RSSL Channel read failed

tao.yang
tao.yang Newcomer

What could possibly cause this error:

(ComponentName) RSSL_Adapter: (Severity) Error: RSSL Channel read failed on connection "RSSL_0x7ff2400dbfc0".  Channel will be closed.  (Internal debug info: "<Impl/ripcsrvr.c:5616> ripcRead() failed: (104)", errno: 104)

Any time this error happens, it always reconnects back.

16:00:14.352519 WARN  mds.mdm3.Mdm3Logger - RFA/ZZZ: Connection is down
16:00:14.355050 INFO mds.mdm3.Mdm3Logger - RFA/ZZZ: User XXX login success, Connection Down
16:00:14.355099 INFO mds.mdm3.Mdm3Logger - RFA/ZZZ: User XXX login success, User: XXX on RSSL_0x7ff2400dbfc0 StreamState: Open(Accepted) plus DataState: Suspect
16:00:14.431764 INFO mds.mdm3.Mdm3Logger - RFA/ZZZ: Connection is up
16:00:14.433272 INFO mds.mdm3.Mdm3Logger - RFA/ZZZ: User XXX login success, Login accepted by host YYY.
16:00:14.433321 INFO mds.mdm3.Mdm3Logger - RFA/ZZZ: User XXX login success, User: XXX on RSSL_0x7ff2400dbfc0 StreamState: Open(Accepted) plus DataState: Ok<br>

Could the heavy traffic slowing down the consumer be the culprit?

Tagged:

Best Answer

  • The error indicates that RFA's transport layer cannot read data from RSSL channel from the socket with system error
    104. System Error 104 should be ECONNRESET
    (Connection reset by peer). It indicates that the peer (server) has
    disconnected the connection.

    This could be a network or server issue so you need to contact
    the network or server team to verify the problem. The server should log the reason of the disconnection.

    The heavy traffic slowing down the consumer could also be the cause of the disconnection. The server such as Elektron or TREP component has a mechanism to cut a connection, once the consumer cannot consume data fast enough (read data from socket connection). In this case, you can verify if the application is able to handle the number of incoming events via the number return from the dispatch() function. It is the estimated number of pending event in the event queue at
    that time. If the number continues increasing during the heavy traffic, you need to make the processEvent() more efficient by moving slow or process-heavy code to another thread.

Answers

  • tao.yang
    tao.yang Newcomer

    Thanks very much for your reply.

    Sometimes the error is 11:

    [Mon Jan  9 16:00:06 2017]: (ComponentName) RSSL_Adapter: (Severity) Error: RSSL Channel read failed on connection "RSSL_0x7f50fa3a65c0".  Channel will be closed.  (Internal debug info: "<Impl/ripcsrvr.c:5616> ripcRead() failed: (11)", errno: 11)

    Does it mean something else?

    Where can I get the explanations of the all the error code? Is there any way I can see the Impl/ripcsrvr.c ?