open-e failover for onapp cloud

Hello,

I have two SAN servers with volume replication with failover setup(active/passive). The failover status looks fine on both primary and secondary nodes. When I manually switch the primary node to suspend mode, the secondary node took the place, but all my onapp VMs are down. The volume replication is Consistent on both nodes before I initial the failover test. After I return the active server state to the primary server, the VMs are still down, I could not do any action like create, start, stop, restore backup, etc.

I am wondering how the onapp know the secondary SAN node. Do I need to manually change the storage configuration when the primary SAN node is down?

Anyone have experience with open-e failover for onapp cloud? Thank you.

Make sure the initiators are using the Virtual IP to make the iSCSI connections.
If there are any allow/deny rules or CHAP authentication being used, this needs to be the same on each node to allow proper connectivity.
This may help in the setup :
http://www.open-e.com/library/webcasts-and-videos/
Open-E DSS V6 AutoFailover with VMware 4.0 - Webcast

If both SAN nodes are connecting to one switch, can open-e swap the Client storage Access link IP ?

Eg:

Primary node:
Client Storage Access: 192.168.2.220

Secondary node:
Client Storage Access: 192.168.2.221

When Primary node is down, the secondary node Client Storage Access IP changed to 192.168.2.220.

Quote:

---

Originally Posted by Gr-R

Make sure the initiators are using the Virtual IP to make the iSCSI connections.
If there are any allow/deny rules or CHAP authentication being used, this needs to be the same on each node to allow proper connectivity.
This may help in the setup :
http://www.open-e.com/library/webcasts-and-videos/
Open-E DSS V6 AutoFailover with VMware 4.0 - Webcast

---

Thanks. I have Virtual IP on Client Storage Access network. Do you mean I need to use the Virtual IP on onapp cloud instead of the actual IP of the Storage Access network? I don't use CHAP authentication.

Yes, you need to use the virtualized IP that you et up when configuring the failover, for iSCSI traffic.
The nodes cant move the real NIC IP, we move the virtual IP to the active node.

Watch the video I provided, it will help make it clear.

Quote:

---

Originally Posted by Gr-R

Yes, you need to use the virtualized IP that you et up when configuring the failover, for iSCSI traffic.
The nodes cant move the real NIC IP, we move the virtual IP to the active node.

Watch the video I provided, it will help make it clear.

---

I just watched the video and found out the answer. Thank you for your support.:D

Glad to help!

Sorry to bother you again. I have the following setup, but I am not able to ping the virtual IP from both SAN servers. In the video, I see they can ping each other. Should I setup 192.168.100.254 as gateway? Thanks.

Primary SAN

eth0 - management network - 192.168.1.207
Gateway - 192.168.1.254
eth1 - SAN storage network - 192.168.2.207
eth2 - Replication network - 192.168.3.207

Secondary SAN

eth0 - management network - 192.168.1.208
Gateway - 192.168.1.254
eth1 - SAN storage network - 192.168.2.208
eth2 - Replication network - 192.168.3.208

Virtual IP is on eth1 for both SANs
IP address: 192.168.100.100
Netmask: 255.255.255.0
Broadcast: 192.168.100.255

It does seem like your gateway is not handling the virtual IP, and its subnet.
You should check this.
Or, if the failover service is not yet started the Virtual IPs are not up and available yet.

The failover status shows the Communication via Storage link is OK. I am not sure how the gateway is handling the virtual IP. Do I need to do any special setup in the Switch(both SAN are connecting to the same switch)? Trunk the 192.168.100.* network?

Please ignore my last reply. I may found the problem.

well...... what was it?

It's my network setup problem. My hypervisor server is using 192.168.2.4, but my SAN virtual IP was 192.168.100.100. I switched the virtual IP with the Storage access IP and now the failover is working.