lock and then later on an m_parts lock when looking for a part by local id
* In the meantime, a scene object being deleted would take an m_parts lock and then later on try to take an m_scripts lock when it tried to trigger script removal
* There may be better ways to resolve this, but I believe that in general, we must always take an m_parts lock before an m_scripts lock
* There was a small window where region logins were allowed before modules were loaded - avatars logins that hit this window could have caused bad things to happen.
* A similar change will follow for grid mode sometime soon
* Monitors / for getting the seed cap for rez_avatar/request on application/llsd+xml && application/xml+llsd && !application/xml
* Experimental, this might break LibOMV temporarily.
- Add necessary dummy Dispose-methods where they are missing
- Implement the SQLite Dispose-methods
(currently only used for unit tests, in the next commit)
Correct a condition where a prim is linked before it has been persisted
for the first time and is then persisted through it's former SOG with
a bad offset position.
XEngine: Change a number of methods to use the new link part method.
Fix ALL_SIDES for llSetColor, llSetTexture, llSetAlpha, llSetPrimitiveParams,
llGetPrimitiveParams(PRIM_TEXTURE, PRIM_COLOR).
XEngine fixes: prevent queue overruns, prevent spamming when no key
is down. Release controls when conflicting permissions are requested
or permissions are refused later. Release when prim or script are deleted.
Fixes Scene script instance deletion semantics.
New patch (terrainedit2.patch) is attached that passes unit tests (
and in fact supplies the new unit test to pass..:) The previous
failure was due to the change of the brush size to be linear instead
of exponential, and the fact that the test parameters were assuming
the exponentional brush size.
This patch also removes dependency on 'duration' argument for
adjustment step size. This should address the widely differing reports
of instability with terrain editing.
Note: 'doing this right' implies ultimately reading the parameters
from the ini file, which this patch does not do.
* the methods were identical except that the Properties one did a check for the user profile beforehand. However, every caller was doing this already anyway.
fixes the terrain spikes, and is the result of mostly a tuning
operation on the smooth and flatten tools. I dug in and found that the
spikes apparently result from smooth's overly aggressive iteration
steps toward the average curvature, which leads to an instability that
blows up the heights. I introduced a scaling factor to dampen the
'duration' parameter which tames progress and seems to keep things
stable.
* There are still a lot of things that are hard coded to use http. They need to be fixed.
* Also includes directions
* A standard junk PEM file to append to app_settings/CA.pem in the client so SSL will work
Moved intialization of appearance from the SendInitialData
event handler into CompleteMovement handler. That ensures
that m_appearance is initialized before the inventory is
retrieved (so there is a place to check on attachments).
after using llTakeControls my sim receives about 200 messages per second,
l of which get queued , this could be because there is no lag as the viewer
and sim are on the same computer. The patch I have included checks to see if
the "Changed" param is 0 then searches the EventQueue for Control messages
being sent to the same localid, if it finds a message already in the Queue
and Changed==0 then the new message is only notifing you the key is being held,
since there is already a message the new one isnt needed so it isnt added
to the queue.
works. Also makes llGetFreeMemory return the constant 16384.
LLGetFreeMemory is useless in SL, as it never goes up, only down.
So, the only thing it is used for, in practice, is to detect an imminent
stack/heap collision, a danger we don't have.
Remove some unused osFunctions that were left in the LSL function
file from the separation way back when. Inline the osSetParcelMediaURL
code to get rid of the osFunction. Really need to add a way for one API
to call another.
functions return CLI strings, which, in the case of lists, don't
get wrapped. Therefore, the list had to be able to deal with
that CLI type here. The correct fix would be to change all LSL
function returns to LSL types.
* Patch attached that adds the check for uninitialized appearance when inventory items are received and processed. Also attempts to ensure that appearance is initialized even
when the profile cache has not been built.
* This will not fix the race condition, but should at least remove the unhandled exception that is being reported in Mantis 0002126.
* Thanks cmickeyb
* This patch aims to introduce look at direction persistence between logins. It won't be active until the second part of the patch is committed in about two weeks time. At
this point, region servers that haven't upgraded past this revision may run into problems
* This checkin upgrades the user database. As always, we recommend you have backups in case something goes wrong.
* Many thanks to tyre for this patch.
* From looking at the code there shouldn't be any difference, and this appears to speed up prim loading and make a certain failure case (where prims seem to be going out of
bounds on startup) less of a problem.
* However, if I've been bad please revert this patch
couple of debug statements for null items while we
diagnose our interrmittent inventory issues. I looked
at this and it seems helpful to others. We can back it
later if desired.
Thank you, ralphos, for a patch the adapts llSetColor and friends to
a change in the underlying types.
Also, thank you for a much cleaner way of casting types out of
lists, which I will adopt throughout.
Floating text, Rotation, Texture animation, Particle System
This will make "Eye Candy" scripts work without modification in
XEngine. The use of the CHANGED_REGION_RESTART hack is no longer
needed. Implemented in MySQL only, hovertext also in SQLite.
* Please note that this implementation has not been mathematically verified to be correct-as-described, it is probable that this implementation may produce differing results to Linden Lab. Testing (and test data) would be appreciated.
duplicate delivery of state_entry if a region is restarted just after
saving the script. Changes script state saves to no longer abort long-
running event handlers. Queues the save instead. Adds shutdown handler
to save script state on irderly shutdown
* Now it should properly inform the user and stop a login if a region server could not be contacted in order to expect a user (the last commit didn't actually quite work correctly)
* These are different types then the OMV types because changing them causes just about all grid comms to break. If these were the libOMV types, then libOMV couldn't change them ever again after that.. or we'd have a breakage whenever they changed them.
* This might introduce a map issue. Still checking it out.
* Testers, please try rezzing your inventory on this revision and sending me feedback in mantis.
* Users, it's not time to update yet. The dust still hasn't settled.
* This is a HUGE OMG update and will definitely have unknown side effects.. so this is really only for the strong hearted at this point. Regular people should let the dust settle.
* This has been tested to work with most basic functions. However.. make sure you back up 'everything' before using this. It's that big!
* Essentially we're back at square 1 in the testing phase.. so lets identify things that broke.
Thank you, openlifegrid, for a patch to move new user connections to
thread pool threads.
Reworked by me to fit current trunk.
I believe that that patch may be beneficial in reducing the cases
in which regions become unresponsive and will no longer accept
new logins.
Types extracted from a LSL_Types.list have to be down-cast initially
to the exact type of value type object that the Object actually is.
This would make for very cumbersome, ugly code when extracting list
parameter items in ll functions where a few implicit conversions
should be applied such as key -> LSLString and LSLInteger -> LSLFloat
(but not LSLFloat -> LSLInteger). This patch adds a set of GetXXXItem
member functions to the LLS_Type.list class, where XXX is the name
of the LSL_Type to be extracted: LSLFLoat, LSLInteger etc. All take
a single, int parameter that is the item number to be extracted.
If the parcel pointed to by the landmark has nothing in the
Description field (of the General tab in About Land...), a
NullReferenceException is thrown and the client is logged out.
I added a check to the calls to Helpers.StringToField() in OpenSim.Region.ClientStack.LindenUDP.SendParcelInfo to avoid
this issue.
Eliminate gretuitious KillObject packets. KillObject is sent to the
viewer for the root part only.
Also prevents the full update on deselect that makes a deleted object
reappear and stay visible until the background deleter got around to it.
We still send 2 KillObject packets for the root prim, that could be
avoided only at a cost in reliability. One packet seems an acceptable
price to pay for consistency.
The attached patch fixes connectivety checking of root- and
child-agents. If an agent's client isn't sending any packets
for some time (not even to StartPingCheck packets), we assume
that we lost connection and try to log them out.
Due to the many problems with not cleaned up child-agents, I thought it
might make sense to be able to see them on the console. 'show users' on
the region-server's console now outputs root- and child-agents (with
"root" or "child" column)
1000 chars to avoid the exception thrown by libomv at 1100 chars.
Change string->int conversion so it copes with non-numeric chars
after the number and no longer uses a float to parse the value.
It wraps constants in new LSLType(x), so that lists with
constant values are processed correctly. Contains changes to
the lsl.parser.cs that are not (yet) reflected in opensim-libs,
since this experimental patch affects XEngine only. Also contains
nuts.
* Unpatched code certainly looks bizarre - attempts to add a new client if we encountered a failure in processing a packet. No apparant ill effects on a sniff test.
* Thanks openlifegrid
* a couple more remain that preferably the original coders should look at in more detail (obsolete warnings in the ogp module and not overriding warnings in the script
engine)
* TimeStamp field of the AgentMovementCompletePacket in LLClientView.MoveAgentIntoRegion contained a fixed number. The attached patch changes this to a current
timestamp.
* Thanks HomerHorwitz
* You can replace "grid" as the asset system with "file" to save and load all your assets from a directory on your hard disk. Files are serialised to XML and saved in the format "/<dir>/0x/0x/0x/0000-0000-000000-0000-0000.xml"
* Directory <dir> is sharing the Asset Server URL path, use a normal path here instead (ie C:\xyz or /var/assets/).
* This probably wont work well in grid mode unless every sim has access to the same directory. This is mostly intended for standalone usage where quick and convenient access to assets is required.
Addresses the problem of ghost avatars. The problem was child-agents
sending data. Due to symmetry reasons, I considered that wrong.
Whenever an avatar A1 in region R1 looks at avatar A2 in adjacent
region R2, we have two possibilities to communicate: A2-root sends
to A1-child (both in region R2), or A2-child sends to A1-root
(both in region R1). Currently, the children send data in some cases,
and I guess the viewer gets puzzled about that and switches the child
to root as consequence (at least partly), so it becomes visible.
Found that during my ghost-hunt: Add some logging and you see that the
counting down happens twice (once at the place that is removed by the
patch, and once caused by the actual removal of the root-/child-agent
during LLClientView.Close). With the patch applied, I end up with correct
numbers, i.e. with 0 roots and 0 children after everyone has logged out.
which is a thin wrapper around a IWorkItemResult from the SmartThreadPool.
However, it is very easy to reimplement on top of basic threading and therefore
makes the IScriptInstance class independent of the specific threading
implementation.
engines that want to use the XEngine's instance handling and state
persistence can do so. IScriptInstance is optional, but it does
require the SmartThreadPool if it is used.
Issue is caused by the terrain height not being queried before
teleporting within the region. Teleporting between regions is
correct. Adding the neccessary checking to intra-region TP code.
* When OGP is active, disable SSL certificate chain validation. I'll add more options here to come, as well as a way to test against a group of known certificate subjects.
* The purpose of the check was to ensure that your home region was sent to your client properly when you connected. However, the desired result is that if the home region is down, but the destination region is online, the user should still get in.
* Changes: if the home region isn't available, set the message to the client so that the home region is the destination region.
This patch adds even more LSL compatibility to llSetPos() As stated in the
wiki http://wiki.secondlife.com/wiki/LlSetPos, [^] movement should also be
capped to 10m per call for unattached root prims.
Beside this issue the attached patch adds (hopefully) all known LSL script
delays (as ScriptSleep(), but still commented out) to LSL_BuiltIn_Commands.cs
and LSL_Api.cs and a lot of format cleanup.
at 1.0, it allows delays to function as coded. Smaller values will
reduce, larger values will increase the delays. A value of 0 is
permitted and results in on delay being applied.
Thank you, salahzar, for a patch that corrects the behavior
of PRIM_TYPE in llGetPrimitiveParams() and improves LSL
conformance in llGetNumberOfSides();
* There are a lot of changes and this is quite experimental. It's off by default, but you can turn it on by examining the bottom of the opensim.ini.example for the proper OpenSim.ini settings. Remember, you still need an agent domain..
* Furthermore, it isn't quite right when it comes to teleporting to remote regions (place_avatar)
The version of llSetLinkPrimitiveParams in LSL_BuiltIn_Commands.cs
has not been kept in sync with the additional functionality made
recently to the llSetLinkPrimitiveParams implementation in
LSL_Api.cs which includes support for setting glow, full bright and
other privative parameters. A patch to LSL_BuiltIn_Commands.cs is
provided the incorperates this functionality (duplicated).
animation, target omega, looped sound, script access pin, allowed
drop state and sale data. Loads it, too. Not all tested.
Code: No Nuts. Data: Cannot Guarantee Nut Free.
also enable LSLconformance on some texture functions as well.
Applied the part of the patch in Shared/.
The part for Common/ needs to be reworked to remove the reference
into Shared/
It is now possible to use module interfaces without referencing Scene.
Place those interfaces in OpenSim/Region/Interfaces. They may not
use any refs from OpenSim.Region.Environment as parameters.
This resolves a circular library ref introduced in r5949
* By default, texture rendering is on. This may be affected by using secure assets since your region hasn't registered with the gridserver before it asks for texture assets. It might also be affected by a slow asset server, so consider this release experimental.
* Defined interface IMapTileTerrainRenderer.
* Extracted "old" shaded maptile terrain rendering into ShadedMapTileRenderer;
streamlined it a bit and added "highlight" rendering to its "shadow"
rendering.
* Added "new" terrain-texture based maptile terrain rendering
(TexturedMapTileRenderer); made TerrainUtil.InterpolatedNoise public
* Adapted MapImageModule to allow switching between those two by configuration
* Added configuration option to OpenSim.ini.example
This patch improves LSL conformance by inserting a 0.2 second delay
into llSetPos/llSetRot, as the function specification states.
Thanks to M. Igarashi for the patch.
the first part of gesture persistence.
----------------------------------------------------------
Attachments no longer vanish on walking crossing. Teleport is still
problematic, but will now be blocked with message "Inconsistent
attachment state" rather than losing the attachment. Detach to be
able to TP in that case.
Change user server to handle attachment assets record properly. Ensure
that attachments are not re-rezzed on region crossing. Persistence
will NOT WORK with earliser UGAI!!
Change region server to match.
Diva Canto