Hello,

my environment is 2x Open-E DSS, 2x ESX Hosts and 2 Switches. All Servers are connected to both switches and they are linked together using Spanning Tree (STP). The iSCSI Backend uses 172.16.* IP's and the Management LAN is 10.1.* - separated by VLANs.

Like this: http://www.dropmocks.com/miB2n (red are my IPs)

IP Configuration:
DSS Node 1
eth0 = WebGUI (10.1.1.1)
eth1 = Storage Replication (172.16.1.101)
eth2 = bond0 (172.16.2.101)
eth3 = bond0 (172.16.3.101)
bond0 = balancerr (172.16.2.101)

DSS Node 2
eth0 = WebGUI (10.1.1.2)
eth1 = Storage Replication (172.16.1.102)
eth2 = bond0 (172.16.2.102)
eth3 = bond0 (172.16.3.102)
bond0 = balancerr (172.16.2.102)

The iSCSI Failover uses 172.16.100.100 as Virtual IP and both Switches as Ping Nodes (172.16.2.1;172.16.2.2)


If I check the DSS WebGUI I can see that the ping nodes goes up and down. If I enable Email Warnings I get a lot of mails with the message "Failover:Failover: Ping node 172.16.2.2 is down."

I can ping 172.16.2.1 and .2 (the Switch Interfaces) without packet loss from both DSS.

I habe 4 implementations like this, everytime the same problem. I use Version 6.0up65.8101.5217 (64bit) with a small patch from the support to disable the Email-Warnings.




Is the bond the problem?
Any ideas?
From my side I would say that the network part looks good.



Best regards,
Manuel