256c8cb62d
Send kill packets for avatars, too
2010-09-17 03:56:31 +01:00
7df4383235
Merge branch 'master' into careminster-presence-refactor
...
Integrate the next large patch.
Don't use this version, it has a ghost avatar issue. Next push
will fix it.
2010-09-17 03:49:30 +01:00
87a1e00f05
Remove the now unused partslock
2010-09-17 04:00:00 +02:00
7762301cea
Revert "* Changed 11 calls for session info to the more optimized API method"
...
This reverts commit 5dc9ea2f24
2010-09-17 01:57:26 +01:00
b49cb3355f
Fixed a regression in SOG.Copy()
2010-09-16 17:57:00 -07:00
5321e0e69f
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2010-09-16 17:31:24 -07:00
860b2a502f
Changed SceneObjectGroup to store parts with the fast and thread-safe MapAndArray collection
2010-09-16 17:30:46 -07:00
e3f1533705
Fix build break by replacing Items.LockItemsForWrite() with lock (Items) {}
2010-09-17 01:21:28 +01:00
5072429263
Send KillPackets on the Task queue rather than the State queue
...
Object updates are sent on the task queue. It's possible for an object update to be placed on the client queue before a kill packet comes along.
The kill packet would then be placed on the state queue and possibly get sent before the update
If the update gets sent afterwards then client get undeletable no owner objects until relog
Placing the kills in the task queue should mean that they are received after updates. The kill record prevents subsequent updates getting on the queue
Comments state that updates are sent via the state queue but this isn't true. If this was the case this problem might not exist.
2010-09-17 01:16:21 +01:00
5b18f8be25
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2010-09-17 01:01:12 +01:00
251f2444d0
Revert "* Changed 11 calls for session info to the more optimized API method"
...
This reverts commit 5dc9ea2f24
2010-09-17 01:50:48 +02:00
670357367e
Removing debug
2010-09-16 23:32:18 +01:00
1b2edfe75f
JustinCC is evil. f7b28dd3 broke script persistence. This fixes it.
2010-09-16 23:19:46 +01:00
1e4e978f7f
Removing debug
2010-09-16 23:26:53 +02:00
a0c87b5af6
JustinCC is evil. f7b28dd3 broke script persistence. This fixes it.
2010-09-16 23:12:32 +02:00
d52b2b9ee7
minor: remove some mono compiler warnings
2010-09-16 20:46:24 +01:00
512ded6eb5
Catch a nullref
2010-09-16 21:07:09 +02:00
d50a072336
Suppress the strange "Result not Dictionary" messages that happen when
...
a region queries for a nonexistent presence.
2010-09-16 20:21:30 +02:00
ec89dc99b7
Some small bug fixes
2010-09-16 18:58:25 +02:00
58b1ee8e2c
Merge branch 'master' into careminster-presence-refactor
2010-09-16 18:04:50 +01:00
f5e3d5a33a
Add the modules include line back that i dropped by mistake
2010-09-15 23:24:58 +01:00
7383173d3d
extend m_entityUpdates.SyncRoot lock in LLClientView.ProcessEntityUpdates() to reduce scope for kill/update race conditions
...
This is necessary because it was still possible for an entity update packet to be constructed, the thread to pause, a kill to be sent on another thread, and then the original thread to resume and send the update
This would result in an update being received after a kill, which results in undeletable ghost objects until the viewer is relogged
Extending the lock looks okay since its only taken by kill, update and reprioritize, and both kill and update do not take further locks
However, evidence suggests that there is still a kill/update race somewhere
2010-09-15 23:06:38 +01:00
e6bc77d832
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2010-09-15 23:03:02 +01:00
39d27fc879
rename SceneObjectGroup.DeleteGroup() to DeleteGroupFromScene() to improve code readability
2010-09-15 22:29:58 +01:00
16f296f489
Instead of locking SOG.Children when a group is being removed from the scene, iterate over an unlocked list instead
...
Previously, deadlock was possible because deleting a group took a SOG.Children lock then an m_entityUpdates.SyncRoot lock in LLClientView
At the same time, a thread starting from LLClientView.ProcessEntityUpdates() could take an m_entityUpdates.SyncRoot lock then later attempt to take a SOG.Children lock in PermissionsModule.GenerateClientFlags() and later on
Taking a children list in SOG appears to be a better solution than changing PermissionsModule to not relook up the prim. Going the permission modules root would require that all downstream modules not take a SOG.Children lock either
2010-09-15 21:51:13 +01:00
c1ccfa6d8d
Mark up OpenSim.ini.example for use with an automatic configuration tool.
...
Remove seldom changed options from OpenSim.ini.example.
Remove non-overridable settings like [Architecture] from OpenSimDefaults.ini
2010-09-15 20:29:28 +01:00
571693d7b6
Revert last commit
2010-09-15 08:16:38 -07:00
77eb8ade8c
Don't throw an exception if the authorizationservice is missing a serviceuri config entry. Not every authorization service will need a URI.
2010-09-15 08:08:29 -07:00
1f74f013b2
Don't fart when deserializing a prim with no ParentGroup set yet
2010-09-15 07:59:28 -07:00
0d034712ba
Make AuthorizationService optional and disabled by default. It depends on
...
external PHP and the default configuration makes it throw for not
having a URL to this PHP
2010-09-15 15:56:03 +01:00
3275fbec54
Fix the above
2010-09-14 23:06:44 +01:00
7f7bbbb84a
Fix the above
2010-09-14 23:05:42 +01:00
18beb796bc
Merge branch 'master' into careminster-presence-refactor
2010-09-14 22:50:38 +01:00
585e575d23
Allow to place the connection strings and providers for estate and region
...
in their oqn sections, for those of us who don't want a catch-all
DatabaseService section.
2010-09-14 22:47:45 +01:00
8ed08a10d9
Remove long unused Region/Framework/ThreadTracker
...
All methods had already been deleted, only GetThreads() was left
2010-09-14 22:34:55 +01:00
609375bf37
Fix "show threads" to show threads now being managed by OpenSim.Framework.Watchdog
2010-09-14 22:24:11 +01:00
540a1912ce
Merge branch 'careminster-presence-refactor' of ssh://3dhosting.de/var/git/careminster into careminster-presence-refactor
2010-09-14 22:14:23 +02:00
356b1ef1d9
Merge branch 'master' into careminster-presence-refactor
2010-09-14 22:16:02 +01:00
e27f59cf42
reorder some code to avoid merge issues in the future
2010-09-14 22:15:16 +01:00
2f41bc4ca7
Fix merge issues
2010-09-14 22:14:08 +01:00
1212cb0c74
Merge branch 'master' into careminster-presence-refactor
2010-09-14 22:03:42 +01:00
cc290abf7a
Adjust the code for ghost prim removal to new information from Kitto
2010-09-14 22:13:44 +02:00
d299f8b55c
Fix a small left over buglet and also add checking of the physics actor
...
position on backup. This way, ant object that has been moved will be
checked for ghost prims as soon as it is persisted.
2010-09-14 19:41:53 +02:00
095d400f5b
fix OpenSim.Tests.ConfigurationLoaderTest to satisfy requirement that OpenSimDefaults.ini is present
...
this should allow the testsuite to run again and the autobuild to complete
2010-09-14 17:01:53 +01:00
5d48e3c0bb
Revert "Move OpenSimDefaults,ini into config-include in order to put it with all the other default files"
...
This reverts commit c3259e9c26
2010-09-14 03:30:07 +01:00
95702129c7
Revert "Repopulate OpenSim.ini.example with OpenSimDefaults.ini until somebody does the work of deciding what users commonly change and what they don't"
...
This reverts commit fc48eb7b54
2010-09-14 03:07:41 +01:00
ec305a4825
LSL compatibility: Don't throw an exception if an invalid key is passed to llInstantMessage, instead shout about it and apply the usual delay. This now matches SL.
2010-09-13 17:43:45 -07:00
c3259e9c26
Move OpenSimDefaults,ini into config-include in order to put it with all the other default files
2010-09-14 01:38:53 +01:00
fc48eb7b54
Repopulate OpenSim.ini.example with OpenSimDefaults.ini until somebody does the work of deciding what users commonly change and what they don't
2010-09-14 01:30:12 +01:00
51411d5664
add the missing ini example
2010-09-14 00:32:49 +01:00
Melanie