HA cost 12 gig of data! $2000 in one month?!!!

[jonathan soong]

Hi everyone,

I have just set up a 2 system heartbeat/drbd fail-over system in a local 
ISP's server room.

I just got charge 12 gig of data for systems that are not yet 'live'!!!

I'm presuming that this is due to the DRBD replication (its going thru 
the ISP's switch), and expect this problem to be clarified 

However, when looking at the network data i noticed.  A LOT of data 
hitting XXX.XXX.XXX.79:ha-cluster - the broadcast address
for our subnet.

As you can see below, both machines are hitting this address EVERY SECOND.

Is this a lot of traffic? Why doe it hit the broadcast that often (i'm 
presuming its doing this to make sure the network is alive
or to see trafffic on the network device).?

Is it necessary for it to do this so often? I'm worried that it is 
contributing to my data charge ($2000!).

Any help would be most appreciated.

Thanks

Jon
------------------------------------------------------------------------------------ 




16:10:57.180528 iha2.XXX.XXX.32818 > XXX.XXX.XXX.79.ha-cluster:  udp 139 
(DF)
16:10:58.109051 iha1.XXX.XXX.33093 > XXX.XXX.XXX.79.ha-cluster:  udp 143 
(DF)
16:10:58.656638 802.1d config c000.00:04:9a:66:d9:bb.8026 root 
c000.00:04:9a:66:d9:bb pathcost 0 age 0 max 20 hello 2 fdelay 15
16:10:59.180572 iha2.XXX.XXX.32818 > XXX.XXX.XXX.79.ha-cluster:  udp 139 
(DF)
16:11:00.108934 iha1.XXX.XXX.33093 > XXX.XXX.XXX.79.ha-cluster:  udp 143 
(DF)
16:11:00.658094 802.1d config c000.00:04:9a:66:d9:bb.8026 root 
c000.00:04:9a:66:d9:bb pathcost 0 age 0 max 20 hello 2 fdelay 15
16:11:01.180505 iha2.XXX.XXX.32818 > XXX.XXX.XXX.79.ha-cluster:  udp 139 
(DF)
16:11:02.108927 iha1.XXX.XXX.33093 > XXX.XXX.XXX.79.ha-cluster:  udp 143 
(DF)
16:11:02.656870 802.1d config c000.00:04:9a:66:d9:bb.8026 root 
c000.00:04:9a:66:d9:bb pathcost 0 age 0 max 20 hello 2 fdelay 1