This discussion is archived
1 2 Previous Next 17 Replies Latest reply: Nov 16, 2012 12:37 PM by Mark Malakanov (user11181920) Go to original post RSS
  • 15. Re: Connection reset from JDBC to 11g
    Mark Malakanov (user11181920) Expert
    Currently Being Moderated
    >
    Thu Nov 15 15:06:40 CET 2012 Client_to_DB: IOException in loop PipedSocketStream Connection reset
    Thu Nov 15 15:06:40 CET 2012 Tunnels = 3
    Thu Nov 15 15:13:56 CET 2012 DB_to_Client: IOException in loop PipedSocketStream DB_to_Client: End of file in communication channel
    >

    Is this communication between Client and Wrapper or between Wrapper and Oracle?
    2) @user11181920 So, connection always is established, but then it is broken.
    Look into network settings of Solaris. May be it includes something that kills TCP connection due to idling after some period.
    only wrapper is too sensible for tcp/ip anomalies
    It sounds weird. Application cannot be "too sensible for tcp/ip anomalies" it opens a TCP Socket and it either works or it does not. All that "retransmissions when packet lost" done in TCP/IP stack and are transparent for the Application. If a failure happens within TCP Socket it will throw IOException to the application.

    As it advised above, check Java versions in Sol 10 and 11. If the Wrapper is "stable" since 2009 it may like old Java and dislike a new one.
    I can do tcpdump, SPAN, RSPAN or whatever, but unfortunately our problem is unpredictable. Fighting with 6-hours-network-capture-file is really ugly thing :) But I have no choice - tcpdump started, I'm waiting for an error.
    you should spool it to a file. then review it.
  • 16. Re: Connection reset from JDBC to 11g
    721832 Newbie
    Currently Being Moderated
    user11181920 wrote:
    It sounds weird. Application cannot be "too sensible for tcp/ip anomalies" it opens a TCP Socket and it either works or it does not. All that "retransmissions when packet lost" done in TCP/IP stack and are transparent for the Application. If a failure happens within TCP Socket it will throw IOException to the application.
    I know it is weird. For example - from my desktop I'll open ssh session to appsrv(with "wrapper"), rdp session to other host, and my app-via-warpper session to db.
    Then I'll do something nasty on network layer (eg. recalculating spanning-tree on switch or add/remove vlan from trunk) - rdp is of course ok, ssh is ok, but wrapper - sorry, connection lost.

    BTW. Today I've opened sqldev session and it is iddle for whole day without any errors. At the same time my app team mailed me that app is crashing due connection lost whole the time.

    Edited by: zawierta on 2012-11-16 10:20
  • 17. Re: Connection reset from JDBC to 11g
    Mark Malakanov (user11181920) Expert
    Currently Being Moderated
    Today I've opened sqldev session and it is iddle for whole day without any errors.
    That means JDBC connection to Oracle works well.
    That means Client-Wrapper connection does not work well.
    Which has nothing to do with Oracle DB or Oracle JDBC.
    At the same time my app team mailed me that app is crashing due connection lost whole the time.
    And what error message with stack(!) did they mail to you?
1 2 Previous Next

Legend

  • Correct Answers - 10 points
  • Helpful Answers - 5 points