[Spread-users] Failure Detection of members on same segment

[Pilling, Michael]

I imaging your problem is running windows XP.
 
We had this problem. Spread is a highly efficient lazy protocol,
it doesn't send pings, only detecting failure on an attempt to send.
 
Windows XP doesn't detect cable disconnection, where as most other OS/s
do.
 
You can fix this with a registry edit.
 
Thanks to Steve Duff for finding this fix.
 
"Just to feed back a quick workaround to the list, Microsoft have
documented a method to disable the Media Sensing feature which causes
the problem with Spread under XP. You can find it here:

http://support.microsoft.com/kb/239924/
<http://support.microsoft.com/kb/239924/>   

"

 

---- 
Dr Michael Pilling 
C3I Division 
DSTO 
PO Box 1500 
Edinburgh SA 5111 
Phone: 08 8259 7017 Fax: 08 8259 5589 
email: Michael.Pilling at dsto.defence.gov.au 


IMPORTANT: This email remains the property of the Australian Defence
Organisation and is subject to the jurisdiction of section 70 of the
CRIMES ACT 1914. If you have received this email in error, you are
requested to contact the sender and delete the email.

 

________________________________

From: spread-users-bounces at lists.spread.org
[mailto:spread-users-bounces at lists.spread.org] On Behalf Of Leandro
Sales Holanda Pinto
Sent: Wednesday, 25 June 2008 10:33
To: spread-users at lists.spread.org
Subject: [Spread-users] Failure Detection of members on same segment



I'm having a problem with failure detection on members that are
configured on the same segment.

This is my scenario:

I Have two different nodes connected to the same daemon that is hosted
on one of the nodes. 

When, for example, the network cable from the node, that is not hosting
the daemon, is disconnected, the other node is not notified about this
failure and the disconnected node freezes when trying to send a message.


Some suggestions? What am I doing wrong?

ps. If the nodes are connected to different daemons then the
notifications are going ok (it detects a network partition).

Leandro

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.spread.org/pipermail/spread-users/attachments/20080625/48eda3cc/attachment.html