EMA InteractiveProvider- frequent disconnects

abhiman.srivastava

Hi, I have a Interactive provider (EMA 3.2) that is trying to publish large volumes of data for internal consumption. It seems to work as expected but we do get disconnected a few times by the ADH servers it seems. This is what we get in our logs:

06Dec 14:17:02.895 [pool-6-thread-1] ERROR com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Warning

Text: Received ChannelDown event on client handle 1

Instance Name Provider_1_1

RsslReactor @6c783614

RsslChannel @3a5110ce

Error Id 0

Internal sysError 0

Error Location Reactor.processWorkerEvent

Error text Ping error for channel: Lost contact with connection...

loggerMsgEnd

06Dec 14:17:02.895 [pool-6-thread-1] WARN com.citi.xcs.common.accessor.ema.publisher.AbstractEMAPublishingClient - Item analytics to be closed is not found

06Dec 14:17:02.900 [pool-6-thread-1] ERROR com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Warning

Text: Received ChannelDown event on client handle 4

Instance Name Provider_1_1

RsslReactor @6c783614

RsslChannel @3c40f29f

Error Id 0

Internal sysError 0

Error Location Reactor.processWorkerEvent

Error text Ping error for channel: Lost contact with connection...

loggerMsgEnd

06Dec 14:17:35.657 [pool-6-thread-1] ERROR com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Warning

Text: Received ChannelDown event on client handle 5

Instance Name Provider_1_1

RsslReactor @6c783614

RsslChannel @3a5110ce

Error Id 0

Internal sysError 0

Error Location Reactor.processWorkerEvent

Error text Error initializing channel: errorId=0 text=Lost contact with connection...

loggerMsgEnd

06Dec 14:17:35.658 [pool-6-thread-1] INFO com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Info

Text: Received ChannelUp event on ClientHandle 6

Instance Name Provider_1_1

Component Version rfa7.6.0.L1.win.rrg 32-bit

loggerMsgEnd

06Dec 14:17:35.658 [pool-6-thread-1] ERROR com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Warning

Text: Received ChannelDown event on client handle 6

Instance Name Provider_1_1

RsslReactor @6c783614

RsslChannel @3c40f29f

Error Id 0

Internal sysError 0

Error Location null

Error text SocketChannel.read returned -1 (end-of-stream)

loggerMsgEnd

06Dec 14:17:35.659 [pool-6-thread-1] ERROR com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Warning

Text: Received ChannelDown event on client handle 7

Instance Name Provider_1_1

RsslReactor @6c783614

RsslChannel @3a5110ce

Error Id 0

Internal sysError 0

Error Location Reactor.processWorkerEvent

Error text Error initializing channel: errorId=0 text=Lost contact with connection...

loggerMsgEnd

06Dec 14:17:35.661 [pool-6-thread-1] ERROR com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Warning

Text: Received ChannelDown event on client handle 8

Instance Name Provider_1_1

RsslReactor @6c783614

RsslChannel @54ccd183

Error Id 0

Internal sysError 0

Error Location Reactor.processWorkerEvent

Error text Error initializing channel: errorId=0 text=Lost contact with connection...

loggerMsgEnd

06Dec 14:17:35.742 [pool-6-thread-1] INFO com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Info

Text: Received ChannelUp event on ClientHandle 9

Instance Name Provider_1_1

Component Version rfa7.6.0.L1.win.rrg 32-bit

loggerMsgEnd

06Dec 14:17:37.493 [pool-6-thread-1] INFO com.thomsonreuters.ema.access.OmmIProviderImpl - loggerMsg

ClientName: ServerChannelHandler

Severity: Info

Text: Received ChannelUp event on ClientHandle 10

Instance Name Provider_1_1

Component Version adh3.3.2.L1.linux.tis.rrg 64-bit

loggerMsgEnd

Can anyone please tell my what these errors means and why i am being disconnected?

zoya faberov

Hello @abhiman.srivastava,

The messages that you are showing are not sufficient, in my view, to provide any insight to the nature of the disconnects.

And tuning large-volume provider can get involving.

You are registered on the forum with you gmail address, so the moderators of this forum are not able to tell if you are a Refinitiv customer, or what type of infra you are connecting to.

Given this, will try my best to share some ideas.

1. Request the infrastructure logs for the approximate time of the disconnects, and try track the disconnects into the infra logs. If you are on hosted, the logs from Refintiv hosted, if you are on deployed infra, from your market data group. See if you can get any pointers as to the cause.

2. Analyse what the app was doing when the disconnects have occurred. Was it publishing more content, or in different pattern then at other times? If the app was doing something that was making it slow, the non-responsive app is likely to be disconnected.

3. Analyse if any network interrupts have occurred at the time of the disconnect, if any CPU or memory resources were overrun on the provider machine at any point of time, making provider slow to respond, and as a result, disconnected by the infra.

There is no simple answer, as many factors may contribute to the disconnect happening, the approach is to examine all possible contributing causes and to start narrowing them down.

I hope this helps.