aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Framework/Servers/HttpServer (unfollow)
Commit message (Collapse)AuthorFilesLines
2012-11-20minor: If logging full incoming HTTP data, don't deceptively print ... at ↵Justin Clark-Casey (justincc)1-2/+5
the end of the body.
2012-11-15Revert "Merge master into teravuswork", it should have been avination, not ↵teravus2-187/+109
master. This reverts commit dfac269032300872c4d0dc507f4f9062d102b0f4, reversing changes made to 619c39e5144f15aca129d6d999bcc5c34133ee64.
2012-11-14A few more AssemblyInfos in a few more dlls.Diva Canto1-0/+33
2012-11-09* Prep work switching the GetMeshModule over to a poll service.teravus1-1/+2
* This still has the image throttler in it.. as is... so it's not suitable for live yet.... The throttler keeps track of the task throttle but doesn't balance the UDP throttle yet.
2012-10-23minor: Get content type handler logger to log "unset" for the content type ↵Justin Clark-Casey (justincc)1-1/+1
instead of blank if no content type was set.
2012-10-23minor: Use LogIncomingToContentTypeHandler() method for incoming HTTP data ↵Justin Clark-Casey (justincc)1-8/+2
where this wasn't already used. This allows log level 5 (log sample or large part of incoming post data) to operate and removes copy/paste.
2012-10-18If we're avoiding printing a long request warning for a GetTexture CAP call, ↵Justin Clark-Casey (justincc)1-1/+1
check we received a request handler first since this is not guaranteed. Resolves harmless logged exception when content type and generic xml rpc requests take more than 3 seconds.
2012-09-30Allow setting max connections for an endpointMelanie1-1/+1
2012-09-30Revert "Added request.Proxy=null everywhere, as discussed in ↵Diva Canto3-4/+0
http://stackoverflow.com/questions/2519655/httpwebrequest-is-extremely-slow." But the patch is here, in case anyone wants to try it. This reverts commit 531edd51d82ecd6a842a2611c99e9919634491ef.
2012-09-30Added request.Proxy=null everywhere, as discussed in ↵Diva Canto3-0/+4
http://stackoverflow.com/questions/2519655/httpwebrequest-is-extremely-slow. Thanks R.Gunther (rigun@rigutech.nl) https://lists.berlios.de/pipermail/opensim-users/2012-September/010986.html
2012-09-29Fix bug where debug http level 6 could not be specified. Also converts ↵Justin Clark-Casey (justincc)1-3/+3
newlines at this level to '\n' to enable them to be logged.
2012-09-27create a new PollServiceHttpRequest req per loop since they can be sent toUbitUmarov1-5/+1
another working thread
2012-09-26just remove the damm thingUbitUmarov3-25/+3
2012-09-26 Seems nothing actually need the request body for getevents. so changeUbitUmarov2-3/+10
control flag to false
2012-09-26keep watchdog happy using it to kill his threadsUbitUmarov1-9/+1
2012-09-26 more changes to PollServiceUbitUmarov2-24/+18
2012-09-23Change the poll service to use a thread pool for replies to make sure theMelanie1-7/+38
event queues aren't blocked by other traffic.
2012-09-22Make BaseHttpServer throws say something useful.Diva Canto1-2/+2
2012-09-21Comment out the long unused afaik HTTP agent handlers.Justin Clark-Casey (justincc)2-84/+84
As far as I know, this was only used by the IBM Rest modules, much of which has been commented out for a very long time now. Other similar code uses HTTP or stream handlers instead. So commenting this out to reduce code complexity and the need to make this facility consistent with the others where it may not be used anyway. If this facility is actually being used then please notify me or uncomment it if you are core.
2012-09-21Add request number counting to incoming HTTP requests in the same way that ↵Justin Clark-Casey (justincc)1-18/+41
this was already being done for outgoing HTTP requests. This allows us to associate debug logging messages with the right request. It also allows us to put a request number on 'long request' logging even if other debug logging is not enabled, which gives us some idea of whether every request is suffering this problem or only some. This is a separate internal number not associated with any incoming number in the opensim-request-id header, this will be clarified when logging of this incoming request number is re-enabled. This commit also adds port number to HTTP IN logging to allow us to distinguish between different request numbers on different ports.
2012-09-20Insert a new log level 4 for HTTP IN and HTTP OUT that will log how long the ↵Justin Clark-Casey (justincc)1-15/+32
request took. This is only printed if debug http level >= 4 and the request didn't take more than the time considered 'long', in which case the existing log message is printed. This displaces the previous log levels 4 and 5 which are now 5 and 6 respectively.
2012-09-20Add ability to turn on/off logging of outgoing HTTP requests flowing through ↵Justin Clark-Casey (justincc)1-5/+5
WebUtil. This is for debugging purposes. This is controlled via the "debug http" command which can already log incoming requests. This now gains a mandatory parameter of in, out or all to control what is logged. Log messages are also shortened and labelled and HTTP IN or HTTP OUT to be consistent with existing UDP PACKET IN and PACKET OUT messages.
2012-09-16Catch a nullref in the code to suppress GetTexture warnign spam we can'tMelanie1-1/+1
do a thing about anyway.
2012-09-14Sequence/throttle asset retrievals.Melanie1-0/+2
2012-09-14Revamp the HTTP textures handler to allow a maximum of four fetchesMelanie3-17/+36
at any time and to drop requests for avatars n longer in the scene
2012-09-14Allow setting max connections for an endpointMelanie1-1/+1
2012-09-12Fix bug in logging sample input at debug http level 4.Justin Clark-Casey (justincc)1-1/+1
Also converts newlines to "\n" text.
2012-09-11Add levels 4 and 5 to "debug http" console command that will log a sample of ↵Justin Clark-Casey (justincc)1-12/+56
incoming request data and the entire incoming data respectively. See "help debug http" for more details.
2012-09-10add a extra httppool thread to compensate for webfetchinventoryUbitUmarov1-1/+2
2012-08-24Fix background inventory loading (Viewer 3) so it won't lag out the simMelanie2-2/+4
2012-07-25Move Watchdog and MemoryWatchdog classes into OpenSim.Framework.Monitoring ↵Justin Clark-Casey (justincc)3-0/+3
with other monitoring code from OpenSim.Framework
2012-07-17fix PollServiceRequestManagerUbitUmarov1-10/+47
2012-07-17Replace PollServiceRequestManager with older version, add extra logging toMelanie1-7/+6
event exceptions to see call path leading up to it.
2012-07-15Eliminate a spurious exception report when a https url is accessed with a wrongMelanie1-0/+2
CN
2012-07-06Convert the slow request queue into a regular queue and add some cleanupMelanie1-4/+18
and locking.
2012-07-05Fix the boo-booMelanie1-3/+3
2012-07-05Add an EventType enum and Type field to the poll service event args. This allowsMelanie2-11/+34
the manager to tell what type of event it is. All events except for lsl http in go to the "slow queue" which is run once per second as before.
2012-07-04 use the pollEvent timeout paramenter on poolingUbitUmarov1-4/+7
2012-07-04 added a timeout paramenter to PollServiceEventArgs, so each type can define ↵UbitUmarov1-1/+4
it's timeout
2012-07-04*test2* http poll: increased again the pool rate do 10/s but increased ↵UbitUmarov1-2/+2
timeout to 1s. So data there is less delay when there is new data, but enought waiting time for it to be avaiable
2012-07-04*test* slow http retries pool rate to original 1sUbitUmarov1-1/+1
2012-07-01*TO CHECK/REVIEW/REVERT/TEST whatever* pollService new requests get enqueued ↵UbitUmarov2-0/+206
to unified requests queue directly. Retries get into that every 100ms. 3 working threads as before plus another that only does retries timming.
2012-06-26Add an event to the poll service manager thread to allow starting itMelanie2-4/+14
when needed rather than once per second. That is just too slow!
2012-06-22Avoid a race condition where an incoming request to a script external URL ↵Justin Clark-Casey (justincc)2-5/+11
can trigger an exception is the URL was being removed at the same time. This involves three steps 1) Return gracefully in UrlModule.HttpRequestHandler() instead of throwing an exception when the url cannot be found in its index 2) Return true instead of false in HasEvents() if no matching request is found in the map. This call will only happen in the first place for raced requests. 3) Return a 404 in GetEvents() if the request is not in the index, rather than a blank 200 OK. Many thanks to Tom Haines in http://opensimulator.org/mantis/view.php?id=6051 for doing some of the work on this.
2012-06-16Fix bug introduced in commit c6e3752 (13 Jun 2012) where poll responses ↵Justin Clark-Casey (justincc)1-3/+1
would always return OK even if some other status code had been set
2012-06-15Make the "debug http" command available for robust as well as the simulator. ↵Justin Clark-Casey (justincc)1-0/+3
This allows one to see incoming requests as they happen. This required making everything use the common MainServer class for registering and retrieving http servers, rather than duplicate structures.
2012-06-15Add main instance to internal MainServer.m_Servers list to simplify internal ↵Justin Clark-Casey (justincc)1-8/+0
logic. This does require the server to be added before it is set as the main Instance
2012-06-15When setting debug http level, do this for all known http servers, not just ↵Justin Clark-Casey (justincc)1-10/+10
the main instance.
2012-06-14Shuffle "debug http" levels so that 1 and 2 now cause different levels of ↵Justin Clark-Casey (justincc)1-8/+27
warn to be logged if we receive invalid xml for xmlrpc.
2012-06-14Fix a regression in BaseHttpServer.HandleXmlRpcRequests() from recent c6e3752Justin Clark-Casey (justincc)1-1/+1
Accidentally make responseString null by default instead of String.Empty. It needs to be something in case the XmlRpcRequest deserialize throws an exception due to bad xml (a failure which we silently swallow!)
generated by cgit v1.1 at 2024-12-19 22:22:03 +0000