We've got a PS6510 we've been using for awhile now. We normally just use eth0 on both interaces, and it works well for us. Recently we've had a customer that wants to have 2TB of space on it, so my idea was to take eth1, and create a seperate network, so customers cannot see into our internal network.
So I hooked eth1 up to another switch, and the customer is connected to that. 192.168.2.250 is our primary network, and 3.250 is the secondary network.
When I start iscsi, and connect to the 192.168.3.250 from the customers machine, it does its thing, and the SAN sends back 192.168.2.250 as the connection point, and then when I go to log in, it complains because eth1 doesn't have access to the 192.168.2.250 network. My goal was to segregate customers from our internal network.
Is this doable with EQuallogics?
Heres from the server.
[root@stats1 init.d]# ping 192.168.3.250 PING 192.168.3.250 (192.168.3.250) 56(84) bytes of data. 64 bytes from 192.168.3.250: icmp_seq=1 ttl=255 time=0.168 ms 64 bytes from 192.168.3.250: icmp_seq=2 ttl=255 time=0.103 ms ^C --- 192.168.3.250 ping statistics --- 2 packets transmitted, 2 received, 0% packet loss, time 999ms rtt min/avg/max/mdev = 0.103/0.135/0.168/0.034 ms [root@stats1 init.d]# iscsiadm -m discovery -t st -p 192.168.3.250 192.168.2.100:3260,1 iqn.2001-05.com.equallogic:0-8a0906-922e2040a-4ed33cc2e155c66f-spamtitan [root@stats1 init.d]# iscsiadm -m node -l Logging in to [iface: default, target: iqn.2001-05.com.equallogic:0-8a0906-922e2040a-4ed33cc2e155c66f-spamtitan, portal: 192.168.2.100,3260] (multiple) iscsiadm: Could not login to [iface: default, target: iqn.2001-05.com.equallogic:0-8a0906-922e2040a-4ed33cc2e155c66f-spamtitan, portal: 192.168.2.100,3260]. iscsiadm: initiator reported error (8 - connection timed out) iscsiadm: Could not log into all portals [root@stats1 init.d]# ping 192.168.2.100 PING 192.168.2.100 (192.168.2.100) 56(84) bytes of data. ^C --- 192.168.2.100 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms
No. The design of the PS Series is that any port can be used for iSCSI traffic. There is only one discovery address per group. So the other customers would have to have access to that network anyways.
You would need your 6510 storage to act as front end for another server to provide storage to that other network. E.g an NFS or SMB server.
Also for proper performance and redundancy both of the 10GbE ports should be enabled on the same network.
Ideally connected to two, interlinked switches for redundancy, and performance as well.
Yeah sorry. Without being able to create multiple Discovery addresses your plan won't work. In order for multiple subnets to work, they have to be routable. If you loose the router/GW to the Group IP new volumes won't connect.
If it *had* to be iSCSI, there are iSCSI targets available for multiple OSs. You can mount the iSCSI EQL volume, then 'share' it out over iSCSI to the customer. SMB/NFS might be cleaner though.