| Author |
Message
|
| Shalini |
 Posted: Wed Oct 13, 2004 5:19 am Post subject: CLOSED_WAIT and TIME_OUT for netstat |
 |
|
Master
Joined: 30 Apr 2002
Posts: 224
Location: India
|
Hi,
Solais 8 MQ5.3 CSD5 we are getting following error
Getting log errors shown below.
Also when I do a
nestat -a | grep 1414 into Solaris box
I get lots of process in "CLOSED_WAIT" and some in "TIME_WAIT" state.
Why such error occurs this box was running fine for past 5 months suddeanly we got the error.
Please let me know the root cause if some one else have got the same error and how to avoid such error in future.
-------------------------------------------------------------------------------
10/13/2004 13:07:36
AMQ9202: Remote host 'frXXX62-X00 (10.X.5X.X6) (1414)' not available,
retry later.
EXPLANATION:
The attempt to allocate a conversation using TCP/IP to host
'frXXX62-X00 (10.X.5X.X6) (1414)' was not successful. However the
error may be a transitory one and it may be possible to successfully allocate a TCP/IP conversation later.
ACTION:
Try the connection again later. If the failure persists, record the error values and contact your systems administrator. The return code from TCP/IP is 10060 (X'274C'). The reason for the failure may be that this host cannot reach the destination host. It may also be possible that the listening program at host frXXX62-X00 (10.X.5X.X6) (1414)' was not running. If this is the case, perform the relevant operations to start the TCP/IP listening program, and try again.
Thanks
 |
|
| Back to top |
|
 |
| JLRowe |
| Posted: Wed Oct 13, 2004 5:32 am Post subject: |
|
|
Yatiri
Joined: 25 May 2002
Posts: 664
Location: South East London
|
You may have orphaned sockets on your channels, this happends when the receiver thinks the socket is still up, the sender has timed out and is trying to reconnect and start a new socket, the received rejects this as the socket is already up!
The soultion is to enable 'AdoptNewMCA' on each receiving queue manager, this tells the receiver to drop the existing socket when it receives a connect. |
|
| Back to top |
|
|
| Shalini |
| Posted: Wed Oct 13, 2004 11:33 pm Post subject: |
|
|
Master
Joined: 30 Apr 2002
Posts: 224
Location: India
|
Hi jlrowe,
Thanks for Reply.
Can anybody suggest any thing more or any other reason.
Please comment
|
|
| Back to top |
|
|
| Shalini |
| Posted: Tue Oct 19, 2004 11:24 pm Post subject: |
|
|
Master
Joined: 30 Apr 2002
Posts: 224
Location: India
|
Hi,
Please let us know, if the connections shown after implementing the AdoptNewMCA feature on the queue manager are the old connections which were existing before disconnecting or whether the connections are new connections.
|
|
| Back to top |
|
|
| Tibor |
| Posted: Thu Oct 21, 2004 3:17 pm Post subject: |
|
|
Grand Master
Joined: 20 May 2001
Posts: 1033
Location: Hungary
|
Shalini,
2nd scenario: you have a connecting client application and this can consume all the sockets . For example, when an MQ client reconnect on every messages 
In this case AdoptNewMCA won't help you because this use a SVRCONN.
Tibor |
|
| Back to top |
|
|
| Shalini |
| Posted: Mon Oct 25, 2004 12:54 am Post subject: |
|
|
Master
Joined: 30 Apr 2002
Posts: 224
Location: India
|
Hi,
How to get rid of that ???? |
|
| Back to top |
|
|
| abiram8 |
| Posted: Mon Oct 25, 2004 12:57 am Post subject: |
|
|
Master
Joined: 27 Mar 2002
Posts: 207
Location: India
|
Hi,
| Quote: |
| In this case AdoptNewMCA won't help you because this use a SVRCONN |
In such cases use "KeepAlive" Option
Turn KeepAlive on for the Queue Manager.
It will use the system value (default is 2 hours) to drop orphaned connections.
Hope this helps. |
|
| Back to top |
|
|
| Nigelg |
| Posted: Mon Oct 25, 2004 1:18 am Post subject: |
|
|
Grand Master
Joined: 02 Aug 2004
Posts: 1046
|
This problem is not an MQ problem. The 10060 return code implies that there was a timeout when we tried a TCP/IP command (it does not necessarily only occur on connect calls). This is more than likely a problem with the network.
If the problem is network related then it's possible to modify the amount of time allowed for connection and data requests with the TcpMaxConnectRetransmissions and the TcpMaxDataRetransmissions - this will aid the situation where the problem is due to a slow or unreliable network. |
|
| Back to top |
|
|
| Tibor |
| Posted: Mon Oct 25, 2004 1:36 am Post subject: |
|
|
Grand Master
Joined: 20 May 2001
Posts: 1033
Location: Hungary
|
| Shalini wrote: |
| How to get rid of that ???? |
We obligated the developers to rewrite their (censored) code 
Workaround: decreasing the OS level tcpip keepalive dramatically, from 2 hours to 10 minutes.
Tibor |
|
| Back to top |
|
|
|
|
