ReadTime java sdk error only when traffic high(market open, close)

kgwaaa

just market open time or close time, in our consumer server above error happened...

not that time, working perfectly...

bandwidth of our provider server to refinitiv is 20Mbps..

is this problem??

Do i have to set compression type??

please help me..

here is our emaConfig file.

<?xml version="1.0" encoding="UTF-8"?>
<EmaConfig>

    <!-- ConsumerGroup provides set of detailed configurations to be used by named consumers-->
    <!-- Application specifies which configuration to use by setting OmmConsumerConfig::consumerName()-->
    <ConsumerGroup>
        <!-- DefaultConsumer parameter defines which consumer configuration is used by OmmConsumer-->
        <!-- if application does not specify it through OmmConsumerConfig::consumerName()-->
        <!-- first consumer on the ConsumerList is a DefaultConsumer if this parameter is not specified-->
        <DefaultConsumer value="WAIKER_CONSUMER_1"/>
        <ConsumerList>
            <Consumer>
                <!-- Name is mandatory-->
                <Name value="WAIKER_CONSUMER_1"/>
                <ChannelSet value="WAIKER_DATA_CHANNEL_1, WAIKER_DATA_CHANNEL_2"/>
                <!-- Channel is optional: defaulted to "RSSL_SOCKET + localhost + 14002"-->
                <!-- Channel or ChannelSet may be specified-->
<!--                <Channel value="WAIKER_DATA_CHANNEL_1"/>-->

                <!-- Dictionary is optional: defaulted to "ChannelDictionary"-->
<!--                <Dictionary value="Dictionary_1"/>-->
<!--                <Dictionary value="ChannelDictionary"/>-->
<!--                <XmlTraceToStdout value="0"/>-->
<!--                <LoginRequestTimeOut value="3000000"/>-->
                <!--
                    Specifies the length of time (in milliseconds) a stream waits to receive an ACK for outstanding post
                    before forwarding a negative acknowledgement to the application.
                    If set to 0, the Enterprise Message API will wait for a response indefinitely.
                -->
                <PostAckTimeout value="0"/>
                <!--
                    Specifies the maximum number of times the consumer and non interactive provider attempt to reconnect to a channel
                    when it fails. If set to -1, the consumer and non-interactive provider continually attempt to reconnect
                -->
                <ReconnectAttemptLimit value="-1"/>

                <ReconnectMaxDelay value="5000"/>
                <ReconnectMinDelay value="1000"/>
                <LoginRequestTimeOut value="2000"/>

                <!--
                     Specifies the maximum allowable number of on-stream posts waiting for an acknowledgment before OmmConsumer disconnects.
                     default : 100000
                -->
<!--                <MaxOutstandingPosts value="9000000"/>-->
<!--                <ReissueTokenAttemptInterval value="30000"/>-->
<!--                <ReissueTokenAttemptLimit value="-1"/>-->
<!--                <RequestTimeout value="480000"/> &lt;!&ndash; 8 min &ndash;&gt;-->
            </Consumer>
            <Consumer>
                <!-- Name is mandatory-->
                <Name value="WAIKER_CONSUMER_2"/>

                <!-- Channel is optional: defaulted to "RSSL_SOCKET + localhost + 14002"-->
                <!-- Channel or ChannelSet may be specified-->
                <Channel value="WAIKER_DATA_CHANNEL_2"/>

                <!-- Dictionary is optional: defaulted to "ChannelDictionary"-->
                <!--                <Dictionary value="Dictionary_1"/>-->
                <!--                <Dictionary value="ChannelDictionary"/>-->
<!--                <XmlTraceToStdout value="0"/>-->
<!--                <LoginRequestTimeOut value="3000000"/>-->
                <!--
                    Specifies the length of time (in milliseconds) a stream waits to receive an ACK for outstanding post
                    before forwarding a negative acknowledgement to the application.
                    If set to 0, the Enterprise Message API will wait for a response indefinitely.
                -->
                <PostAckTimeout value="0"/>
                <!--
                    Specifies the maximum number of times the consumer and non interactive provider attempt to reconnect to a channel
                    when it fails. If set to -1, the consumer and non-interactive provider continually attempt to reconnect
                -->
                <ReconnectAttemptLimit value="-1"/>

                <ReconnectMaxDelay value="5000"/>
                <ReconnectMinDelay value="1000"/>
<!--                <LoginRequestTimeOut value="960000"/>-->
                <LoginRequestTimeOut value="2000"/>

                <!--
                     Specifies the maximum allowable number of on-stream posts waiting for an acknowledgment before OmmConsumer disconnects.
                     default : 100000
                -->
<!--                <MaxOutstandingPosts value="9000000"/>-->

<!--                <ReissueTokenAttemptInterval value="30000"/>-->
<!--                <ReissueTokenAttemptLimit value="-1"/>-->
<!--                <RequestTimeout value="480000"/> &lt;!&ndash; 8 min &ndash;&gt;-->
            </Consumer>
        </ConsumerList>
    </ConsumerGroup>


    <!-- IProviderGroup provides set of detailed configurations to be used by named providers                -->
    <!-- Application specifies which configuration to use by setting OmmIProviderConfig::providerName()      -->
<!--    <IProviderGroup>-->
<!--        &lt;!&ndash; DefaultIProvider parameter defines which provider configuration is used by OmmProvider          &ndash;&gt;-->
<!--        &lt;!&ndash; if application does not specify it through OmmIProviderConfig::providerName()                   &ndash;&gt;-->
<!--        &lt;!&ndash; first provider on the IProviderList is a default provider if this parameter is not specified    &ndash;&gt;-->
<!--        <DefaultIProvider value="Provider_1"/>-->
<!--        <IProviderList>-->
<!--            <IProvider>-->
<!--                &lt;!&ndash; Name is mandatory                                                                       &ndash;&gt;-->
<!--                <Name value="Provider_1"/>-->
<!--                &lt;!&ndash; Server is optional: defaulted to "RSSL_SOCKET + 14002"                                  &ndash;&gt;-->
<!--                <Server value="Server_1"/>-->
<!--                &lt;!&ndash; Directory is optional: defaulted to AdminControl::UserControlEnum                       &ndash;&gt;-->
<!--                &lt;!&ndash; source directory configuration to use if OmmIProviderConfig::adminModel()               &ndash;&gt;-->
<!--                &lt;!&ndash; was set to ApiControlEnum                                                               &ndash;&gt;-->
<!--                &lt;!&ndash; this configuration also decides which dictionaries will be loaded at startup            &ndash;&gt;-->
<!--                &lt;!&ndash; this configuration may be overwritten by OmmIProviderConfig::addAdminMsg()              &ndash;&gt;-->
<!--&lt;!&ndash;                <Directory value="Directory_2"/>&ndash;&gt;-->

<!--                <ItemCountHint value="10000"/>-->

<!--                <ServiceCountHint value="10000"/>-->

<!--                <DispatchTimeoutApiThread value="500" />-->

<!--                <MaxDispatchCountApiThread value="500" />-->

<!--                <MaxDispatchCountUserThread value="500" />-->

<!--                <PipePort value="9009" />-->

<!--                <RefreshFirstRequired value="1" />-->
<!--                &lt;!&ndash; 추가)-->
<!--                        Default is 15000. OmmProvider waits for a response to a request before sending another request.-->
<!--                        the DICTIONARY domain will not send another request.-->
<!--                        If set to 0, the Message API waits for a response indefinitely.-->
<!--                 &ndash;&gt;-->
<!--                <RequestTimeout value="0"/>-->
<!--                &lt;!&ndash; 추가)-->
<!--                        Default is 15000. OmmProvider waits for a response to a request before sending another request.-->
<!--                        the DICTIONARY domain will not send another request.-->
<!--                        If set to 0, the Message API waits for a response indefinitely.-->
<!--                 &ndash;&gt;-->
<!--            </IProvider>-->
<!--        </IProviderList>-->
<!--    </IProviderGroup>-->

    <ChannelGroup>
        <ChannelList>
            <Channel>
                <Name value="WAIKER_DATA_CHANNEL_1"/>

                <!-- ChannelType possible values are:-->
                <!-- ChannelType::RSSL_SOCKET    - TCP IP connection type-->
                <ChannelType value="ChannelType::RSSL_SOCKET"/>

                <!-- CompressionType is optional: defaulted to None-->
                <!-- possible values: None, ZLib, LZ4-->
                <CompressionType value="CompressionType::None"/>
                <!--                <GuaranteedOutputBuffers value="5000"/>-->
                <GuaranteedOutputBuffers value="50000"/>
                <SysRecvBufSize value="50000"/>
                <SysSendBufSize value="50000"/>

                <!-- ConnectionPingTimeout is optional: defaulted to 30000-->
                <!--                <ConnectionPingTimeout value="30000000"/>-->
                <ConnectionPingTimeout value="10000"/> <!-- 10초 -->

                <!-- TcpNodelay is optional: defaulted to 1 -->
                <!-- possible values: 1 (tcp_nodelay option set), 0 (tcp_nodelay not set)-->
                <TcpNodelay value="1"/>
                <Host value="providerIP"/>
                <Port value="14002"/>
                <InitializationTimeout value="600"/> <!-- 10분 -->
            </Channel>
            <Channel>
                <Name value="WAIKER_DATA_CHANNEL_2"/>

                <!-- ChannelType possible values are:-->
                <!-- ChannelType::RSSL_SOCKET    - TCP IP connection type-->
                <ChannelType value="ChannelType::RSSL_SOCKET"/>

                <!-- CompressionType is optional: defaulted to None-->
                <!-- possible values: None, ZLib, LZ4-->
                <CompressionType value="CompressionType::None"/>
                <!--                <GuaranteedOutputBuffers value="5000"/>-->
                <GuaranteedOutputBuffers value="50000"/>
                <SysRecvBufSize value="50000"/>
                <SysSendBufSize value="50000"/>

                <!-- ConnectionPingTimeout is optional: defaulted to 30000-->
<!--                <ConnectionPingTimeout value="30000000"/>-->
                <ConnectionPingTimeout value="10000"/> <!-- 10초 -->

                <!-- TcpNodelay is optional: defaulted to 1 -->
                <!-- possible values: 1 (tcp_nodelay option set), 0 (tcp_nodelay not set)-->
                <TcpNodelay value="1"/>
                <Host value="Provider_Ip2"/>
                <Port value="14002"/>
                <InitializationTimeout value="600"/> <!-- 10분 -->
            </Channel>

        </ChannelList>
    </ChannelGrou

</EmaConfig>

umer.nalla

Hi @kgwaaa

From your mention of 20Mbps, I assume you are connecting to either our cloud-based Refinitiv Real-Time Optimised or one of our remote managed services such as Real-Time Managed Distribution Service.

Firstly, I would recommend you raise a Support ticket at My.Refinitiv for whichever service you are using. The support team could then check the server logs etc to help determine why your connection is being reset.

If it is happening during peak market activity periods such as open or close time. then it could be that your Consumer is what we call a 'slow consumer'. It is not consuming the data quickly enough so the buffer for your connection on the Refinitiv server overflows and the Refinitiv server disconnects you. This could be because your code is spending too much time in the OnRefresh and OnUpdate callbacks - which are executed in the API Thread context. This results in the API thread being starved and therefore unable to read the data promptly from the server.

The usual recommendations for Slow consumer are to reduce the amount of processing being done in the callback on the main API thread context - by offloading any processing to worker threads. Ypu can also use 'Horizontal Scaling' i..e more than one OmmConsumer instance to split the watchlist and processing between the OmmConsumer instances. This is assuming your PC has multi-cores to support the additional thread used by each OmmConsumer instance. See Consumer example 410 MP Horizontal scaling from the EMA Examples provided in the RT-SDK