StarWind V5.9/6.0 Beta

[Vavaga]

"Bohdan (staff)"]Thanks!
Auto restore is disabled in current 5.9 beta build by Dev team for debug purposes.
I'm sorry for misleading.

Ok, can I get 5.9 beta with enabled restore option? For me important test how it works after power down.

[anton (staff)]

Sure you can. Just wait for the next 5.9 build where it would be enabled.

[Vavaga]

Could you tell please, will be any option in verssion 5.9 for automaticly rescan of iSCSI software adapter in esxi 5 hosts after power down. For time beeing if both esxi 5 hosts, where starwin installed, will be rebooted, the iSCSI software adapter will not be foud till manual rescan and as result HA storage will not be availible.

[Bohdan (staff)]

You have created the HA device with StarWind installed inside VMs and connected that device to the same ESXi servers that host the VMs. Am I right?

[Vavaga]

You have created the HA device with StarWind installed inside VMs and connected that device to the same ESXi servers that host the VMs. Am I right?

I have three physical servers with esxi 5 hosts, where VMVsphere 5 and HA cluster installed. On the first host I have vcenter 5 and users VM with datastore on HA storadge. On second and third hosts I have created the HA storadge device with StarWind installed inside VMs. If all three hosts will be rebooted due power down, then after power up no one user VMs will be available, because esxi hosts scaning iSCSI software adapters immediately after reboot and at that time StarWind is not started yet to work on second and third hosts.

[Bohdan (staff)]

I think ESXi expects iSCSI storage to be booted first and it must be available before ESXi boot process, but this doesn't happen because the storage nodes are VMs on the same hosts.

That is related to ESXi behavior and I would recommend you to ask the question on the VMware community forum.
AFAIK you are not alone with such problem...

http://communities.vmware.com/message/1984194
http://communities.vmware.com/thread/300527
http://communities.vmware.com/message/1831384
http://communities.vmware.com/thread/399137

I think the startup procedure should be:

1) Boot ESXi hosts
2) Auto-start storage VMs
3) Reconnect share storage provided by storage VMs to ESXi hosts
4) Start VMs located on the shared storage

The typical workarounds are manual rescan with a script (for step 3) and VM boot delay (for step 4) .

[Vavaga]

Thank you, I have resolve this problem. I put scripts in sheduler on startup in VMs where Starwind is running. The details is here http://communities.vmware.com/thread/405529?tstart=0

[Bohdan (staff)]

Thanks for keeping us updated!
As fo "Starwind iSCSI не поддерживаются в VSphere 5", StarWind is vSphere 4.1 certified. We are going to perform vSphere 5 certification with the next StarWind release.

[kmax]

Do we have an ETA on official 5.9 yet?

[anton (staff)]

1) We don't have any public milestone maps.

2) We'll skip 5.9 and will "gold" it as V6.

Do we have an ETA on official 5.9 yet?

[kmax]

Will 5.9/6.0 allow online updating..meaning won't have to take entire HA down like we did from 5.7 to 5.8?

[anton (staff)]

Live updates are available since 5.8 or so. Should work with 5.7 as well (some restrictions applied). Please drop me a message telling why you had to put cluster down with 5.7 -> 5.8 upgrade. Thanks!

Will 5.9/6.0 allow online updating..meaning won't have to take entire HA down like we did from 5.7 to 5.8?

[kmax]

From the release notes on 5.8:

Warning: The existing HA devices cannot be upgraded to version 5.8 without bringing the storage offline for a short time.
Please, take the following steps to update the existing HA devices:
1. To prevent data loss, disconnect clients from HA (if it is possible).
2. Update StarWind service on the first HA node. Wait until the service starts. At this step the HA node is unable to synchronize its HA devices and is not accepting the client connections. Client requests are processed by the second HA node. The next step will disconnect the existing client connections.
3. Update the StarWind service on the second HA node. Wait until the service starts.
4. Start synchronization on the first HA node. The second node changes its state to “ready” and starts processing client connections. Now, you can safely connect to the HA device.
5. Wait for synchronization to finish. Now the first HA node can process the client connections too.

[anton (staff)]

Shame on me!

From the release notes on 5.8:

Warning: The existing HA devices cannot be upgraded to version 5.8 without bringing the storage offline for a short time.
Please, take the following steps to update the existing HA devices:
1. To prevent data loss, disconnect clients from HA (if it is possible).
2. Update StarWind service on the first HA node. Wait until the service starts. At this step the HA node is unable to synchronize its HA devices and is not accepting the client connections. Client requests are processed by the second HA node. The next step will disconnect the existing client connections.
3. Update the StarWind service on the second HA node. Wait until the service starts.
4. Start synchronization on the first HA node. The second node changes its state to “ready” and starts processing client connections. Now, you can safely connect to the HA device.
5. Wait for synchronization to finish. Now the first HA node can process the client connections too.

[kmax]

No problem, after you wrote that I thought I was nuts.

But the question still stands...5.8 to 6.0 without interruption?