Start a Conversation

Unsolved

This post is more than 5 years old

1696

July 15th, 2011 05:00

Network related errors

Hi!

Our customer uses Nw 7.6.2

He recognized, when he run the backup group and a network error occurs (Network configuration, intermittent connection problem

between the server and client)  the backup

job doesnt complete, it will run forever.

The only way to solve it to restart the server and restart  the group, but it isnt

a good solution. The ideal behaviour would be if the group stopped

if the Network error occurs.

Do you have any idea, which can prevent the group running forever?

Group inactivity timeout? Client inactivity timeout? NSR_KEEPALIVE setting?

BR: Paul

2.4K Posts

July 15th, 2011 08:00

There are really lots of possibilities which will result in similar errors. Debugging will not be easy. However, to stop a group, try the following methods:

  - for each client, stop the NW client daemon until the group will become aborted

  - on the server, stop the savegroup with the new jobkill command.

116 Posts

July 15th, 2011 09:00

Hi Paul,

Try set in group client retries to 0, maybe help in your problem.

client retries (read/write)
          The number of times failed clients should be retried before
          savegroup gives up and declare them failed.  Zero means don't retry.
          Abandoned saves are not retried, because they may eventually
          complete.  A client's save sets are retried by savegroup whenever
          savegroup would otherwise not be able to start a new save set.  That
          is, savegroup prefers to start new save sets first, and only retries
          when there is nothing else to do.
          Example: client retries: 1;

Regards

Luis Rogério

202 Posts

July 27th, 2011 05:00

Hi Roberto!

Thanks.

I dont fully understand that,

BR: Paul

202 Posts

July 27th, 2011 06:00

Hi!

I checked the Networker settings. The client retries is 0 in all groups.

Do you have any new idea?

Maybe the host keepalive settings or something similar?

Thanks in advance.

BR: Paul

July 27th, 2011 07:00

Hello Paul,

I would suggest you to apply the TCP/IP best practices. Please create the following entries in the registry (Windows OS):

DWROD entries with Decimal values

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize=256000

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\GlobalMaxTcpWindowSize=16777216

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\KeepAliveInterval=1000

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\KeepAliveTime=1800000

For more informaiton about these settings please refer to the following information:

http://technet.microsoft.com/en-us/library/cc938219.aspx

http://technet.microsoft.com/en-us/library/cc957546.aspx

http://technet.microsoft.com/en-us/library/cc957548.aspx

http://technet.microsoft.com/en-us/library/cc957549.aspx

Hope that helps.

Thank you.

Carlos.

14.3K Posts

July 29th, 2011 16:00

The group is probably still "running" as network connection is left hanging (you can verify it with netstat).  Addressing the network issue would be the way to go.

No Events found!

Top