A question Jason - what happens if you yank one of the RAID-0 drives out of one node and replace it with another drive to simulate a drive failure? I'm mainly interested in whether you have to manually re-create the RAID-0 array and re-synchronise StarWind? Cheers, Rob. PS. Of course, don't do this ...

Above you said that sync priority is always enabled and given the fastest setting still has some component of throttling, then the sync channel always has some form of throttling - so you'll never use the full throughput of 10GbE sync networks. Which I can't quite believe.

Cheers, Rob.

So much for thin provisioning.

Quite

Cheers, Rob.

One for the wishlist - a tag indicating what is using the connections:

StarWind sync
StarWind heartbeat
External iSCSI

Or whatever... I wondered the same myself.

Cheers, Rob.

I'm still concerned about that very real scenario I last posted. Without manual intervention, the entire infrastructure cannot recover from power outage. This is supposed to be high-availability! Compared to single node, this HA is actually worse from one point of view. Imagine the discussion with o...

May I doublecheck why do we want it? I mean, what`s the reason and purpose? Because of the confusion of node synchronisation after clean power down. Just an idea so that one can control the function manually. If auto-synchronisation worked better then it might not be needed. This is possibly the mo...

It is always applicable.

So sync traffic is always throttled? That seems a rather strange thing to do as it's artificially putting a bottleneck in.

Cheers, Rob.

So this is how it ends up:
storage3 and HAImage3 have disappeared entirely from SAN91 even though the device is still there. What I'm suggesting is that whilst the target can be removed, the orphan device should still appear so you know it's still there.

Cheers, Rob.

In the screenshot showing lsfs1 without a target, I had to re-add the device to get that screenshot. By default, the target and device disappear entirely. They are still on disk so you can re-add the device.

Cheers, Rob.

I think we`ll just take away ability to choose from users since AutoDefrag is the thing that need to be turned on always.

I agree - you'd have to be slightly mad to not have it turned on

Cheers, Rob.

Yes it will work but I guess your overall performance will limited by the slowest drives. By SAS, do you mean 15k drives or the higher capacity but slower rotation speed nearline SAS? I guess you mean NL SAS as you need a lot of 600GB SAS drives to match the higher capacity of SATA.

Cheers, Rob.

Okay I understand but maybe not remove the option entirely but simply grey out the "Add replica" and "Remove replica" buttons at the top (possibly add text saying function disabled during sync).

Cheers, Rob.

One for the documentation but what are the various synchronisation types and what do they mean? In the attached example, it's "Fast".

Cheers, Rob.

I have noticed that the replication manager option is not there whilst a node is synchronising. So in the lab, node #1 was marked as synchronised and the replication manager option is there but on node #2 which was synchronising, the option is missing. Maybe that's what I noticed before. Any particu...

Yup - removed Users from the E: drive and restarted StarWind and all is fine. So just SYSTEM & Administrators I'm guessing...

Cheers, Rob.