Interesting. BTW, do you turn on the debug switch at that time?
James Pan wrote:
Hi,
The attached file is a network traffic diagram of a 3-node cluster.
the cluster nodes commnunicate with each other via udp broadcast.
The following is a piece of ha.cf, this piece specifies the UDP
communication port and the interface.
udpport 812
bcast eth0
At the begining, all nodes were stopped, and I started `tcpdump port
812' on hadev5 to monitor the traffic. hadev5 is a node
of the cluster.
At time 10:25:01, the nodes were started one by one. at about
10:26:29, the DC node (hadev6) was stopped. At about 10:27:01,
hadev6 was started again. The pink in the diagram shows the total
taffic that captured by tcpdump, and the blue one shows the packets sent
by hadev5 only.
I do this to give meself a basic concept about how big the traffic
overhead is. And now I'd say the traffic overhead is really small,
especially when the cluster is in stable state.
------------------------------------------------------------------------
_______________________________________________
Linux-HA mailing list
[EMAIL PROTECTED]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
--
BRs,
Sun Jiang Dong
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
