Application.xml client settings, serverToclient not working

[Guest]

Hi,

Can anybody tell me about what is wrong with ServerToClient value in application.xml as:

When i set ClientToServer to 25600 the server works fine and stops clients from sending data over 25600.

Whereas this does not apply to ServerToClient property. I set it to 25600 but the server keeps on sending data very much over 25600 to clients.

I have tried all permutation combinations, ServerToClient Bandwidth limit does not seem to work for me.

Please help.

Thanks,

Sahil.

[Guest]

Its still not working!!!!!!!

[ShivenK]

Hi,

Are you using the server-wide Application.xml located under RootInstall/conf/_defaultRoot_/_defaultVhost_ directory. Or is the application specific xml.

Can you please send me your configuration in both cases?

Thanks,

Shiven

[Guest]

Hi ShivenK,

I am so sorry i was not able to get back to your reply, Here is the application.xml of RootInstall followed by app-specific xml:

<Application>

    <Process>

    <!-- The process scope in which instances of this application will  -->

    <!-- run. An application instance may run by itself in a process,   -->

    <!-- along with other instances or even be distributed among        -->

    <!-- multiple processes. The process scope may be one of adaptor,   -->

    <!-- vhost, app or inst, with the default being vhost. These values -->

    <!-- determine how various application instances are grouped into   -->

    <!-- processes, in the following manner,                            -->

    <!--                                                                -->

    <!--    adaptor       all instances in an adaptor run in a process  -->

    <!--    vhost         all instances in a vhost run in a process     -->

    <!--    app              all instances in an app run in a process      -->

    <!--    inst          each instance runs in a process               -->

    <Scope></Scope>

    <!-- Whether to distribute the above process scope into multiple    -->

    <!-- processes. One can distribute vhosts, applications, instances  -->

    <!-- or even clients across the process group. These entities have  -->

    <!-- a complete enclosing relationship or a strict ordering among   -->

    <!-- them, for e.g. adaptors contain vhosts, which contain apps     -->

    <!-- etc. The complete ordering is,                                 -->

    <!--                                                                -->

    <!--   . Adaptors                                                   -->

    <!--   . VHosts                                                     -->

    <!--   . Applications                                               -->

    <!--   . Instances                                                  -->

    <!--   . Clients                                                    -->

    <!--                                                                -->

    <!-- The entity chosen for <Distribute> tag must be lower or        -->

    <!-- contained in the process scope specified in the <Scope> tag,   -->

    <!-- and is specified as vhosts, apps, insts and clients. In other  -->

    <!-- words, if the process scope is set to adaptor, one can choose  -->

    <!-- vhosts, apps, insts, or clients. On the other hand, if the     -->

    <!-- process scope is set to "app" then one can specify only insts  -->

    <!-- or clients for distribution.                                   -->

    <!--                                                                -->

    <!-- Choosing to distribute clients indicates a stateless           -->

    <!-- application, wherein all clients need not rendezvous on a      -->

    <!-- single instance. Typical VOD (Video on Demand) applications    -->

    <!-- are stateless, as each user plays some content independent of  -->

    <!-- all other instances. On the other hand a chat or gaming        -->

    <!-- application is not stateless as all participants need to       -->

    <!-- rendezvous on a single instance wherein they share the         -->

    <!-- application state. By default, entities are distributed over   -->

    <!--  3 cores, which maximizes the server's ability to utilize      -->

    <!--  system resources.  Distribution may be turned off by setting  -->

    <!-- numprocs to 0 or 1. Also by default the server chooses to      -->

    <!--  distribute the immediately lower entity to the one specified  -->

    <!--  in <Scope> tag.                                               -->

    <Distribute numprocs="3">Instances</Distribute>

    <!-- The lifetime of a core process can be specified to not exceed  -->

    <!-- a certain period, wherein any connections after this will      -->

    <!-- result in rolling over to a new core process. The <RollOver>   -->

    <!-- tag determines this time period which is specified in seconds. -->

    <!-- For stateless applications, the older core process continues   -->

    <!-- to serve the earlier connections. In this case one can limit   -->

    <!-- the maximum number of core processes that may be running       -->

    <!-- simultaneously by specifying an appropriate value in the       -->

    <!-- <MaxCores> tag.                                                -->

    <!--                                                                -->

    <!-- An application is considered to be stateless if it is          -->

    <!-- configured to distribute clients over multiple processes. This -->

    <!-- may be done explicitly specifying "clients" for distribution,  -->

    <!-- or implicitly by choosing "inst" process scope, of course      -->

    <!-- distribution must be enabled in the first place by setting     -->

    <!-- numprocs to a value greater than 1. If the application is not  -->

    <!-- stateless, the server ignores the <MaxCores> value and forces  -->

    <!-- it to 1. This ensures that such an application instance is not -->

    <!-- split across multiple processes, but results in clients        -->

    <!-- getting disconnected on a periodic basis. As such this is      -->

    <!-- mostly useful for stateless applications, and may have limited -->

    <!-- applicability otherwise.                                       -->

    <!--                                                                -->

    <!-- Process rollover may be disabled by setting the rollover time  -->

    <!-- to 0 and is the default. Limiting the number of core processes -->

    <!-- may be disabled by setting it to 0 and is the default too.     -->

    <LifeTime>

      <RollOver></RollOver>

      <MaxCores></MaxCores>

    </LifeTime>

    <!-- Determines the maximum number of process failures that will be -->

        <!-- tolerated before a core process is disabled. Once disabled the -->

        <!-- the master will not launch a core process till some minimum    -->

        <!-- recovery time elapses. This may be used to guard against a DOS -->

        <!-- behaviour where a faulty core consumes all cpu time by being   -->

        <!-- repeatedly launched very quickly. The recovery time is         -->

        <!-- specified in seconds, and setting it to 0 disables checking    -->

        <!-- for process failures. Note loading an application via admin    -->

        <!-- tools or APIs bypasses this check.                             -->

        <MaxFailures>2</MaxFailures>

        <RecoveryTime>0</RecoveryTime>

    </Process>

    <!-- Specifies whether or not to load this app when the server starts. -->

    <LoadOnStartup>false</LoadOnStartup>

    <!-- Max time an app instance can be idle (no clients connected)  -->

    <!-- before being unloaded. Note: an app instance only goes idle  -->

    <!-- after the last client disconnects. If the instance is loaded -->

    <!-- with no clients connected, it is not considered to be idle.  -->

    <!-- The max idle time is specified in seconds. (20 minutes)      -->

    <MaxAppIdleTime>1200</MaxAppIdleTime>

  <!-- You can override the settings specified in Vhost.xml for  -->

  <!-- this vhost by uncommenting the tag below. You can disable -->

  <!-- auto-close idle clients for this app, or change the max   -->

  <!-- idle time for clients connected to this vhost. If the max -->

  <!-- idle time is not specified here, or invalid (i.e. <= 0),  -->

  <!-- then we use whatever was set at the vhost level.          -->

  <!--

  <AutoCloseIdleClients enable="false">

    <MaxIdleTime>3600</MaxIdleTime>

  </AutoCloseIdleClients>

  -->

  <!-- This section contains information about configuring the script engine. -->

    <JSEngine>

        <!-- This specifies the max size (Kb.) the runtime can grow to before -->

        <!-- garbage collection is performed.                                 -->

        <RuntimeSize>1024</RuntimeSize>

        <!-- This specifies the max number that the server will skip the GC   -->

        <!-- when the JS engine is busy.  By default, the server will only    -->

        <!-- do the GC when the JS engine is not busy.  If MaxGCSkipCount is  -->

        <!-- set to 0, the server will force a GC regardless of the state of  -->

        <!-- the JS engine.  And if MaxGCSkipCount is positive, the server    -->

        <!-- will force a GC when the number of skip count exceeds the value  -->

        <!-- in MaxGCSkipCount.                                               -->

        <MaxGCSkipCount>-1</MaxGCSkipCount>

        <!-- This specifies the max amount of time (in seconds) that a script -->

        <!-- function can take to execute. If it takes longer than this max   -->

        <!-- timeout period, then the script is considered "runaway", and     -->

        <!-- script execution will be terminated. By default, this value is 0 -->

        <!-- which means there is no timeout and we do not check for runaway  -->

        <!-- scripts. Normally you would only want to check for this in a     -->

        <!-- development environment while debugging and testing apps. In a   -->

        <!-- production environment, presumably, your apps have been tested   -->

        <!-- and should not go into any infinite loops (or become "runaway"). -->

        <MaxTimeOut>0</MaxTimeOut>

        <!-- FMS provides the load API which allows loading other script        -->

        <!-- files within main.asc. File specified is first resolved using the  -->

        <!-- main.asc, if the load file is not found in that location, path     -->

        <!-- specified in this tag is used to resolve the file. This tag can    -->

        <!-- contain multiple and they must be separated by a ';' Resolution is -->                 

        <!-- is done left to right                                              -->

        <ScriptLibPath>${APP.JS_SCRIPTLIBPATH}</ScriptLibPath>

        <!-- This section specifies virtual directory mappings for file      -->

        <!-- objects in the script. By using virtual directories, you can    -->

        <!-- specify file directories for different applications. If the     -->

        <!-- beginning portion of a file path matches the virtual directory  -->

        <!-- that is specified, then the storage location of the file maps   -->

        <!-- to the location specified by the virtual directory mapping. To  -->

        <!-- specify a virtual directory mapping, you first specify the      -->

        <!-- virtual directory, followed by a semi-colon, followed by the    -->

        <!-- actual file location. <virtual dir>;<actual dir>                -->

        <FileObject>

            <!--

            <VirtualDirectory>/flashapps;C:\dev\Code\Flash\tincan\flashapps\</VirtualDirectory>

            -->

        </FileObject>

        <XMLSocket>

            <!-- XMLSocket will close the connection if it receives this much -->

            <!-- data from XML server without receiving an end tag. This can  -->

            <!-- be overridden by each XMLSocket by specifying the proeprty,  -->

            <!-- XMLSocket.maxUnprocessedChars. However, this cannot exceed   -->

            <!-- what is specified here.                                      -->

            <MaxUnprocessedChars>4096</MaxUnprocessedChars>

        </XMLSocket>

        <NetConnection>

            <!-- Default object encoding to use for SSAS NetConnection. This  -->

            <!-- can be AMF0 or AMF3.  The default is AMF3.  It can be        -->

            <!-- overridden for each individual NetConnection by setting the  -->

            <!-- NetConnection.objectEncoding property to either 0 for AMF0,  -->

            <!-- or 3 for AMF3.                                               -->

            <ObjectEncoding>AMF3</ObjectEncoding>

        </NetConnection>

    </JSEngine>

    <StreamManager>

        <!-- Specifies the physical location where recorded streams are stored. -->

        <!-- For example, <StorageDir>c:\myapp\streams\</StorageDir>            -->

        <StorageDir></StorageDir>

        <!-- Enables mapping of virtual stream paths to different physical locations. -->

        <!-- This tag is applicable when using the Authorization plug-in.  -->

        <QualifiedStreamsMapping enable="false" />

        <!-- Specifies the physical location where streams are duplicated           -->

        <!-- This location serves as a backup for Stream files                   -->

        <!-- This location must already exist, and when a stream            -->

        <!-- is copied here, it will be categorized by instance name by default        -->

        <!-- To include application name in the file path created the tag        -->

        <!-- change the appName attribute to "true", if not "false"            -->

        <!-- Example<DuplicateDir appName="true">c:\backupStreams</DuplicateDir>    -->

        <DuplicateDir appName="true"></DuplicateDir>

        <!-- Specifies the cache prefix that is passed from the origin server to the      -->

        <!-- proxy server.  This is used by the proxy server as a relative path to       -->

        <!-- locate the cache file in the CacheDir                                       -->

        <!-- The type can be set to "path" or "sname".  If type="path", the physical     -->

        <!-- path of the recorded stream will be appended to the prefix.                 -->

        <!-- If type = "sname", the stream name will be appended to the prefix           -->

        <!-- The prefix can be any text with/without preset paramaters, all parameters   -->

        <!-- are sourrounded by ?.  Everything within the ? will be substituted by the   -->

        <!-- server.                                                                     -->

        <!--                                                                             -->

        <!-- ?IP?      represents the IP address of the server.                           -->

        <!-- ?APP?     represents the application name.                                   -->

        <!-- ?APPINST? represents the application instance.                               -->

        <!-- ?VHOST?   represents the vhost name.                                         -->

        <!--                                                                             -->

        <!-- Since the prefix is used as a relative path in the proxy CacheDir, all      -->

        <!-- parameters should be separate by '/'.  For example, if type = "path" and    -->

        <!-- the recorded stream is located in c:/fms/flvs/foo.flv.                      -->

        <!-- data/?IP? will be resolved into data/xxx.xxx.xxx.xxx/c/fms/flvs/foo.flv     -->

        <!-- ?APPINST?/data will be resolved into app1/inst1/data/c/fms/flvs/foo.flv     -->

        <!-- origin1/data/ will be resolved into origin1/data/c/fms/flvs/foo.flv         -->

        <!--                                                                             -->

        <!-- By default, type is set to "path" and the prefix is set to ?IP?. The IP in  -->

        <!-- the prefix can be used to avoid file collision.  For example, the proxy     -->

        <!-- server might be connecting to two different origin servers with the same    -->

        <!-- file in c:\data\foo.flv.  And you want to keep the cache files separate in  -->

        <!-- the proxy server.  Similarly, if you want to share the cache file among     -->

        <!-- origin servers, you should not include the IP as parameter.                 -->

        <!-- Remember this prefix is a relative path used by proxy server to lookup the  -->

        <!-- cache stream file.  The prefix can be any name except some special          -->

        <!-- characters such as \, :, *, ?, ", <, >, |.                                  -->

        <CachePrefix type="path">?IP?</CachePrefix>

        <!-- Specifies the interval for updating the cache streaming in the proxy        -->

        <!-- server.  Value is in millisecond.  Minimum = 10 sec, maximum = 24 hours,    -->

        <!-- default = 10 min.                                                           -->

        <CacheUpdateInterval></CacheUpdateInterval>

        <MaxBufferRetries>128</MaxBufferRetries>

        <ThrottleBoundaryRequest enable="false">8</ThrottleBoundaryRequest>

        <ThrottleLoads enable="true">8</ThrottleLoads>

        <ThrottleDisplayInterval>64</ThrottleDisplayInterval>

        <!-- If there is no key frame at the point of seek server will perform     -->

        <!-- accurate seeking based on the nearest key and intermediate frames.    -->

        <!-- Default = true. The zone attribute can be set to "client", "server".  -->

        <!-- If zone is set to "client", no keyframe will be generated in the      -->

        <!-- the server, but all the information for accurate seeking will passed  -->

        <!-- and processed on the client side.                                     -->

        <!-- If zone is set to "server", a new keyframe will be created on the     -->

        <!-- server side based on the previous keyframe and intermediate frames.   -->

        <!-- If the zone attribute is missing, server will fall back to the old    -->

        <!-- behavior where enhanced seeking will be handled by the server for     -->

        <!-- Sorenson codec.  For the other video codec, enhanced will be handled  -->

        <!-- by the client.                                                        -->

        <EnhancedSeek zone="client">true</EnhancedSeek>

        <!-- Specifies the interval for saving generated keyframes in an .flv file  -->

        <!-- Value is in milliseconds.  Default = 60000 millisecond.  Minimum = 0   -->

        <!-- millisecond and maximum = 24 hours.  If value is unspecified or out of -->

        <!-- the range, it is set to 1000 millisecond.                              -->

        <KeyFrameInterval>60000</KeyFrameInterval>

        <!-- Force a GC if stream list grows over the specified value  -->

        <!-- Default = 100. Will be GC'd during application GC         -->

        <!-- interval.  If value is unspecified or -1, it means        -->

        <!-- unlimited.                                                -->       

        <MaxStreamsBeforeGC>100</MaxStreamsBeforeGC>

        <Audio>

            <CombineSamples>

                <!-- Combine samples if there are more than N subscribers.  -->

                <Subscribers>8</Subscribers>

                <!-- Stop combining samples if cpu is lower than this.      -->

                <LoCPU>60</LoCPU>

                <!-- Start combining samples if cpu is more than this.      -->

                <HiCPU>80</HiCPU>

                <!-- Combine N audio samples into one message.  Zero means  -->

                <!-- that audio is never collapsed.  Default is 0.          -->

                <MaxSamples>0</MaxSamples>

            </CombineSamples>

            <SendSilence>

                <!-- Specifies the interval (in milliseconds) for sending silence message -->

                <!-- when no audio is publishing to a live stream. Default is 3 second    -->

                <!-- Set to 0 to disable the silence message                              -->

                <!-- Silence message is used to support older version of Flash players.   -->

                <!-- The server will only send the silence message to clients which are   -->

                <!-- specified in <UserAgent> inside the <Client> section.  And bit 0x01  -->

                <!-- is used to control the silence message.                              -->

                <Interval></Interval>

            </SendSilence>

            <NotifyAudioStop enabled="false">

                <!-- Specifies how long to wait before the server notifies the client     -->

                <!-- when audio stops in the middle of a live/recorded stream.            -->

                <!-- Default is 3 second and minimum is 1 second.                         -->

                <Duration></Duration>

            </NotifyAudioStop>

        </Audio>

        <Live>

            <AccumulatedIFrames>

                <!-- Specifies the maximum duration (in second) of intermediate frames a  -->

                <!-- live stream can hold in the buffer.  The buffer contains a history   -->

                <!-- of the video messages up to the last keyframe so that clients are    -->

                <!-- able to catch up to the latest message even when they join in        -->

                <!-- between keyframes.  If the duration in the buffer is bigger than     -->

                <!-- MaxTime, server will clear the messages.  This setting is to avoid   -->

                <!-- the buffer from getting too big in the server and should be set      -->

                <!-- bigger than the keyframe interval.  Default value is set to -1 which -->

                <!-- means unlimited.                                                     -->

                <MaxTime>-1</MaxTime>

                <!-- Specifies the maximum size (in kilobytes) of intermediate frames a   -->

                <!-- live stream can hold in the buffer.  The buffer contains a history   -->

                <!-- of the video messages up to the last keyframe so that clients are    -->

                <!-- able to catch up to the latest message even when they join in        -->

                <!-- between keyframes.  If the size in the buffer is bigger than         -->

                <!-- MaxSize, server will clear the messages.  This setting is to avoid   -->

                <!-- the buffer from getting too big in the server and should be set      -->

                <!-- bigger than the total size of intermediate frame between keyframes.  -->

                <!-- Default value is set to -1 which means unlimited.                    -->

                <MaxSize>-1</MaxSize>

            </AccumulatedIFrames>

            <!-- A message queue is used to buffer the incoming messages from the publisher   -->

            <!-- so that the server can send messages in chunks to the subscribers.  Queuing  -->

            <!-- can be disabled so that individual messages will be sent out immediately to  -->

            <!-- the subscribers.  The "enabled" attribute can be set to "false" to disable   -->

            <!-- queuing.                                                                     -->

            <Queue enabled="true">

                <!-- Specifies how often the server will flush the message queue by size   -->

                <!-- (in bytes).   Set value to 0 also disable queuing.  Default is 4096.  -->

                <MaxQueueSize>4096</MaxQueueSize>

                <!-- Specifies how often the server will flush the message queue by time   -->

                <!-- (in milliseconds).   The default value is set to 500 milliseconds.    -->

                <MaxQueueDelay>500</MaxQueueDelay>

                <!-- Specifies whether the server will flush the message queue when a      -->

                <!-- data message arrives.  This is important when streaming only data     -->

                <!-- messages, so the server will send out the messages immediately.       -->

                <!-- Default is set to true.                                               -->

                <FlushOnData>true</FlushOnData>

                <!-- When queuing is enabled, messages in the queue can be combined to     -->

                <!-- form aggregate messages.  The "enabled" attribute will determine      -->

                <!-- whether aggregate messages will be created.  Default is "true".       -->

                <AggregateMessages enabled="false">

                    <!-- This setting determines the maximum size (in bytes) of  -->

                    <!-- aggregate messages created from the message queue when  -->

                    <!-- when aggregate messages are enabled.  Default is -1,    -->

                    <!-- which means that the aggregates can be as large as the  -->

                    <!-- queue size allows.                                      -->

                    <MaxAggMsgSize>-1</MaxAggMsgSize>

                </AggregateMessages>

            </Queue>

            <!-- Ping the existing publisher if another client tries to publish to the -->

            <!-- same live stream.  The value specifies how long (in milliseconds)     -->

            <!-- the server expects a response from the existing publisher.  If the    -->

            <!-- publisher fails to response within the timeout.  The server will      -->

            <!-- allow the second publisher to take over the live stream.  Default is  -->

            <!-- 2000 milliseconds.  And set value to -1 to disable ping.              -->

            <PublishTimeout>2000</PublishTimeout>

            <!-- This specifies the max latency (in milliseconds)of incoming messages  -->

            <!-- in a live stream.  When a data message is injected to a live stream   -->

            <!-- through the server side script, the max latency is used to determine  -->

            <!-- the timestamp of the message.  If the injection time is bigger than   -->

            <!-- the maximum latency, the time difference between the last message     -->

            <!-- time and the injection time will be added to the timestamp of the     -->

            <!-- data message.  Otherwise the last message time will be used.  Default -->

            <!-- is 500, and minimum is 0.                                             -->

            <MaxLatency>500</MaxLatency>

            <!-- When this is enabled, the server will start the stream timeline of a -->

            <!-- live stream as soon as it is published. This means that any delay    -->

            <!-- publish time and the time the first message arrives will be added to -->

            <!-- stream time as a fixed offset. This is useful if the publisher wishes-->

            <!-- to wait before sending messgages on a published stream, and wants    -->

            <!-- that delay to be accounted for in the stream timeline. This should   -->

            <!-- not be enabled when using DynamicStreaming to switch between live    -->

            <!-- streams, since this can cause the streasm to be appear out of sync.  -->

            <StartClockOnPublish>false</StartClockOnPublish>

            <!-- When this is disabled, the server will not erase any previously      -->

            <!-- recorded stream when a new publisher arrives in live mode.  By       -->

            <!-- default, this is enabled and the recorded stream will be erased.     -->

            <EraseOnPublish>true</EraseOnPublish>

        </Live>

        <!-- Whether we send status message NetStream.Play.Start for all the commands     -->

        <!-- including play, seek and unpause.  If false, only the play command will get  -->

        <!-- the status message                                                           -->

        <SendDuplicateStart>true</SendDuplicateStart>

        <!-- Controls commands we send OnMetaData with.  The value can be set to true,    -->

        <!-- false, or once.  These values determines when the onMetaData message is      -->

        <!-- sent, in the following manner,                                               -->

        <!--                                                                              -->

        <!--     true     Sends onMetaData for play and seek commands.                    -->

        <!--     false    Sends onMetaData for play only.                                 -->

        <!--     once     Falls back to FMS 1.x behavior, sends onMetaData based on the   -->

        <!--              start position regardless of the command.  If no onMetaData is  -->

        <!--              found from the start position, no onMetaData will be sent.      -->

        <!-- By default, the value is set to true.                                        -->

        <SendDuplicateOnMetaData>true</SendDuplicateOnMetaData>

        <Recording>

            <!-- The default maximum size (in KB) of a recorded file.    -->

            <!-- recording completely. -1 means that no maximum size is to be enforced. Default is -1. -->

            <MaxSize>-1</MaxSize>

            <!-- Sets a cap on the maximum size (in KB) of a recorded file.        -->

            <!-- Values set using the MaxSize field or SSAS will not be allowed to exceed this value.-->

            <!-- 0 disables recording completely. -1 means that no size cap is to be enforced. -->

            <!-- Default is -1. -->

            <MaxSizeCap>-1</MaxSizeCap>

            <!-- The default maximum duration (in seconds) of a recorded file.          -->

            <!-- 0 disables recording completely. -1 means that no maximum duration is to be enforced. -->

            <!-- Default is -1. -->

            <MaxDuration>-1</MaxDuration>

            <!-- Sets a cap on the maximum duration (in seconds) of a recorded file.    -->

            <!-- Values set using the MaxDuration field or SSAS will not be allowed to exceed this value.-->

            <!-- 0 disables recording completely. -1 means that no size cap is to be enforced. -->

            <!-- Default is -1. -->

      <MaxDurationCap>-1</MaxDurationCap>

  </Recording>

    </StreamManager>

    <SharedObjManager>

        <!-- Specifies the physical location where shared objects are stored -->

        <!-- on the server.                                                  -->

        <!-- For example, <StorageDir>c:\myapp\sharedobjects\</StorageDir>   -->

        <StorageDir></StorageDir>

        <!-- Specifies the physical location where shared objects are duplicated       -->

        <!-- This location serves as a backup for SharedObject files                   -->

        <!-- This location must already exist, and when a shared object            -->

        <!-- is copied here, it will be categorized by instance name by default        -->

        <!-- To include application name in the file path created the tag        -->

        <!-- change the appName attribute to "true", if not "false"            -->

        <!-- Example<DuplicateDir appName="true">c:\backupSharedObjects</DuplicateDir>    -->

        <DuplicateDir appName="true"></DuplicateDir>

        <!-- Deleted slots will be purged if the current version - deleted version is greater -->

        <!-- than the value specified.  -->

        <ResyncDepth>-1</ResyncDepth>

        <!-- Specifies the timeout value in seconds to automatically unlock -->

        <!-- sharedobject if there is a client waiting for update. -1 to wait for ever. -->

        <LockTimeOut>-1</LockTimeOut>

        <!-- SharedObjects are automatically commited if they are dirty, this can be disabled -->

        <!-- for all shared objects within this instance by specifying false. SASS will have to call-->

        <!-- "save" or SharedObject.commit to persist the shared object else all data will be lost when the app shuts down-->

        <AutoCommit>true</AutoCommit>

        <!-- Maximum properties that can be set per shared object -->

        <!-- -1 is unlimited.  -->

        <MaxProperties>-1</MaxProperties>

        <!-- Each property within each shared object cannot exceed value specified in bytes. -->

        <!-- -1 is unlimited.  -->

        <MaxPropertySize>-1</MaxPropertySize>

    </SharedObjManager>

    <!-- Specifies whether or not to allow tunneling connections into this   -->

    <!-- application. The acceptable values are true, false, http1.1only and -->

    <!-- keepalive. The following describes the effect of each value,        -->

    <!--    true        allow all tunneling connections                      -->

    <!--    false       disallow all tunneling connections                   -->

    <!--    http1.1only allow only http 1.1 connections                      -->

    <!--    keepalive   allow http 1.1 or http 1.0 keep alive connections    -->

    <AllowHTTPTunnel>true</AllowHTTPTunnel>

    <Client>

        <!-- Use 10 if you have bandwidth values for FMS 2, 2 otherwise. -->

        <WindowsPerAck>2</WindowsPerAck>

        <Bandwidth override="yes">

            <!-- Specified in bytes/sec. The default provides about 20Mbps. -->

            <ServerToClient>65536</ServerToClient>

            <!-- Specified in bytes/sec -->

            <ClientToServer>65536</ClientToServer>

        </Bandwidth>

        <!-- Bandwidth cannot be set higher than the value set here. -->

        <BandwidthCap override="no">

            <!-- Specified in bytes/sec -->

            <ServerToClient>131072</ServerToClient>

            <!-- Specified in bytes/sec -->

            <ClientToServer>131072</ClientToServer>

        </BandwidthCap>

    <!-- Configure downstream (server to client) bw detection. -->

    <!-- To disable native bw detection and use script-based   -->

    <!-- bw detection, set the enabled attribute to false. By  -->

    <!-- default, native bw detection is enabled.              -->

    <BandwidthDetection enabled="true">

      <!-- Don't send data to client greater than this rate (in Kbps). -->

      <!-- Default of -1 means don't need to throttle; just send the   -->

      <!-- data at whatever rate is necessary to measure bw.           -->

      <MaxRate>-1</MaxRate>

      <!-- Amount of data sent to client. Server sends a series of random bytes -->

      <!-- to the client, each time sending this much more data. For example,   -->

      <!-- x bytes are sent, followed by 2x bytes, followed by 3x bytes, etc.   -->

      <!-- until <MaxWait> time has elapsed.                                    -->

      <DataSize>16384</DataSize>

      <!-- Specifies how long (in sec.) server will send data to the client.    -->

      <MaxWait>2</MaxWait>

    </BandwidthDetection>

        <MsgQueue>

            <Live>

                <!-- Drop live audio if audio q exceeds time specified. time in milliseconds -->

                <MaxAudioLatency>2000</MaxAudioLatency>

                <!-- Default buffer length in millisecond for live audio and video queue. -->

                <MinBufferTime>8000</MinBufferTime>

            </Live>

            <Recorded>

                <!-- Default buffer length in millisecond for live audio and video, value cannot be set below this by Flash player. -->

                <MinBufferTime>2000</MinBufferTime>

            </Recorded>

            <Server>

                <!-- Ratio of the buffer length used by server side stream -->

                <!-- to live buffer.  The value is between 0 and 1.  To    -->

                <!-- avoid break up of audio, the ratio should not be more -->

                <!-- than 0.5 of the live buffer.                          -->

                <BufferRatio>0.5</BufferRatio>

                <!-- This specifies whether outgoing messages are sent in       -->

                <!-- different priorities for server to server connection.  By  -->

                <!-- default, prioritization is set to false which means        -->

                <!-- messages are sent out through one channel with the same    -->

                <!-- priority.  If the value is set to true, messages will be   -->

                <!-- sent through multiple channels with different priorities.  -->

                <!-- And the priority is based on the message type.             -->

                <Prioritization>false</Prioritization>   

                <!-- The minimum number of video messages to queue at the start -->

                <!-- of a stream.  When an H264 stream or other pipelined codec -->

                <!-- stream begins it needs some number of messages, at most 64 -->

                <!-- to begin playback and this setting ensures that regardless -->

                <!-- of the buffer set there are enough messages at the start   -->

                <!-- to begin playback quickly.  Making this less than 64, down -->

                <!-- to 0 won't queue up extra messages but may cause H264      -->

                <!-- content with a low FramesPerSecond to delay before start   -->

                <MinQueuedVideo>64</MinQueuedVideo>

            </Server>

            <Diagnostic>

                <!-- Whether diagnostic logging specific to message queue is    -->

                <!-- enabled.  Default is false.                                -->

                <Enable>false</Enable>

            </Diagnostic>

        </MsgQueue>

        <HTTPTunnel>

            <!-- The following two parameters affect the latency observed by -->

            <!-- a client tunneling into the server. Low values will reduce  -->

            <!-- the latency but increase the network bandwidth overhead.    -->

            <!-- Applications desiring low latency may use 128/256 for idle  -->

            <!-- post and ack intervals, while those not susceptible to high -->

            <!-- latencies may use 1024/2048. The default setting provides   -->

            <!-- medium latency and is set at 512/512.                       -->

            <!-- This specifies the interval at which the client should send -->

            <!-- idle posts to the server to indicate that the player has no -->

            <!-- data to send. This is needed to enable the server to send   -->

            <!-- downstream data. This interval is specified in milliseconds -->

            <!-- and must be in the range 0 to 4064 msec.                    -->

            <IdlePostInterval>512</IdlePostInterval>

            <!-- This specifies the max time the server may wait before it   -->

            <!-- sends back an ack for a client idle post.                   -->

            <IdleAckInterval>512</IdleAckInterval>

            <!-- This specifies the default mime type header sent on tunnel  -->

            <!-- responses. In general the server always uses the mime type  -->

            <!-- specified by the incoming requests. The server will use the -->

            <!-- following only if it could not determine the mime type from -->

            <!-- incoming requests.                                          -->

            <MimeType>application/x-fcs</MimeType>

            <!-- This specifies the size (in KB) of the write buffer. The    -->

            <!-- default is 16Kb.                                            -->

            <WriteBufferSize>16</WriteBufferSize>

        </HTTPTunnel>

        <!-- This specifies the max size of messages for screen-sharing      -->

        <!-- video packets.  Messages larger than this limit will be broken  -->

        <!-- into multiple messages, which can allow for updates to appear   -->

        <!-- faster but also creates partial image updates.  The size is     -->

        <!-- bytes.  Set to zero for to not alter screen-sharing messages    -->

        <MaxMessageSizeLosslessVideo>0</MaxMessageSizeLosslessVideo>

        <!-- Specifies the RTMP chunk size to use in all streams for this    -->

        <!-- application.  Stream content breaks into chunks of this size    -->

        <!-- in bytes.  Larger values reduce CPU usage, but also commit to    -->

        <!-- larger writes that can delay other content on lower bandwidth    -->

        <!-- connections.  This can have a minimum value of 128 (bytes) and    -->

        <!-- a maximum value of 65536 (bytes) with a default of 4096 bytes    -->

        <!-- Note that older clients may not support chunk sizes largee than    -->

        <!-- 1024 bytes. If the chunk setting is larger than these clients can    -->

        <!-- support, the chunk setting will be capped at 1024 bytes.        -->

        <OutChunkSize>4096</OutChunkSize>

        <!-- An application can be configured to deliver aggregate messages to       -->

        <!-- clients that support them by setting the "enabled" attribute to "true". -->

        <!-- The server will attempt to send aggregate messages to these supported   -->

        <!-- clients based whenever possible.                                        -->

        <!-- When this setting is disabled, aggregate messages will always be broken -->

        <!-- up into individual messages before being delivered to clients.          -->

        <!--  The default is "true".                                     -->

        <AggregateMessages enabled="false"></AggregateMessages>

        <!-- Controls libconnect.dll access configurations             -->

        <Access>

            <!-- This controls if readAccess and writeAccess use the format  -->

            <!-- where folder level permissions are enforced         -->

            <!-- if true, individual files cannot be marked specifically for -->

            <!-- read or write access. Default is false, new format allows   -->

            <!-- access control at the file level                 -->

            <FolderAccess>false</FolderAccess>

            <!-- This tag specifies a semicolon delimited list of folder     -->

            <!-- levels which are given with audio sample access.  By        -->

            <!-- default, this tag is disabled and the value is empty which  -->

            <!-- means no one will have audio access.  This tag can be       -->

            <!-- enabled by setting "enabled" to "true".                     -->

            <!-- When enabled, everyone have audio sample access to the      -->

            <!-- access folders specified in the list.  If value is set to   -->

            <!-- to "/", full sample access will be granted.  User can also  -->

            <!-- override this setting by using the access adaptor or server -->

            <!-- side script.                                                -->

            <AudioSampleAccess enabled="false"></AudioSampleAccess>

            <!-- This tag specifies a semicolon delimited list of folder     -->

            <!-- levels which are given with video sample access.  By        -->

            <!-- default, this tag is disabled and the value is empty which  -->

            <!-- means no one will have video access.  This tag can be       -->

            <!-- enabled by setting "enabled" to "true".                     -->

            <!-- When enabled, everyone have video sample access to the      -->

            <!-- access folders specified in the list.  If value is set to   -->

            <!-- to "/", full sample access will be granted.  User can also  -->

            <!-- override this setting by using the access adaptor or server -->

            <!-- side script.                                                -->

            <VideoSampleAccess enabled="false"></VideoSampleAccess>

        </Access>

        <!-- Bit 0x01 represents clients will get the silence message if the -->

        <!-- client falls within the range specified.                        -->

        <UserAgent override="yes">

            <Bits from="WIN 6,0,0,0" to="WIN 7,0,55,0">0x01</Bits>

            <Bits from="MAC 6,0,0,0" to="MAC 7,0,55,0">0x01</Bits>

        </UserAgent>

    </Client>

    <Debug>

        <!-- This specifies the max number of pending debug connections.     -->

        <!-- This is by default 50.  If this is 0, then debugging         -->

        <!-- connections are disabled.  Once this number is met, the         -->

        <!-- oldest pending debug connection will be rejected to make room   -->

        <MaxPendingDebugConnections>50</MaxPendingDebugConnections>

        <!-- This sets the default value for application.allowDebug          -->

        <!-- allowDebug is a gate allowing debug connections on a per app    -->

        <!-- basis.  The default value is false                     -->

        <AllowDebugDefault>false</AllowDebugDefault>

    </Debug>

    <HTTP>

        <!-- Specifies whether or not to use HTTP 1.0. Default is false.     -->

        <HTTP_1_0>false</HTTP_1_0>

        <!-- Specifies whether or not to output verbose information during   -->

        <!-- operation. This is useful for debugging.                        -->

        <Verbose>false</Verbose>

        <Connections>

            <!-- Specifies max time (in secs.) that a transfer can take to   -->

            <!-- complete. Default is 60 secs. Since operations such as DNS  -->

            <!-- lookups may take some time, setting this to a low value     -->

            <!-- runs the risk of aborting perfectly normal operations.      -->

            <MaxTimeout>60</MaxTimeout>

            <!-- Specifies whether or not to explicitly close the connection -->

            <!-- after every transfer. Default is to reuse connections. Set  -->

            <!-- this to false to explicitly use a new connection each time. -->

            <Reuse>true</Reuse>

            <!-- Specifies the interface name to use as the outgoing network -->

            <!-- interface. The name can be an interface name, an IP address -->

            <!-- or a host name.                                             -->

            <Interface></Interface>

        </Connections>

        <Proxy>

            <!-- Specifies the HTTP proxy to use. This can be the hostname   -->

            <!-- a dotted IP address. To specify port number in this string, -->

            <!-- append :[port] to the end of the host name. The port number -->

            <!-- may optionally be specified separately in the <Port> tag.   -->

            <Host></Host>

            <!-- Specifies the proxy port to connect to if not specified as  -->

            <!-- part of the host in <Host>.                                 -->

            <Port></Port>

            <!-- Specifies the type of proxy being connected to. This can be -->

            <!-- This can be HTTP or SOCKS5. Default is HTTP.                -->

            <Type>HTTP</Type>

            <!-- Specifies whether or not to tunnel all operations through a -->

            <!-- given HTTP proxy. Default is false.                         -->

            <Tunnel>false</Tunnel>

            <!-- Specifies the username to use for connecting to the proxy.  -->

            <Username></Username>

            <!-- Specifies the password to use for connecting to the proxy.  -->

            <Password></Password>

        </Proxy>

        <Redirect>

            <!-- Specifies whether or not to allow following any Location:   -->

            <!-- header that is sent as part of the HTTP header. Default is  -->

            <!-- true; allow redirects.                                      -->

            <Allow>true</Allow>

            <!-- Specifies the max number of redirects allowed.              -->

            <Max>10</Max>

            <!-- Specifies whether or not to allow sending username/password -->

            <!-- again with each redirect. This is only useful if redirects  -->

            <!-- are allowed as defined by <Allow>. Default is true.         -->

            <UnrestrictedAuth>true</UnrestrictedAuth>

        </Redirect>

    </HTTP>

    <!-- SWFVerification configures how a SWF connecting to this app is verified. -->

    <!-- If enabled, all connections to this app will require verification, with  -->

    <!-- some exceptions which is configured in <UserAgentExceptions>.            -->

  <!-- This setup is completely independent of the Administrator level SWF      -->

  <!-- verification that may be found in Server.xml.  When SWFVerification is   -->

  <!-- enabled, it first looks to the settings here and only upon failure to    -->

  <!-- verify will the independent Adminstrator level SWF Verification settings -->

  <!-- be consulted.                                                            -->

    <SWFVerification enabled="false">

        <!-- Any SWFs found in this folder will be authenticated for connecting -->

        <!-- to this application on this server.  You can use a semicolon to     -->

        <!-- seperate multiple directories here as well. For example         -->

        <!-- C:\apps\app1\SWFs;D:\apps\app1\SWFs will allow swfs from either    -->

        <!-- of these two directories to connect.  SWFs located right at this     -->

        <!-- directory location will be allowed to connect to any application     -->

        <!-- instance.  Those sWFs located under an instance named folder will     -->

        <!-- be allowed to connect only to that specific instance.  The default -->

        <!-- here is the application's folder appended with /SWFs - thus if my     -->

        <!-- application is located at C:\applications\foo1 then the default SWF-->

        <!-- verification folder is C:\applications\foo1\SWFs            -->

        <SWFFolder></SWFFolder>

        <!-- This specifies the minimum accepted version of SWFVerification allowed by the server.    -->

        <!-- Initial version is 1.                                                                      -->

        <MinGoodVersion></MinGoodVersion>

        <!-- This specifies the number of levels for subfolders in SWFFolder to scan          -->

        <!-- for SWFs files. Negative value means all existing subfolders level are scanned.       -->

        <!-- Default is 1 which means the only one subfolder level is scanned.                     -->

        <DirLevelSWFScan>1</DirLevelSWFScan>

        <!-- The maximum time (sec) for processing swf files. Default value is 5 sec -->

        <MaxInitDelay>5</MaxInitDelay>

        <!-- The maximum allowable time, before disconnect, that a client may     -->

        <!-- take to provide its first verification hash to FMS.        -->

        <!-- Default is 5 seconds.  Set -1 for no timeout.             -->

        <!-- Note: Adjusting this time lower than 5 seconds risks disconnecting    -->

        <!-- valid users with slower connections                -->

        <FirstHashTimeout>5</FirstHashTimeout>

        <!-- The maximum allowable time, before disconnect, that a client may     -->

        <!-- take to provide its final verification hash to FMS.  Final hash is for -->

        <!-- the completely downloaded SWF file.                -->

        <!-- Default is 60 seconds.  Set -1 for no timeout.            -->

        <!-- Note: Adjusting this time lower than 60 seconds risks disconnecting-->

        <!-- valid users with slower connections.                -->

        <FinalHashTimeout>60</FinalHashTimeout>

        <!-- If it is necessary to allow certain types of connection to bypass    -->

        <!-- the requirement for SWFVerification, then those types of connection    -->

        <!-- should have their distinguishing UserAgent type added to this     -->

        <!-- configuration.  To add an exception to the SWFVerification rule one    -->

        <!-- inserts an Exception block within the UserAgentExceptions tag.    -->

        <!-- By default there are no exceptions.                -->

        <UserAgentExceptions>

            <!-- This is an example exception, UserAgents are specified in    -->

            <!-- a range.  An empty range, such as in the example below is    -->

            <!-- disregarded.  Ranges specify a string name and a version     -->

            <!-- number.  If one wanted to allow Flash Media Encoder to     -->

            <!-- be an exception, its UserAgent is FME/1.0 at time of this    -->

            <!-- writing.  Thus one could specify From="FME/0.0" and      -->

            <!-- To="FME/1.0" allowing the range FME falls into.            -->

            <!-- Another example using a specific Flash Player        -->

            <!-- will report WIN 9,0,28,0 as its UserAgent.  One may specify-->

            <!-- From="WIN 9,0,28,0" and To="WIN 9,0,28,0" and only that    -->

            <!-- specific version will be an exception.  One may list     -->

            <!-- as many exceptions as necessary, and they may overlap in     -->

            <!-- range as needed.  Users should take great care in allowing    -->

            <!-- exceptions as those listed here are not verified via SWF    -->

            <!-- verification at all.                                            -->

            <Exception from="" to=""/>

        </UserAgentExceptions>

        <!-- SWFVerification data are cache in the server.  This configures how the cache behaves.  -->

        <Cache>

            <!-- TTL is the time to live of each SWF in the cache.  Specified in minutes.       -->

            <!-- Default value is 1440 minutes (24 hours)                                       -->

            <TTL>1440</TTL>

            <!-- The maximum time to wait for the server to scan the SWF folders for updates    -->

            <!-- when there is a miss in the cache.  Specified in minutes.  Default value is    -->

            <!-- 5 minutes.                                                                     -->

            <UpdateInterval>5</UpdateInterval>

        </Cache>

    </SWFVerification>

    <!-- This section controls some of the logging behaviors.                      -->

    <Logging>

        <!-- This section controls access log.                                 -->

        <Access>

            <!-- You can override the settings specified in Vhost.xml for  -->

            <!-- this application by uncommenting the tag below. You can   -->

            <!-- disable logging checkpoints for this application, or      -->

            <!-- change the checkpoint interval of this application.  If   -->

            <!-- the checkpoint interval is not specified here, or invalid -->

            <!-- (i.e. <= 0), then we use whatever was set at the vhost    -->

            <!-- level.                                                    -->

            <!--

            <Checkpoints enable="false">

                <LogInterval>3600</LogInterval>

            </Checkpoints>

            -->

        </Access>

    </Logging>

    <!-- Specifes which protocols cannot be used to connect to this app. -->

    <!-- Protocols can be rtmp, rtmps, or rtmpe.  Protocols are specified -->

    <!-- in a comma delimited list.  Any protocol not specified is allowed. -->

    <!-- Example, <DisallowedProtocols>rtmp,rtmps</DisallowedProtocols> -->

    <!-- disallows rtmp and rtmps and only allows clients to connect via -->

    <!-- rtmpe.  Default is to allow all protocols. -->

    <DisallowedProtocols></DisallowedProtocols>

</Application>

Application Specific file xml :

<Application>

    <Process>

    <!-- The process scope in which instances of this application will  -->

    <!-- run. An application instance may run by itself in a process,   -->

    <!-- along with other instances or even be distributed among        -->

    <!-- multiple processes. The process scope may be one of adaptor,   -->

    <!-- vhost, app or inst, with the default being vhost. These values -->

    <!-- determine how various application instances are grouped into   -->

    <!-- processes, in the following manner,                            -->

    <!--                                                                -->

    <!--    adaptor       all instances in an adaptor run in a process  -->

    <!--    vhost         all instances in a vhost run in a process     -->

    <!--    app              all instances in an app run in a process      -->

    <!--    inst          each instance runs in a process               -->

    <Scope></Scope>

    <!-- Whether to distribute the above process scope into multiple    -->

    <!-- processes. One can distribute vhosts, applications, instances  -->

    <!-- or even clients across the process group. These entities have  -->

    <!-- a complete enclosing relationship or a strict ordering among   -->

    <!-- them, for e.g. adaptors contain vhosts, which contain apps     -->

    <!-- etc. The complete ordering is,                                 -->

    <!--                                                                -->

    <!--   . Adaptors                                                   -->

    <!--   . VHosts                                                     -->

    <!--   . Applications                                               -->

    <!--   . Instances                                                  -->

    <!--   . Clients                                                    -->

    <!--                                                                -->

    <!-- The entity chosen for <Distribute> tag must be lower or        -->

    <!-- contained in the process scope specified in the <Scope> tag,   -->

    <!-- and is specified as vhosts, apps, insts and clients. In other  -->

    <!-- words, if the process scope is set to adaptor, one can choose  -->

    <!-- vhosts, apps, insts, or clients. On the other hand, if the     -->

    <!-- process scope is set to "app" then one can specify only insts  -->

    <!-- or clients for distribution.                                   -->

    <!--                                                                -->

    <!-- Choosing to distribute clients indicates a stateless           -->

    <!-- application, wherein all clients need not rendezvous on a      -->

    <!-- single instance. Typical VOD (Video on Demand) applications    -->

    <!-- are stateless, as each user plays some content independent of  -->

    <!-- all other instances. On the other hand a chat or gaming        -->

    <!-- application is not stateless as all participants need to       -->

    <!-- rendezvous on a single instance wherein they share the         -->

    <!-- application state. By default, entities are distributed over   -->

    <!--  3 cores, which maximizes the server's ability to utilize      -->

    <!--  system resources.  Distribution may be turned off by setting  -->

    <!-- numprocs to 0 or 1. Also by default the server chooses to      -->

    <!--  distribute the immediately lower entity to the one specified  -->

    <!--  in <Scope> tag.                                               -->

    <Distribute numprocs="3">Instances</Distribute>

    <!-- The lifetime of a core process can be specified to not exceed  -->

    <!-- a certain period, wherein any connections after this will      -->

    <!-- result in rolling over to a new core process. The <RollOver>   -->

    <!-- tag determines this time period which is specified in seconds. -->

    <!-- For stateless applications, the older core process continues   -->

    <!-- to serve the earlier connections. In this case one can limit   -->

    <!-- the maximum number of core processes that may be running       -->

    <!-- simultaneously by specifying an appropriate value in the       -->

    <!-- <MaxCores> tag.                                                -->

    <!--                                                                -->

    <!-- An application is considered to be stateless if it is          -->

    <!-- configured to distribute clients over multiple processes. This -->

    <!-- may be done explicitly specifying "clients" for distribution,  -->

    <!-- or implicitly by choosing "inst" process scope, of course      -->

    <!-- distribution must be enabled in the first place by setting     -->

    <!-- numprocs to a value greater than 1. If the application is not  -->

    <!-- stateless, the server ignores the <MaxCores> value and forces  -->

    <!-- it to 1. This ensures that such an application instance is not -->

    <!-- split across multiple processes, but results in clients        -->

    <!-- getting disconnected on a periodic basis. As such this is      -->

    <!-- mostly useful for stateless applications, and may have limited -->

    <!-- applicability otherwise.                                       -->

    <!--                                                                -->

    <!-- Process rollover may be disabled by setting the rollover time  -->

    <!-- to 0 and is the default. Limiting the number of core processes -->

    <!-- may be disabled by setting it to 0 and is the default too.     -->

    <LifeTime>

      <RollOver></RollOver>

      <MaxCores></MaxCores>

    </LifeTime>

    <!-- Determines the maximum number of process failures that will be -->

        <!-- tolerated before a core process is disabled. Once disabled the -->

        <!-- the master will not launch a core process till some minimum    -->

        <!-- recovery time elapses. This may be used to guard against a DOS -->

        <!-- behaviour where a faulty core consumes all cpu time by being   -->

        <!-- repeatedly launched very quickly. The recovery time is         -->

        <!-- specified in seconds, and setting it to 0 disables checking    -->

        <!-- for process failures. Note loading an application via admin    -->

        <!-- tools or APIs bypasses this check.                             -->

        <MaxFailures>2</MaxFailures>

        <RecoveryTime>0</RecoveryTime>

    </Process>

    <!-- Specifies whether or not to load this app when the server starts. -->

    <LoadOnStartup>false</LoadOnStartup>

    <!-- Max time an app instance can be idle (no clients connected)  -->

    <!-- before being unloaded. Note: an app instance only goes idle  -->

    <!-- after the last client disconnects. If the instance is loaded -->

    <!-- with no clients connected, it is not considered to be idle.  -->

    <!-- The max idle time is specified in seconds. (20 minutes)      -->

    <MaxAppIdleTime>1200</MaxAppIdleTime>

  <!-- You can override the settings specified in Vhost.xml for  -->

  <!-- this vhost by uncommenting the tag below. You can disable -->

  <!-- auto-close idle clients for this app, or change the max   -->

  <!-- idle time for clients connected to this vhost. If the max -->

  <!-- idle time is not specified here, or invalid (i.e. <= 0),  -->

  <!-- then we use whatever was set at the vhost level.          -->

  <!--

  <AutoCloseIdleClients enable="false">

    <MaxIdleTime>3600</MaxIdleTime>

  </AutoCloseIdleClients>

  -->

  <!-- This section contains information about configuring the script engine. -->

    <JSEngine>

        <!-- This specifies the max size (Kb.) the runtime can grow to before -->

        <!-- garbage collection is performed.                                 -->

        <RuntimeSize>64</RuntimeSize>

        <!-- This specifies the max number that the server will skip the GC   -->

        <!-- when the JS engine is busy.  By default, the server will only    -->

        <!-- do the GC when the JS engine is not busy.  If MaxGCSkipCount is  -->

        <!-- set to 0, the server will force a GC regardless of the state of  -->

        <!-- the JS engine.  And if MaxGCSkipCount is positive, the server    -->

        <!-- will force a GC when the number of skip count exceeds the value  -->

        <!-- in MaxGCSkipCount.                                               -->

        <MaxGCSkipCount>-1</MaxGCSkipCount>

        <!-- This specifies the max amount of time (in seconds) that a script -->

        <!-- function can take to execute. If it takes longer than this max   -->

        <!-- timeout period, then the script is considered "runaway", and     -->

        <!-- script execution will be terminated. By default, this value is 0 -->

        <!-- which means there is no timeout and we do not check for runaway  -->

        <!-- scripts. Normally you would only want to check for this in a     -->

        <!-- development environment while debugging and testing apps. In a   -->

        <!-- production environment, presumably, your apps have been tested   -->

        <!-- and should not go into any infinite loops (or become "runaway"). -->

        <MaxTimeOut>0</MaxTimeOut>

        <!-- FMS provides the load API which allows loading other script        -->

        <!-- files within main.asc. File specified is first resolved using the  -->

        <!-- main.asc, if the load file is not found in that location, path     -->

        <!-- specified in this tag is used to resolve the file. This tag can    -->

        <!-- contain multiple and they must be separated by a ';' Resolution is -->                 

        <!-- is done left to right                                              -->

        <ScriptLibPath>${APP.JS_SCRIPTLIBPATH}</ScriptLibPath>

        <!-- This section specifies virtual directory mappings for file      -->

        <!-- objects in the script. By using virtual directories, you can    -->

        <!-- specify file directories for different applications. If the     -->

        <!-- beginning portion of a file path matches the virtual directory  -->

        <!-- that is specified, then the storage location of the file maps   -->

        <!-- to the location specified by the virtual directory mapping. To  -->

        <!-- specify a virtual directory mapping, you first specify the      -->

        <!-- virtual directory, followed by a semi-colon, followed by the    -->

        <!-- actual file location. <virtual dir>;<actual dir>                -->

        <FileObject>           

            <VirtualDirectory>/FmsData;D:\FmsData\</VirtualDirectory>

        </FileObject>

        <XMLSocket>

            <!-- XMLSocket will close the connection if it receives this much -->

            <!-- data from XML server without receiving an end tag. This can  -->

            <!-- be overridden by each XMLSocket by specifying the proeprty,  -->

            <!-- XMLSocket.maxUnprocessedChars. However, this cannot exceed   -->

            <!-- what is specified here.                                      -->

            <MaxUnprocessedChars>4096</MaxUnprocessedChars>

        </XMLSocket>

        <NetConnection>

            <!-- Default object encoding to use for SSAS NetConnection. This  -->

            <!-- can be AMF0 or AMF3.  The default is AMF3.  It can be        -->

            <!-- overridden for each individual NetConnection by setting the  -->

            <!-- NetConnection.objectEncoding property to either 0 for AMF0,  -->

            <!-- or 3 for AMF3.                                               -->

            <ObjectEncoding>AMF3</ObjectEncoding>

        </NetConnection>

    </JSEngine>

    <StreamManager>

        <!-- Specifies the physical location where recorded streams are stored. -->

        <!-- For example, <StorageDir>c:\myapp\streams\</StorageDir>            -->

        <StorageDir>D:\FmsData\</StorageDir>

        <!-- Enables mapping of virtual stream paths to different physical locations. -->

        <!-- This tag is applicable when using the Authorization plug-in.  -->

        <QualifiedStreamsMapping enable="false" />

        <!-- Specifies the physical location where streams are duplicated           -->

        <!-- This location serves as a backup for Stream files                   -->

        <!-- This location must already exist, and when a stream            -->

        <!-- is copied here, it will be categorized by instance name by default        -->

        <!-- To include application name in the file path created the tag        -->

        <!-- change the appName attribute to "true", if not "false"            -->

        <!-- Example<DuplicateDir appName="true">c:\backupStreams</DuplicateDir>    -->

        <DuplicateDir appName="true"></DuplicateDir>

        <!-- Specifies the cache prefix that is passed from the origin server to the      -->

        <!-- proxy server.  This is used by the proxy server as a relative path to       -->

        <!-- locate the cache file in the CacheDir                                       -->

        <!-- The type can be set to "path" or "sname".  If type="path", the physical     -->

        <!-- path of the recorded stream will be appended to the prefix.                 -->

        <!-- If type = "sname", the stream name will be appended to the prefix           -->

        <!-- The prefix can be any text with/without preset paramaters, all parameters   -->

        <!-- are sourrounded by ?.  Everything within the ? will be substituted by the   -->

        <!-- server.                                                                     -->

        <!--                                                                             -->

        <!-- ?IP?      represents the IP address of the server.                           -->

        <!-- ?APP?     represents the application name.                                   -->

        <!-- ?APPINST? represents the application instance.                               -->

        <!-- ?VHOST?   represents the vhost name.                                         -->

        <!--                                                                             -->

        <!-- Since the prefix is used as a relative path in the proxy CacheDir, all      -->

        <!-- parameters should be separate by '/'.  For example, if type = "path" and    -->

        <!-- the recorded stream is located in c:/fms/flvs/foo.flv.                      -->

        <!-- data/?IP? will be resolved into data/xxx.xxx.xxx.xxx/c/fms/flvs/foo.flv     -->

        <!-- ?APPINST?/data will be resolved into app1/inst1/data/c/fms/flvs/foo.flv     -->

        <!-- origin1/data/ will be resolved into origin1/data/c/fms/flvs/foo.flv         -->

        <!--                                                                             -->

        <!-- By default, type is set to "path" and the prefix is set to ?IP?. The IP in  -->

        <!-- the prefix can be used to avoid file collision.  For example, the proxy     -->

        <!-- server might be connecting to two different origin servers with the same    -->

        <!-- file in c:\data\foo.flv.  And you want to keep the cache files separate in  -->

        <!-- the proxy server.  Similarly, if you want to share the cache file among     -->

        <!-- origin servers, you should not include the IP as parameter.                 -->

        <!-- Remember this prefix is a relative path used by proxy server to lookup the  -->

        <!-- cache stream file.  The prefix can be any name except some special          -->

        <!-- characters such as \, :, *, ?, ", <, >, |.                                  -->

        <CachePrefix type="path">?IP?</CachePrefix>

        <!-- Specifies the interval for updating the cache streaming in the proxy        -->

        <!-- server.  Value is in millisecond.  Minimum = 10 sec, maximum = 24 hours,    -->

        <!-- default = 10 min.                                                           -->

        <CacheUpdateInterval></CacheUpdateInterval>

        <MaxBufferRetries>128</MaxBufferRetries>

        <ThrottleBoundaryRequest enable="false">8</ThrottleBoundaryRequest>

        <ThrottleLoads enable="true">8</ThrottleLoads>

        <ThrottleDisplayInterval>64</ThrottleDisplayInterval>

        <!-- If there is no key frame at the point of seek server will perform     -->

        <!-- accurate seeking based on the nearest key and intermediate frames.    -->

        <!-- Default = true. The zone attribute can be set to "client", "server".  -->

        <!-- If zone is set to "client", no keyframe will be generated in the      -->

        <!-- the server, but all the information for accurate seeking will passed  -->

        <!-- and processed on the client side.                                     -->

        <!-- If zone is set to "server", a new keyframe will be created on the     -->

        <!-- server side based on the previous keyframe and intermediate frames.   -->

        <!-- If the zone attribute is missing, server will fall back to the old    -->

        <!-- behavior where enhanced seeking will be handled by the server for     -->

        <!-- Sorenson codec.  For the other video codec, enhanced will be handled  -->

        <!-- by the client.                                                        -->

        <EnhancedSeek zone="client">true</EnhancedSeek>

        <!-- Specifies the interval for saving generated keyframes in an .flv file  -->

        <!-- Value is in milliseconds.  Default = 60000 millisecond.  Minimum = 0   -->

        <!-- millisecond and maximum = 24 hours.  If value is unspecified or out of -->

        <!-- the range, it is set to 1000 millisecond.                              -->

        <KeyFrameInterval>60000</KeyFrameInterval>

        <!-- Force a GC if stream list grows over the specified value  -->

        <!-- Default = 100. Will be GC'd during application GC         -->

        <!-- interval.  If value is unspecified or -1, it means        -->

        <!-- unlimited.                                                -->       

        <MaxStreamsBeforeGC>100</MaxStreamsBeforeGC>

        <Audio>

            <CombineSamples>

                <!-- Combine samples if there are more than N subscribers.  -->

                <Subscribers>8</Subscribers>

                <!-- Stop combining samples if cpu is lower than this.      -->

                <LoCPU>60</LoCPU>

                <!-- Start combining samples if cpu is more than this.      -->

                <HiCPU>80</HiCPU>

                <!-- Combine N audio samples into one message.  Zero means  -->

                <!-- that audio is never collapsed.  Default is 0.          -->

                <MaxSamples>0</MaxSamples>

            </CombineSamples>

            <SendSilence>

                <!-- Specifies the interval (in milliseconds) for sending silence message -->

                <!-- when no audio is publishing to a live stream. Default is 3 second    -->

                <!-- Set to 0 to disable the silence message                              -->

                <!-- Silence message is used to support older version of Flash players.   -->

                <!-- The server will only send the silence message to clients which are   -->

                <!-- specified in <UserAgent> inside the <Client> section.  And bit 0x01  -->

                <!-- is used to control the silence message.                              -->

                <Interval></Interval>

            </SendSilence>

            <NotifyAudioStop enabled="false">

                <!-- Specifies how long to wait before the server notifies the client     -->

                <!-- when audio stops in the middle of a live/recorded stream.            -->

                <!-- Default is 3 second and minimum is 1 second.                         -->

                <Duration></Duration>

            </NotifyAudioStop>

        </Audio>

        <Live>

            <AccumulatedIFrames>

                <!-- Specifies the maximum duration (in second) of intermediate frames a  -->

                <!-- live stream can hold in the buffer.  The buffer contains a history   -->

                <!-- of the video messages up to the last keyframe so that clients are    -->

                <!-- able to catch up to the latest message even when they join in        -->

                <!-- between keyframes.  If the duration in the buffer is bigger than     -->

                <!-- MaxTime, server will clear the messages.  This setting is to avoid   -->

                <!-- the buffer from getting too big in the server and should be set      -->

                <!-- bigger than the keyframe interval.  Default value is set to -1 which -->

                <!-- means unlimited.                                                     -->

                <MaxTime>-1</MaxTime>

                <!-- Specifies the maximum size (in kilobytes) of intermediate frames a   -->

                <!-- live stream can hold in the buffer.  The buffer contains a history   -->

                <!-- of the video messages up to the last keyframe so that clients are    -->

                <!-- able to catch up to the latest message even when they join in        -->

                <!-- between keyframes.  If the size in the buffer is bigger than         -->

                <!-- MaxSize, server will clear the messages.  This setting is to avoid   -->

                <!-- the buffer from getting too big in the server and should be set      -->

                <!-- bigger than the total size of intermediate frame between keyframes.  -->

                <!-- Default value is set to -1 which means unlimited.                    -->

                <MaxSize>-1</MaxSize>

            </AccumulatedIFrames>

            <!-- A message queue is used to buffer the incoming messages from the publisher   -->

            <!-- so that the server can send messages in chunks to the subscribers.  Queuing  -->

            <!-- can be disabled so that individual messages will be sent out immediately to  -->

            <!-- the subscribers.  The "enabled" attribute can be set to "false" to disable   -->

            <!-- queuing.                                                                     -->

            <Queue enabled="true">

                <!-- Specifies how often the server will flush the message queue by size   -->

                <!-- (in bytes).   Set value to 0 also disable queuing.  Default is 4096.  -->

                <MaxQueueSize>4096</MaxQueueSize>

                <!-- Specifies how often the server will flush the message queue by time   -->

                <!-- (in milliseconds).   The default value is set to 500 milliseconds.    -->

                <MaxQueueDelay>500</MaxQueueDelay>

                <!-- Specifies whether the server will flush the message queue when a      -->

                <!-- data message arrives.  This is important when streaming only data     -->

                <!-- messages, so the server will send out the messages immediately.       -->

                <!-- Default is set to true.                                               -->

                <FlushOnData>true</FlushOnData>

                <!-- When queuing is enabled, messages in the queue can be combined to     -->

                <!-- form aggregate messages.  The "enabled" attribute will determine      -->

                <!-- whether aggregate messages will be created.  Default is "true".       -->

                <AggregateMessages enabled="false">

                    <!-- This setting determines the maximum size (in bytes) of  -->

                    <!-- aggregate messages created from the message queue when  -->

                    <!-- when aggregate messages are enabled.  Default is -1,    -->

                    <!-- which means that the aggregates can be as large as the  -->

                    <!-- queue size allows.                                      -->

                    <MaxAggMsgSize>-1</MaxAggMsgSize>

                </AggregateMessages>

            </Queue>

            <!-- Ping the existing publisher if another client tries to publish to the -->

            <!-- same live stream.  The value specifies how long (in milliseconds)     -->

            <!-- the server expects a response from the existing publisher.  If the    -->

            <!-- publisher fails to response within the timeout.  The server will      -->

            <!-- allow the second publisher to take over the live stream.  Default is  -->

            <!-- 2000 milliseconds.  And set value to -1 to disable ping.              -->

            <PublishTimeout>2000</PublishTimeout>

            <!-- This specifies the max latency (in milliseconds)of incoming messages  -->

            <!-- in a live stream.  When a data message is injected to a live stream   -->

            <!-- through the server side script, the max latency is used to determine  -->

            <!-- the timestamp of the message.  If the injection time is bigger than   -->

            <!-- the maximum latency, the time difference between the last message     -->

            <!-- time and the injection time will be added to the timestamp of the     -->

            <!-- data message.  Otherwise the last message time will be used.  Default -->

            <!-- is 500, and minimum is 0.                                             -->

            <MaxLatency>500</MaxLatency>

            <!-- When this is enabled, the server will start the stream timeline of a -->

            <!-- live stream as soon as it is published. This means that any delay    -->

            <!-- publish time and the time the first message arrives will be added to -->

            <!-- stream time as a fixed offset. This is useful if the publisher wishes-->

            <!-- to wait before sending messgages on a published stream, and wants    -->

            <!-- that delay to be accounted for in the stream timeline. This should   -->

            <!-- not be enabled when using DynamicStreaming to switch between live    -->

            <!-- streams, since this can cause the streasm to be appear out of sync.  -->

            <StartClockOnPublish>false</StartClockOnPublish>

            <!-- When this is disabled, the server will not erase any previously      -->

            <!-- recorded stream when a new publisher arrives in live mode.  By       -->

            <!-- default, this is enabled and the recorded stream will be erased.     -->

            <EraseOnPublish>true</EraseOnPublish>

        </Live>

        <!-- Whether we send status message NetStream.Play.Start for all the commands     -->

        <!-- including play, seek and unpause.  If false, only the play command will get  -->

        <!-- the status message                                                           -->

        <SendDuplicateStart>true</SendDuplicateStart>

        <!-- Controls commands we send OnMetaData with.  The value can be set to true,    -->

        <!-- false, or once.  These values determines when the onMetaData message is      -->

        <!-- sent, in the following manner,                                               -->

        <!--                                                                              -->

        <!--     true     Sends onMetaData for play and seek commands.                    -->

        <!--     false    Sends onMetaData for play only.                                 -->

        <!--     once     Falls back to FMS 1.x behavior, sends onMetaData based on the   -->

        <!--              start position regardless of the command.  If no onMetaData is  -->

        <!--              found from the start position, no onMetaData will be sent.      -->

        <!-- By default, the value is set to true.                                        -->

        <SendDuplicateOnMetaData>true</SendDuplicateOnMetaData>

        <Recording>

            <!-- The default maximum size (in KB) of a recorded file.    -->

            <!-- recording completely. -1 means that no maximum size is to be enforced. Default is -1. -->

            <MaxSize>-1</MaxSize>

            <!-- Sets a cap on the maximum size (in KB) of a recorded file.        -->

            <!-- Values set using the MaxSize field or SSAS will not be allowed to exceed this value.-->

            <!-- 0 disables recording completely. -1 means that no size cap is to be enforced. -->

            <!-- Default is -1. -->

            <MaxSizeCap>-1</MaxSizeCap>

            <!-- The default maximum duration (in seconds) of a recorded file.          -->

            <!-- 0 disables recording completely. -1 means that no maximum duration is to be enforced. -->

            <!-- Default is -1. -->

            <MaxDuration>-1</MaxDuration>

            <!-- Sets a cap on the maximum duration (in seconds) of a recorded file.    -->

            <!-- Values set using the MaxDuration field or SSAS will not be allowed to exceed this value.-->

            <!-- 0 disables recording completely. -1 means that no size cap is to be enforced. -->

            <!-- Default is -1. -->

      <MaxDurationCap>-1</MaxDurationCap>

  </Recording>

    </StreamManager>

    <SharedObjManager>

        <!-- Specifies the physical location where shared objects are stored -->

        <!-- on the server.                                                  -->

        <!-- For example, <StorageDir>c:\myapp\sharedobjects\</StorageDir>   -->

        <StorageDir>D:\FmsData\</StorageDir>

        <!-- Specifies the physical location where shared objects are duplicated       -->

        <!-- This location serves as a backup for SharedObject files                   -->

        <!-- This location must already exist, and when a shared object            -->

        <!-- is copied here, it will be categorized by instance name by default        -->

        <!-- To include application name in the file path created the tag        -->

        <!-- change the appName attribute to "true", if not "false"            -->

        <!-- Example<DuplicateDir appName="true">c:\backupSharedObjects</DuplicateDir>    -->

        <DuplicateDir appName="true"></DuplicateDir>

        <!-- Deleted slots will be purged if the current version - deleted version is greater -->

        <!-- than the value specified.  -->

        <ResyncDepth>-1</ResyncDepth>

        <!-- Specifies the timeout value in seconds to automatically unlock -->

        <!-- sharedobject if there is a client waiting for update. -1 to wait for ever. -->

        <LockTimeOut>-1</LockTimeOut>

        <!-- SharedObjects are automatically commited if they are dirty, this can be disabled -->

        <!-- for all shared objects within this instance by specifying false. SASS will have to call-->

        <!-- "save" or SharedObject.commit to persist the shared object else all data will be lost when the app shuts down-->

        <AutoCommit>true</AutoCommit>

        <!-- Maximum properties that can be set per shared object -->

        <!-- -1 is unlimited.  -->

        <MaxProperties>-1</MaxProperties>

        <!-- Each property within each shared object cannot exceed value specified in bytes. -->

        <!-- -1 is unlimited.  -->

        <MaxPropertySize>-1</MaxPropertySize>

    </SharedObjManager>

    <!-- Specifies whether or not to allow tunneling connections into this   -->

    <!-- application. The acceptable values are true, false, http1.1only and -->

    <!-- keepalive. The following describes the effect of each value,        -->

    <!--    true        allow all tunneling connections                      -->

    <!--    false       disallow all tunneling connections                   -->

    <!--    http1.1only allow only http 1.1 connections                      -->

    <!--    keepalive   allow http 1.1 or http 1.0 keep alive connections    -->

    <AllowHTTPTunnel>true</AllowHTTPTunnel>

    <Client>

        <!-- Use 10 if you have bandwidth values for FMS 2, 2 otherwise. -->

        <WindowsPerAck>2</WindowsPerAck>

        <Bandwidth override="yes">

            <!-- Specified in bytes/sec. The default provides about 20Mbps. -->

            <ServerToClient>38400</ServerToClient>

            <!-- Specified in bytes/sec -->

            <ClientToServer>89600</ClientToServer>

        </Bandwidth>

        <!-- Bandwidth cannot be set higher than the value set here. -->

        <BandwidthCap override="no">

            <!-- Specified in bytes/sec -->

            <ServerToClient>2500000</ServerToClient>

            <!-- Specified in bytes/sec -->

            <ClientToServer>2500000</ClientToServer>

        </BandwidthCap>

    <!-- Configure downstream (server to client) bw detection. -->

    <!-- To disable native bw detection and use script-based   -->

    <!-- bw detection, set the enabled attribute to false. By  -->

    <!-- default, native bw detection is enabled.              -->

    <BandwidthDetection enabled="true">

      <!-- Don't send data to client greater than this rate (in Kbps). -->

      <!-- Default of -1 means don't need to throttle; just send the   -->

      <!-- data at whatever rate is necessary to measure bw.           -->

      <MaxRate>-1</MaxRate>

      <!-- Amount of data sent to client. Server sends a series of random bytes -->

      <!-- to the client, each time sending this much more data. For example,   -->

      <!-- x bytes are sent, followed by 2x bytes, followed by 3x bytes, etc.   -->

      <!-- until <MaxWait> time has elapsed.                                    -->

      <DataSize>16384</DataSize>

      <!-- Specifies how long (in sec.) server will send data to the client.    -->

      <MaxWait>2</MaxWait>

    </BandwidthDetection>

        <MsgQueue>

            <Live>

                <!-- Drop live audio if audio q exceeds time specified. time in milliseconds -->

                <MaxAudioLatency>2000</MaxAudioLatency>

                <!-- Default buffer length in millisecond for live audio and video queue. -->

                <MinBufferTime>8000</MinBufferTime>

            </Live>

            <Recorded>

                <!-- Default buffer length in millisecond for live audio and video, value cannot be set below this by Flash player. -->

                <MinBufferTime>2000</MinBufferTime>

            </Recorded>

            <Server>

                <!-- Ratio of the buffer length used by server side stream -->

                <!-- to live buffer.  The value is between 0 and 1.  To    -->

                <!-- avoid break up of audio, the ratio should not be more -->

                <!-- than 0.5 of the live buffer.                          -->

                <BufferRatio>0.5</BufferRatio>

                <!-- This specifies whether outgoing messages are sent in       -->

                <!-- different priorities for server to server connection.  By  -->

                <!-- default, prioritization is set to false which means        -->

                <!-- messages are sent out through one channel with the same    -->

                <!-- priority.  If the value is set to true, messages will be   -->

                <!-- sent through multiple channels with different priorities.  -->

                <!-- And the priority is based on the message type.             -->

                <Prioritization>false</Prioritization>   

                <!-- The minimum number of video messages to queue at the start -->

                <!-- of a stream.  When an H264 stream or other pipelined codec -->

                <!-- stream begins it needs some number of messages, at most 64 -->

                <!-- to begin playback and this setting ensures that regardless -->

                <!-- of the buffer set there are enough messages at the start   -->

                <!-- to begin playback quickly.  Making this less than 64, down -->

                <!-- to 0 won't queue up extra messages but may cause H264      -->

                <!-- content with a low FramesPerSecond to delay before start   -->

                <MinQueuedVideo>64</MinQueuedVideo>

            </Server>

            <Diagnostic>

                <!-- Whether diagnostic logging specific to message queue is    -->

                <!-- enabled.  Default is false.                                -->

                <Enable>false</Enable>

            </Diagnostic>

        </MsgQueue>

        <HTTPTunnel>

            <!-- The following two parameters affect the latency observed by -->

            <!-- a client tunneling into the server. Low values will reduce  -->

            <!-- the latency but increase the network bandwidth overhead.    -->

            <!-- Applications desiring low latency may use 128/256 for idle  -->

            <!-- post and ack intervals, while those not susceptible to high -->

            <!-- latencies may use 1024/2048. The default setting provides   -->

            <!-- medium latency and is set at 512/512.                       -->

            <!-- This specifies the interval at which the client should send -->

            <!-- idle posts to the server to indicate that the player has no -->

            <!-- data to send. This is needed to enable the server to send   -->

            <!-- downstream data. This interval is specified in milliseconds -->

            <!-- and must be in the range 0 to 4064 msec.                    -->

            <IdlePostInterval>512</IdlePostInterval>

            <!-- This specifies the max time the server may wait before it   -->

            <!-- sends back an ack for a client idle post.                   -->

            <IdleAckInterval>512</IdleAckInterval>

            <!-- This specifies the default mime type header sent on tunnel  -->

            <!-- responses. In general the server always uses the mime type  -->

            <!-- specified by the incoming requests. The server will use the -->

            <!-- following only if it could not determine the mime type from -->

            <!-- incoming requests.                                          -->

            <MimeType>application/x-fcs</MimeType>

            <!-- This specifies the size (in KB) of the write buffer. The    -->

            <!-- default is 16Kb.                                            -->

            <WriteBufferSize>16</WriteBufferSize>

        </HTTPTunnel>

        <!-- This specifies the max size of messages for screen-sharing      -->

        <!-- video packets.  Messages larger than this limit will be broken  -->

        <!-- into multiple messages, which can allow for updates to appear   -->

        <!-- faster but also creates partial image updates.  The size is     -->

        <!-- bytes.  Set to zero for to not alter screen-sharing messages    -->

        <MaxMessageSizeLosslessVideo>0</MaxMessageSizeLosslessVideo>

        <!-- Specifies the RTMP chunk size to use in all streams for this    -->

        <!-- application.  Stream content breaks into chunks of this size    -->

        <!-- in bytes.  Larger values reduce CPU usage, but also commit to    -->

        <!-- larger writes that can delay other content on lower bandwidth    -->

        <!-- connections.  This can have a minimum value of 128 (bytes) and    -->

        <!-- a maximum value of 65536 (bytes) with a default of 4096 bytes    -->

        <!-- Note that older clients may not support chunk sizes largee than    -->

        <!-- 1024 bytes. If the chunk setting is larger than these clients can    -->

        <!-- support, the chunk setting will be capped at 1024 bytes.        -->

        <OutChunkSize>4096</OutChunkSize>

        <!-- An application can be configured to deliver aggregate messages to       -->

        <!-- clients that support them by setting the "enabled" attribute to "true". -->

        <!-- The server will attempt to send aggregate messages to these supported   -->

        <!-- clients based whenever possible.                                        -->

        <!-- When this setting is disabled, aggregate messages will always be broken -->

        <!-- up into individual messages before being delivered to clients.          -->

        <!--  The default is "true".                                     -->

        <AggregateMessages enabled="false"></AggregateMessages>

        <!-- Controls libconnect.dll access configurations             -->

        <Access>

            <!-- This controls if readAccess and writeAccess use the format  -->

            <!-- where folder level permissions are enforced         -->

            <!-- if true, individual files cannot be marked specifically for -->

            <!-- read or write access. Default is false, new format allows   -->

            <!-- access control at the file level                 -->

            <FolderAccess>false</FolderAccess>

            <!-- This tag specifies a semicolon delimited list of folder     -->

            <!-- levels which are given with audio sample access.  By        -->

            <!-- default, this tag is disabled and the value is empty which  -->

            <!-- means no one will have audio access.  This tag can be       -->

            <!-- enabled by setting "enabled" to "true".                     -->

            <!-- When enabled, everyone have audio sample access to the      -->

            <!-- access folders specified in the list.  If value is set to   -->

            <!-- to "/", full sample access will be granted.  User can also  -->

            <!-- override this setting by using the access adaptor or server -->

            <!-- side script.                                                -->

            <AudioSampleAccess enabled="false"></AudioSampleAccess>

            <!-- This tag specifies a semicolon delimited list of folder     -->

            <!-- levels which are given with video sample access.  By        -->

            <!-- default, this tag is disabled and the value is empty which  -->

            <!-- means no one will have video access.  This tag can be       -->

            <!-- enabled by setting "enabled" to "true".                     -->

            <!-- When enabled, everyone have video sample access to the      -->

            <!-- access folders specified in the list.  If value is set to   -->

            <!-- to "/", full sample access will be granted.  User can also  -->

            <!-- override this setting by using the access adaptor or server -->

            <!-- side script.                                                -->

            <VideoSampleAccess enabled="false"></VideoSampleAccess>

        </Access>

        <!-- Bit 0x01 represents clients will get the silence message if the -->

        <!-- client falls within the range specified.                        -->

        <UserAgent override="yes">

            <Bits from="WIN 6,0,0,0" to="WIN 7,0,55,0">0x01</Bits>

            <Bits from="MAC 6,0,0,0" to="MAC 7,0,55,0">0x01</Bits>

        </UserAgent>

    </Client>

    <Debug>

        <!-- This specifies the max number of pending debug connections.     -->

        <!-- This is by default 50.  If this is 0, then debugging         -->

        <!-- connections are disabled.  Once this number is met, the         -->

        <!-- oldest pending debug connection will be rejected to make room   -->

        <MaxPendingDebugConnections>50</MaxPendingDebugConnections>

        <!-- This sets the default value for application.allowDebug          -->

        <!-- allowDebug is a gate allowing debug connections on a per app    -->

        <!-- basis.  The default value is false                     -->

        <AllowDebugDefault>false</AllowDebugDefault>

    </Debug>

    <HTTP>

        <!-- Specifies whether or not to use HTTP 1.0. Default is false.     -->

        <HTTP_1_0>false</HTTP_1_0>

        <!-- Specifies whether or not to output verbose information during   -->

        <!-- operation. This is useful for debugging.                        -->

        <Verbose>false</Verbose>

        <Connections>

            <!-- Specifies max time (in secs.) that a transfer can take to   -->

            <!-- complete. Default is 60 secs. Since operations such as DNS  -->

            <!-- lookups may take some time, setting this to a low value     -->

            <!-- runs the risk of aborting perfectly normal operations.      -->

            <MaxTimeout>60</MaxTimeout>

            <!-- Specifies whether or not to explicitly close the connection -->

            <!-- after every transfer. Default is to reuse connections. Set  -->

            <!-- this to false to explicitly use a new connection each time. -->

            <Reuse>true</Reuse>

            <!-- Specifies the interface name to use as the outgoing network -->

            <!-- interface. The name can be an interface name, an IP address -->

            <!-- or a host name.                                             -->

            <Interface></Interface>

        </Connections>

        <Proxy>

            <!-- Specifies the HTTP proxy to use. This can be the hostname   -->

            <!-- a dotted IP address. To specify port number in this string, -->

            <!-- append :[port] to the end of the host name. The port number -->

            <!-- may optionally be specified separately in the <Port> tag.   -->

            <Host></Host>

            <!-- Specifies the proxy port to connect to if not specified as  -->

            <!-- part of the host in <Host>.                                 -->

            <Port></Port>

            <!-- Specifies the type of proxy being connected to. This can be -->

            <!-- This can be HTTP or SOCKS5. Default is HTTP.                -->

            <Type>HTTP</Type>

            <!-- Specifies whether or not to tunnel all operations through a -->

            <!-- given HTTP proxy. Default is false.                         -->

            <Tunnel>false</Tunnel>

            <!-- Specifies the username to use for connecting to the proxy.  -->

            <Username></Username>

            <!-- Specifies the password to use for connecting to the proxy.  -->

            <Password></Password>

        </Proxy>

        <Redirect>

            <!-- Specifies whether or not to allow following any Location:   -->

            <!-- header that is sent as part of the HTTP header. Default is  -->

            <!-- true; allow redirects.                                      -->

            <Allow>true</Allow>

            <!-- Specifies the max number of redirects allowed.              -->

            <Max>10</Max>

            <!-- Specifies whether or not to allow sending username/password -->

            <!-- again with each redirect. This is only useful if redirects  -->

            <!-- are allowed as defined by <Allow>. Default is true.         -->

            <UnrestrictedAuth>true</UnrestrictedAuth>

        </Redirect>

    </HTTP>

    <!-- SWFVerification configures how a SWF connecting to this app is verified. -->

    <!-- If enabled, all connections to this app will require verification, with  -->

    <!-- some exceptions which is configured in <UserAgentExceptions>.            -->

  <!-- This setup is completely independent of the Administrator level SWF      -->

  <!-- verification that may be found in Server.xml.  When SWFVerification is   -->

  <!-- enabled, it first looks to the settings here and only upon failure to    -->

  <!-- verify will the independent Adminstrator level SWF Verification settings -->

  <!-- be consulted.                                                            -->

    <SWFVerification enabled="false">

        <!-- Any SWFs found in this folder will be authenticated for connecting -->

        <!-- to this application on this server.  You can use a semicolon to     -->

        <!-- seperate multiple directories here as well. For example         -->

        <!-- C:\apps\app1\SWFs;D:\apps\app1\SWFs will allow swfs from either    -->

        <!-- of these two directories to connect.  SWFs located right at this     -->

        <!-- directory location will be allowed to connect to any application     -->

        <!-- instance.  Those sWFs located under an instance named folder will     -->

        <!-- be allowed to connect only to that specific instance.  The default -->

        <!-- here is the application's folder appended with /SWFs - thus if my     -->

        <!-- application is located at C:\applications\foo1 then the default SWF-->

        <!-- verification folder is C:\applications\foo1\SWFs            -->

        <SWFFolder></SWFFolder>

        <!-- This specifies the minimum accepted version of SWFVerification allowed by the server.    -->

        <!-- Initial version is 1; default is 0 which means to accept all forward versions.         -->

        <MinGoodVersion></MinGoodVersion>

        <!-- This specifies the number of levels for subfolders in SWFFolder to scan          -->

        <!-- for SWFs files. Negative value means all existing subfolders level are scanned.       -->

        <!-- Default is 1 which means the only one subfolder level is scanned.                     -->

        <DirLevelSWFScan>1</DirLevelSWFScan>

        <!-- The maximum time (sec) for processing swf files. Default value is 5 sec -->

        <MaxInitDelay>5</MaxInitDelay>

        <!-- The maximum allowable time, before disconnect, that a client may     -->

        <!-- take to provide its first verification hash to FMS.        -->

        <!-- Default is 5 seconds.  Set -1 for no timeout.             -->

        <!-- Note: Adjusting this time lower than 5 seconds risks disconnecting    -->

        <!-- valid users with slower connections                -->

        <FirstHashTimeout>5</FirstHashTimeout>

        <!-- The maximum allowable time, before disconnect, that a client may     -->

        <!-- take to provide its final verification hash to FMS.  Final hash is for -->

        <!-- the completely downloaded SWF file.                -->

        <!-- Default is 60 seconds.  Set -1 for no timeout.            -->

        <!-- Note: Adjusting this time lower than 60 seconds risks disconnecting-->

        <!-- valid users with slower connections.                -->

        <FinalHashTimeout>60</FinalHashTimeout>

        <!-- If it is necessary to allow certain types of connection to bypass    -->

        <!-- the requirement for SWFVerification, then those types of connection    -->

        <!-- should have their distinguishing UserAgent type added to this     -->

        <!-- configuration.  To add an exception to the SWFVerification rule one    -->

        <!-- inserts an Exception block within the UserAgentExceptions tag.    -->

        <!-- By default there are no exceptions.                -->

        <UserAgentExceptions>

            <!-- This is an example exception, UserAgents are specified in    -->

            <!-- a range.  An empty range, such as in the example below is    -->

            <!-- disregarded.  Ranges specify a string name and a version     -->

            <!-- number.  If one wanted to allow Flash Media Encoder to     -->

            <!-- be an exception, its UserAgent is FME/1.0 at time of this    -->

            <!-- writing.  Thus one could specify From="FME/0.0" and      -->

            <!-- To="FME/1.0" allowing the range FME falls into.            -->

            <!-- Another example using a specific Flash Player        -->

            <!-- will report WIN 9,0,28,0 as its UserAgent.  One may specify-->

            <!-- From="WIN 9,0,28,0" and To="WIN 9,0,28,0" and only that    -->

            <!-- specific version will be an exception.  One may list     -->

            <!-- as many exceptions as necessary, and they may overlap in     -->

            <!-- range as needed.  Users should take great care in allowing    -->

            <!-- exceptions as those listed here are not verified via SWF    -->

            <!-- verification at all.                                            -->

            <Exception from="" to=""/>

        </UserAgentExceptions>

        <!-- SWFVerification data are cache in the server.  This configures how the cache behaves.  -->

        <Cache>

            <!-- TTL is the time to live of each SWF in the cache.  Specified in minutes.       -->

            <!-- Default value is 1440 minutes (24 hours)                                       -->

            <TTL>1440</TTL>

            <!-- The maximum time to wait for the server to scan the SWF folders for updates    -->

            <!-- when there is a miss in the cache.  Specified in minutes.  Default value is    -->

            <!-- 5 minutes.                                                                     -->

            <UpdateInterval>5</UpdateInterval>

        </Cache>

    </SWFVerification>

    <!-- This section controls some of the logging behaviors.                      -->

    <Logging>

        <!-- This section controls access log.                                 -->

        <Access>

            <!-- You can override the settings specified in Vhost.xml for  -->

            <!-- this application by uncommenting the tag below. You can   -->

            <!-- disable logging checkpoints for this application, or      -->

            <!-- change the checkpoint interval of this application.  If   -->

            <!-- the checkpoint interval is not specified here, or invalid -->

            <!-- (i.e. <= 0), then we use whatever was set at the vhost    -->

            <!-- level.                                                    -->

            <!--

            <Checkpoints enable="false">

                <LogInterval>3600</LogInterval>

            </Checkpoints>

            -->

        </Access>

    </Logging>

    <!-- Specifes which protocols cannot be used to connect to this app. -->

    <!-- Protocols can be rtmp, rtmps, or rtmpe.  Protocols are specified -->

    <!-- in a comma delimited list.  Any protocol not specified is allowed. -->

    <!-- Example, <DisallowedProtocols>rtmp,rtmps</DisallowedProtocols> -->

    <!-- disallows rtmp and rtmps and only allows clients to connect via -->

    <!-- rtmpe.  Default is to allow all protocols. -->

    <DisallowedProtocols></DisallowedProtocols>

</Application>

[Guest]

Hi ShivenK, please find time to answer this post....

[ShivenK]

Hi Sahil,

I apologize for not getting back earlier.

I did look into this and verified that the functionality is working fine. You have set your ServertoClient value as 38400. Note that this value is in bytes per second. So essentially this value is 38.4 KBps, which is equal to 307.2 Kbps. This means that this is the bandwidth available from the Server to Client, PER CLIENT and not all clients in total.

Also, this does not mean that the client will be unable to view streams more than 307.2 Kbps in bitrate, but in fact the client will view the stream with some buffering as the bandwidth available for that connection is 307.2 Kbps.

Hope this clarifies.

Thanks,

Shiven

[Guest]

Hi ShivenK,

Thanks for your time once again!!!

I got what you elaborated about ServerToClient. I wanted to use this property in the following way:

1) Publisher publishes at bitrate say 400 kbps. If reciever has bandwidth less than 400 then reciever will surely face latency, right??.

2) I thought to set ServerToClient for such receivers as equal to their bandwidth capacity to avoid latency. Because i thought the server will not send data to receiver over that limit then.

But this does not work. The receiver gets the all 400 kbps not accounting the ServerToClient property.

Won't this property will work in this way?

Regards,

Sahil.

[ShivenK]

Hi Sahil,

If the sender can send data at a rate greater than 400 kbps and the receiver is only capable of streaming at 400kbps or less, then you will encounter some buffering on the receiver side.

This is how this property is designed to work.

Thanks,

Shiven

[Guest]

Thanks Shiven! I will try to use this property with this idea.

You are a great help.

Can you also help me on an another query that i have posted regarding edge-origin issue I am facing. I am sure you will be able to clarify that as well.

http://forums.adobe.com/message/5125634#5125634

Regards,

Sahil.

[ShivenK]

Sure Sahil.

Please allow me a couple of days to test this out. Dont hesitate to ping me with reminders 🙂

-Shiven

[Guest]

Hi Shiven,

Were you able to find time to have a look at this:
http://forums.adobe.com/message/5125634#5125634

Regards,

Sahil.