aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/ClientStack/LindenUDP/LLPacketServer.cs (unfollow)
Commit message (Collapse)AuthorFilesLines
2009-02-12* optimized usings.lbsa711-3/+0
2009-02-09From Alan Webb <awebb@linux.vnet.ibm.com>Sean Dague1-2/+2
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.
2009-01-12* Separate starting a client thread into a separate Start() method (which ↵Justin Clarke Casey1-0/+2
matches the existing Stop() and Restart() methods)
2009-01-12* Extend PacketHandlerTest to fire in a packet.Justin Clarke Casey1-3/+4
* Can't test result yet since the Client thread handles it with unpredictable timing
2008-11-06* test: Add assert checking that the circuit which trigger the socket ↵Justin Clarke Casey1-1/+2
exception has been closed
2008-10-24* minor: eliminate some mono compiler warningsJustin Clarke Casey1-2/+2
2008-10-24* Stop passing in unnecessary pameters to CreateNewCircuitJustin Clarke Casey1-12/+9
2008-10-24* Stop creating a circuit if the client fails authentication (i.e. the ↵Justin Clarke Casey1-21/+51
region server wasn't told that it was coming) * 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
2008-10-09* minor: initialize udp server in unit testJustin Clarke Casey1-6/+0
2008-10-09* minor: very small doc additionJustin Clarke Casey1-1/+1
2008-10-03* Put in some infrastructure to allow tweaking of packet queue throttle ↵Justin Clarke Casey1-2/+11
values for the total throttle (the one that throttles all packet output) * Not complete yet
2008-10-01* refactor: rename interface to put standard I prefix in frontJustin Clarke Casey1-2/+2
2008-10-01* Fix bug in LLPacketHandler where the tickcount when a packet was ↵Justin Clarke Casey1-1/+1
approximately sent was not being stored in the ack record * This meant that acks were being sent out every 250ms when the ack timer fired, rather than when they had actually aged past m_ResendTimeout
2008-09-14* Converted a number of methods within the login processes from private to ↵Adam Frisby1-3/+2
protected. * Made several methods virtual to allow derivative overrides. * Minor cleanups.
2008-09-07* minor: Clean up of logging messages to make following the client login ↵Justin Clarke Casey1-8/+34
process easier * documentation
2008-09-06* This is the fabled LibOMV update with all of the libOMV types from JHurlimanTeravus Ovares1-4/+4
* 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.
2008-07-21* eliminated some warnings and added some const and readonlieslbsa711-1/+1
* refactored some member names for readability and ccc (code convention conformance) * took away two refs from Rest.Inventory since * System.IO is part of System * System.Xml.Serialization is part of System.Xml
2008-05-16Formatting cleanup.Jeff Ames1-6/+6
2008-05-02fixed the dos line endingsSean Dague1-149/+149
2008-05-02* Commit 2/3 - Please dont attempt to update to this revision until all 3 ↵Adam Frisby1-149/+150
are in.
2008-04-21* Optimised using statements and namespace references across entire project ↵Adam Frisby1-1/+0
(this took a while to run).
2008-03-18Formatting cleanup.Jeff Ames1-26/+25
2008-03-04Merged 3Di code that provides scene and avatar serialization, and plugin ↵Johan Berntsson1-4/+4
support for region move/split/merge. See ThirdParty/3Di/README.txt. Unless the new modules are used there should be no noticeable changes when running OpenSim.
2008-02-20Minor cleanup.Jeff Ames1-2/+2
2008-02-09* Change logger to handle [<entry>] where <entry> contains non alphabetic ↵Justin Clarke Casey1-1/+1
characters * Change logger to not print extra line if [<entry>] <text> like string is not logged * Remove more of my previous chatty debugging statements
2008-02-08Still chasing logout memory leak. Putting in small changes and temporary ↵Justin Clarke Casey1-3/+9
light verbosity to this end
2008-02-06* Chasing down memory leak where memory used by a client is not returned on ↵Justin Clarke Casey1-1/+7
client logout * This code may or may not be on the right track, but I want to save my work so far.
2007-12-27* Optimized usingslbsa711-6/+7
* shortened references * Removed redundant 'this' * Normalized EOF
2007-12-27* Added slightly better object sit handlingTeravus Ovares1-2/+2
* Added sit handling for sit targets * Implemented llSitTarget() * Implemented llAvatarOnSitTarget() * Sit targets do not persist sim restart.
2007-12-21* Remapped the Connection shutdown path.Teravus Ovares1-1/+1
* This fixes a *bunch* of Mantis bugs related to the following * -- Neighbouring simulators not appearing after relog * -- Login to simulator only to be logged off by simulator. * -- ThreadAbort * -- Unable to shutdown circuitCode: x
2007-12-18* Fix for mantis 0000040 After client logout remote host closed connection ↵Teravus Ovares1-1/+4
on Simulator makes sim unuseable->'Closed Connection Called' * I've fundamentally changed a few things, so this is experimental * The routine that I used needs to be tested on Linux. I don't expect it to cause a problem, but hey, it might. * Child agents are still not logged off properly, so when the first set time out, the second set get logged off also, on the second log in if the second login is initiated before the first one fully times out.
2007-12-15* some work on not storing the circuitPack (bad thing if we're going to ↵lbsa711-3/+4
reuse packets) * some work on encapsulation, code convention compliance and beautification. * also, some ignores
2007-12-13Minor cleanupJeff Ames1-1/+0
2007-12-10saved OpenSim source code from the giant rampaging unterminated copyright ↵Jeff Ames1-2/+2
notice of doom
2007-10-30* Optimized usingslbsa711-18/+15
* Shortened type references * Removed redundant 'this' qualifier
2007-10-29as per the "Filesystem cleanup for OpenSim repository" mailing list thread. ↵MW1-1/+1
Have flattened the OpenSim.Framework project/namespace. The problem is that the namespace is still wrong as its "OpenSim.Framework" while the directory is "OpenSim\Framework\General" , so we need to decide if we change the directory or correct the namespace. Note this has lead to a big flat project, but I think a lot of the files we most likely don't even use any longer. And others belong in other projects/namespaces anyway.
2007-09-25* Now the OGS1GridServices has a LocalBackEndServices that it forwards ↵lbsa711-26/+26
intra-instance requests to * Every Scene has a ClientManager (as every dog it's day) since two scenes can have the same circuit as client.
2007-09-24* Encapsulated all CommunicationsManager serviceslbsa711-1/+1
2007-09-19* Added AddToPhysicalScene and RemoveFromPhysicalScene and added to ↵lbsa711-47/+0
MakeAvatar and MakeChildAgent respectively * Removed unused m_child from ClientView
2007-09-18* Moved SendLogoutPacket back to IClientAPI.lbsa711-7/+1
2007-09-18* Yet some more connectivity restructuringlbsa711-13/+28
* We now have CloseAllAgents( circuit ) and CloseAllCircuits( agentId ) for great justice ( but alas, still only closing on one single scene - be brave! ) * Login and ConnectionClosed now eventified and moveified awayified * Killed off unused NullClientAPI * Now the client is almost only responsible for its own closing. ( I will get that scene out of there ) * Lookin' good!
2007-09-18* Handlerized ViewerEffectlbsa711-16/+12
* Now there-is-only-client-manager * First step towards moving Logout and ConnectionClosed out of Client and into something else (which will let us get rid of ClientView reference to ClientManager * General posititvity, peace, love and understanding
2007-09-18* Replaced usage of ClientView with IClientAPIlbsa711-4/+4
* Some propertification and hideousness of fields.
2007-09-10mass update of urls in source code to new websiteSean Dague1-1/+1
2007-09-08Converted the LSL scripting engine into a IRegionModule, so now all ↵MW1-0/+8
"modules" share a common base interface and are loaded from the single loader. (It seems to work fine, but I have left the old scriptengine loader, incase we have to change back). Removed the reference to OpenJpeg in the DynamicTextureModule, to see if that was causing the build problem someone is having. Added a Temporary fix for the "existing connection was forcibly closed by the remote host" exception on windows when a user logs out of a multiregion instance. Some early work to prepare for improving the way clients are updated (about prims etc).
2007-08-16Deleted old inventoryCache.csMW1-5/+4
2007-08-06* renamed some scene to worldlbsa711-2/+2
* passing on NotImplemented from Update()
2007-07-26* Started renaming world to Scenelbsa711-17/+5
* Update and UpdateMovement now first stores array to avoid collection update exceptions * Ignored some bins
2007-07-22* Some work in progress code: Inventory cache, start of inventory ↵MW1-1/+1
server/service, userprofile cache, inventory handling. (non of it is enabled yet (or at least it shouldn't be). * Fixed some of the problems with crossing regions when flying: you should no longer sink to ground level when crossing (should keep roughly your right height). Should no longer sometimes get sent back to the centre of the current region when attempting to border cross. But instead sometimes you will find you avatar stop at the edge of region and you will need to start moving again to retry the crossing (which should then work). This code is partly based on Babblefrog's issue #212 patch. [I think I have some ideas of how to solve the stopping at edges problem, just want to get the inventory code done first] * Capabilities code has now been moved to the OpenSim.Framework.Communications project as some of the caps code will be tightly tied to inventory/asset handling and it was causing a two way reference problem when it was in its own project/dll. This is a Big commit as I was going to keep my inventory work local until I had it in a working state, in case it brakes anything, but its getting harder to keep in sync with svn.
2007-07-18*New Configuration System, much easier and less buggy compared to the ↵mingchen1-0/+1
original system in place *View RegionInfo.cs for an example on how it works! *This hopefully copies all the files over, but who knows :)