aboutsummaryrefslogtreecommitdiffstatshomepage
Commit message (Collapse)AuthorAgeFilesLines
* Make "terrain save" more friendly by telling the user if we have saved and ↵Justin Clark-Casey (justincc)2011-11-171-18/+32
| | | | putting out a useful complaint message if we haven't for some reason
* If the entire simulator is shutting down then don't bother to unload the ↵Justin Clark-Casey (justincc)2011-11-172-17/+35
| | | | | | | scripts from the appdomain in XEngine. All the other actions (script state save, etc.) still occur. This makes shutdown where there are many scripts vastly quicker.
* Stop OdePrim and OdeCharacter insanely overriding set LocalID to set their ↵Justin Clark-Casey (justincc)2011-11-169-44/+40
| | | | | | | own private m_localID property but leaving get to return the then unset PhysicsActor.LocalId! Instead, just have both subclasses use the PhysicsActor.LocalID property. This restores collision functionality that fell away in 45c7789 yesterday
* Call public ForEach instead of using m_entities directly. No semantic ↵Dan Lake2011-11-161-2/+2
| | | | changes, just cleanup
* In AddNewClient, iterator over copy of entities rather than copying under ↵Dan Lake2011-11-151-2/+3
| | | | read lock
* Comment out the vebose logging on HGFriendsModule.Justin Clark-Casey (justincc)2011-11-161-12/+11
| | | | Recent issues in http://opensimulator.org/mantis/view.php?id=5794 were not related to HG friends
* Fix the build breakJustin Clark-Casey (justincc)2011-11-161-1/+1
|
* Look up a homeURL only when it's not null, in addition to not being string.EmptyJustin Clark-Casey (justincc)2011-11-161-1/+1
|
* Enable log message to tell us what type of url is being requested from a ↵Justin Clark-Casey (justincc)2011-11-151-3/+3
| | | | user's homeurl
* Make tracked per scene thread names conform to the majorirty format.Justin Clark-Casey (justincc)2011-11-152-4/+9
| | | | This is <thread-name> (<region-name>)
* fix build break on UserManagementModule.Justin Clark-Casey (justincc)2011-11-153-15/+23
| | | | This also adds time since started to "show threads". Unfortunately these two changes got mixed in.
* Do proper locking of UserManagementModule.m_UserCache when getting.Justin Clark-Casey (justincc)2011-11-151-29/+57
| | | | | | | | This might help with [USER AGENT CONNECTOR]: new connector to () [USER AGENT CONNECTOR]: Unable to contact remote server for GetServerURLs [USER AGENT CONNECTOR]: Malformed Uri : Argument cannot be null. Parameter name: uriString
* improve formatting of "show threads"Justin Clark-Casey (justincc)2011-11-151-1/+8
|
* send a watchdog heartbeat for a poll worker thread when it's actually runJustin Clark-Casey (justincc)2011-11-151-0/+3
|
* temporarily increasing logging in HGFriendsModule for debuggingJustin Clark-Casey (justincc)2011-11-151-1/+24
|
* distinguish between FriendsSimConnector and FriendsServiceConnector in logJustin Clark-Casey (justincc)2011-11-152-17/+17
|
* Add number of milliseconds since last update to "show threads"Justin Clark-Casey (justincc)2011-11-152-4/+18
|
* Dont' bother with a userAgentService != null check right after we've ↵Justin Clark-Casey (justincc)2011-11-151-11/+10
| | | | constructed it
* Add comments about trying to avoid synchronous work off the ↵Justin Clark-Casey (justincc)2011-11-153-5/+16
| | | | EventManager.OnMakeRootAgent event since this is on the critical path for transfer of avatars from one region to another.
* Instead of having scene add/remove collision events directly to the OdeScene ↵Justin Clark-Casey (justincc)2011-11-151-20/+38
| | | | | | | | | collision event dictionary, marshall them via a change dictionary first. This is to avoid a complicated tri-thread deadlock on region crossing for avatars with attachments, where 1) XEngine starting up scripts can lock XEngine.m_Scripts and then try to lock OdeScene._collisionEventPrim while starting up a script due to avatar border crossing 2) An existing collision event will lock OdeScene._collisionEventPrim and then try to lock SP.m_attachments while trying to send the collision event to attachments 3) The avatar still entering the region will lock SP.m_attachments and then try to lock m_Scripts to start more attachment scripts.
* use a more efficient dictionary in OdeScene._collisionEventPrim rather than ↵Justin Clark-Casey (justincc)2011-11-151-7/+4
| | | | a list
* Removed unused and mostly commented out SceneCommunicationService methodsJustin Clark-Casey (justincc)2011-11-152-116/+3
| | | | As far as I can see, the SCS is only now used for informing neighbours of up/down status and possibly sending child agent updates and close requests
* Remove prebuild reference to now gone PumaCode.SvnDotNet.dllJustin Clark-Casey (justincc)2011-11-151-1/+0
|
* Remove unused RegionCommsListener/IRegionCommsListener.Justin Clark-Casey (justincc)2011-11-153-275/+0
| | | | All this is now being handled through IEntityTransferModule and SimulationService instead, and has been for some time.
* remove SceneCommunicationService.OnAvatarCrossingIntoRegion. This stuff is ↵Justin Clark-Casey (justincc)2011-11-154-11/+6
| | | | not being used any more - it's now IEntityTransferModule and SimulationService instead
* Rename FetchFriendslist() -> CacheFriends() and RefetchFriends() -> ↵Justin Clark-Casey (justincc)2011-11-152-15/+13
| | | | RecacheFriends() to reflect their intended function
* refactor: rename m_NeedsListOfFriends => m_NeedsListOfOnlineFriends to ↵Justin Clark-Casey (justincc)2011-11-151-5/+5
| | | | better reflect its actual function
* For clients that are entering a simulator from initial login, stop executing ↵Justin Clark-Casey (justincc)2011-11-153-11/+34
| | | | | | | | FriendsModule.FetchFriendslist() asychronously. Executing this asynchronously allows a race condition where subsequent friends fetches hit a cache that FetchFriendsList() had not yet populated. Changing this to synchronous may improve issues where a user does not see friends as online even though they are. I don't believe synchronous is a problem here, but if it is, then a more complicated signalling mechanism is required. Locking the cache isn't sufficient.
* refactor: Don't create a new UUID for passing uuids to client - UUIDs are ↵Justin Clark-Casey (justincc)2011-11-151-4/+6
| | | | structs are so not passed by reference (and they're immutable!)
* Add test for removing a friendship.Justin Clark-Casey (justincc)2011-11-144-17/+51
|
* Merge branch 'remove-scene-viewer'Dan Lake2011-11-147-427/+22
|\
| * Remove SceneViewer from ScenePresence to reduce quadruple queueing ofDan Lake2011-11-117-427/+22
| | | | | | | | | | | | | | | | | | | | | | | | prim update to only triple queuing. Existing method was: 1. Schedule prim for update, adding to scene update list 2. Update on SOGs during heartbeat queues update onto each SceneViewer 3. Update on SPs during heartbeat queues update onto each IClientAPI 4. ProcessEntityUpdates queues updates into UDP send stack Now the SceneViewer has been eliminated so updates are scheduled at any time and then put onto the IClientAPI priority queues immediately during SceneGraph.UpdateObjectGroups.
* | Improved method doc for AddFriend() - it actually does set up a two-way ↵Justin Clark-Casey (justincc)2011-11-143-9/+7
| | | | | | | | | | | | relationship. Rename IFriendsModule.AddFriend() to AddFriendship()
* | minor: remove some mono compiler warningsJustin Clark-Casey (justincc)2011-11-144-11/+2
| |
* | Add test for adding a friend whilst onlineJustin Clark-Casey (justincc)2011-11-144-7/+53
| |
* | Add very simple FriendsModuleTests.TestNoFriends()Justin Clark-Casey (justincc)2011-11-144-3/+91
| |
* | If a friends identifier which is too short is given to ↵Justin Clark-Casey (justincc)2011-11-142-1/+10
| | | | | | | | | | | | HGFriendsModule.GetOnlineFriends() then spit out a warning rather than failing on the String.Substring(). This is to progress http://opensimulator.org/mantis/view.php?id=5789
* | Do a ScenePresence null check in HGMessageTransferModule.SendIMToScene() to ↵Justin Clark-Casey (justincc)2011-11-142-1/+4
|/ | | | | | stop a NullReferenceException being thrown if an HG IM is sent to a simulator running multiple regions This is an attempt to address http://opensimulator.org/mantis/view.php?id=5791
* doh - correct build breakJustin Clark-Casey (justincc)2011-11-121-6/+7
|
* Add threads to the watchdog thread list before we start them.Justin Clark-Casey (justincc)2011-11-121-5/+5
| | | | Hopefully this wil make "WATCHDOG: Asked to update thread ## which is not being monitored" messages." go away.
* Make log messages on authentication failure more explicitJustin Clark-Casey (justincc)2011-11-111-13/+26
|
* extract common ScenePresence setup code into Init() method for ↵Justin Clark-Casey (justincc)2011-11-111-39/+31
| | | | ScenePresenceSitTests
* add SP.PhysicsActor checks to other sit/stand testsJustin Clark-Casey (justincc)2011-11-111-0/+4
|
* add sit and stand on ground testJustin Clark-Casey (justincc)2011-11-111-0/+23
|
* extract ground sit code into SP.HandleAgentSitOnGround() for consistency ↵Justin Clark-Casey (justincc)2011-11-111-12/+19
| | | | with other sitting code.
* Pick up the intended ConsolePrompt from [Startup] instead of console_prompt.Justin Clark-Casey (justincc)2011-11-111-1/+1
| | | | Addresses http://opensimulator.org/mantis/view.php?id=5786
* As with prim sitting avatars, make an avatar phantom when it sits on the ↵Justin Clark-Casey (justincc)2011-11-111-6/+5
| | | | | | | ground and solid again when it stands. This is to avoid http://opensimulator.org/mantis/view.php?id=5783 when a collision with a ground sitting avatar causes that avatar to automatically stand and sometimes not be able to move A better solution may be to keep gound sitting avatars solid but remove their collision status. However, this requires some physics code work.
* Restore sending of OutPacket() for object kills removed in commit c7dd7b1.Justin Clark-Casey (justincc)2011-11-111-26/+20
| | | | | | | | | | OutPacket() must be called within the m_killRecord lock. Otherwise the following event sequence is possible 1) LLClientView.ProcessEntityUpdates() passes the kill record check for a particular part suspends before OutPacket() 2) Another thread calls LLClientView.SendKillObject() to delete the same part and modifies the kill record 3) The same thread places the kill packet on the Task queue. 4) The earlier thread resumes and places the update packet on the Task queue after the kill packet. This results in a ghost part in the sim that only goes away after client relog. This commit also removes the unnecessary m_entityUpdates.SyncRoot locking in SendKillObject.
* remove unncessary IClientAPI parameter from SP.SendSitResponse()Justin Clark-Casey (justincc)2011-11-111-4/+3
|
* As per mailing list last week, remove facility that would automatically move ↵Justin Clark-Casey (justincc)2011-11-112-228/+198
| | | | | | | the avatar if prim with no sit target was out of sitting range. Now, no movement occurs. Note that you can still sit on a prim with an explicit sit target from any distance, as was the case before.
generated by cgit v1.1 at 2024-12-25 02:53:07 +0000