iSCSI Failover

[enealDC]

You are describing an active/active configuration. Unfortunately, this won't work with automatic HA, but it may work if you can deal with manual failover.

[To-M]

enealDC is correct you would want to use only volume replication side of it.

[Slijfer]

enealDC is correct you would want to use only volume replication side of it.

So it is not possible to use a Failover configuration for active/active?

Because the whole idea was to set up a configuration in which the 2nd server will take over the tasks of the first server and vice versa. This is not possible through volume replication because the server will than only backup the data and client computers wont be able to access the storage when there primary server is down.


Greetings

[Gr-R]

So it is not possible to use a Failover configuration for active/active?

Because the whole idea was to set up a configuration in which the 2nd server will take over the tasks of the first server and vice versa. This is not possible through volume replication because the server will than only backup the data and client computers wont be able to access the storage when there primary server is down.


Greetings

Active / Active is not possible yet......

In failover, if Pri goes down, Sec WILL take over, no problem.
When Primary comes back you have to fail back... manually.
The only issue here is the manual part of failing back...... but keep in mind, the hosts will continue to use the targets the whole time, with no problem.

[enealDC]

If you are willing to make an investment in some scripts and learning the API, you can have your cake and eat it too

1. Externalize the HA (e.g. the virtual ip) using a load balancer
2. Create a script or program that leverages the API to cause the volume replication to "flip"

Open-E, the API should allow for one to control volume replication and iSCSI Targets correct?

[Gr-R]

Open-E, the API should allow for one to control volume replication and iSCSI Targets correct?

Yes, and No.....
Target manipulation can be achieved in the API, but the replication switching can be a bit tricky.. I havent tested it, but it may be possible.
got a test box around, and want to test this, i can throw ya some commands if u want.....

[enealDC]

I'm going to take you up on that offer next week. I need to do some prep of my dev environment.

[To-M]

I have some more but let me know if this helps.

create_iscsilv - Creates a logical iSCSI Volume.
volume_iscsi_remove - Removes a logical iSCSI volume
volume_replication - Adds and removes replication to volume.
volume_status - Displays storage info.


FAILOVER:

Usage: failover --status | --actions | --start | --stop | --failover | --synchronize | --failback

This function allows you to stop, run or change the operation mode for the
given server.

Options:
-h, --help show this help message and exit
--status show iSCSI Failover status.
--actions show available iSCSI Failover actions.
--start start iSCSI Failover. On the primary server the Start
action
controls both servers. On the secondary server, it
controls
the secondary server only.
--stop stop iSCSI Failover. On the primary server the Stop action
controls both servers. On the secondary server, it
controls
the secondary server only.
--failover failover action delegate (switch) active server state
to the
passive server.
--synchronize synchronize data from the secondary/active server to the
primary server
--failback failback action will return the active server state to the
primary server.