StarWind iSCSI SAN Version 8.0 Beta-2

[Bohdan (staff)]

@@@ beta-3 is out @@@

This topic is dedicated for StarWind iSCSI SAN Version 8.0 Beta-2

Improvements over first beta release:
A lot of small fixes and updates, several new features.
Log-structured File System container: fixes and performance improvements.
Flash cache: fix for the issue when non-default path has been used for L2 cache storage device.
Core: VAAI support for single-node devices and devices with synchronous replication. WRITE SAME, EXTENDED COPY, ATS commands are supported.
Scale-out cluster: organize your servers to SAN cluster and create multi-node devices across the cluster easily.
Asynchronous replication: asynchronous mode added to replication engine. Asynchronous replication can be used to replicate data to remote site. Information on snapshots that are stored on the primary storage is replicated too. Replication channel bandwidth doesn’t affect the performance of primary storage which is replicated. State of the primary storage can be easily restored from replica.

Other features of StarWind iSCSI SAN Version 8.0 Beta:

New Log-structured File System container:
LSFS container implements thin-provisioning, snapshots and optional online deduplication features. Synchronous replication for high availability can be configured for a LSFS device. Snapshots are replicated too.
LSFS device is flash-friendly. It eliminates multiple rewrites of same data blocks.
LSFS provides high performance for random write access patterns, eliminating I/O blender effect in virtualized environments.

Flash cache:
L2 cache can be used with LSFS or flat storage devices for acceleration.

Management console:
User interface has been reworked to provide super simple and clean user experience.

Default storage pool path can be set up for the server. This path will be used for all new devices.

New simple device creation wizard allows create virtual device with target in three steps.

High availability device can be created from any existing device by adding replication partner through replication manager.

Network settings for replication can be changed on fly without rebuilding replication partners.

Installation notes

Installation: Previous versions can be updated by installing this version over the existing installation.
Warning: Mirror, IBV and Deduplication devices of version 6.x and earlier versions have limited support in v8.0! Data from existing Mirror, IBV and Deduplication devices must be migrated to new ImageFile or LSFS devices after installation.
Use replication manager to add synchronous replication functionality for your device. Use LSFS virtual device type when thin-provisioning and snapshots features are needed.
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.


Download link:
http://starwindsoftware.com/tmplink/StarWindBeta.exe
License:
http://starwindsoftware.com/tmplink/sw8 ... tatest.swk
http://www.starwindsoftware.com/tmplink ... tatest.swk

[anton (staff)]

Thank you Bohdan!

Gentlemen, please post your V8 Beta-2 reports here.

Thank you!

[lohelle]

When I upgraded from beta1 to beta2 ESXi did not reconnect (even after resync). I could select "add storage" and choose "keep signature" to get access to my ISO-datastore, but this did not work for my VM datastore.
Add storage + new signature worked. I then had to register all VMs again.

[Bohdan (staff)]

Am I right saying that this problem occurs for both virtual disk types: thick (ImageFile) and thin (LSFS)?
I mean the disks which are used for HA (synchronous replica) device.

[lohelle]

This was standard image file devices (non-HA). I created the targets on the latest v6-release, then upgraded to Beta1 (success) and finally to beta2.

I do not have any logs now. Sorry.. :\

[fbifido]

Sorry did not see this post

Anyway, some correction from by first post about beta 2:

1. 4k works when I created a 100GB LUN, but failed when tried 1655GB, so I use 8k instead.
2. lun's created in beta1, does not auto connect in beta2 on esxi 5.5.0

The lun that is still active was created after updated to Beta 2, in picture below, as you can see no VM is loaded.



I can manually add the LUNs back to esxi storage, but as soon as I reboot my esxi server, I am back to the picture above.



In the picture below I told it to keep the previous stuff



Now with all the LUNs re-added to esxi the VMs now show up and I can start them up. see pic below



How do I go about fixing this problem?
I will try to re-create the LUNs and move the VMs to these new LUNs:

[fbifido]

Sorry the full picture does not show so here is the links to the full pictures:

Sorry did not see this post

Anyway, some correction from by first post about beta 2:

1. 4k works when I created a 100GB LUN, but failed when tried 1655GB, so I use 8k instead.
2. lun's created in beta1, does not auto connect in beta2 on esxi 5.5.0

The lun that is still active was created after updated to Beta 2, in picture below, as you can see no VM is loaded.

http://i39.tinypic.com/28sse9f.png

I can manually add the LUNs back to esxi storage, but as soon as I reboot my esxi server, I am back to the
picture above.

http://i39.tinypic.com/14smyr8.png

In the picture below I told it to keep the previous stuff

http://i43.tinypic.com/fwow8j.png

Now with all the LUNs re-added to esxi the VMs now show up and I can start them up. see pic below

http://i43.tinypic.com/wgymtj.jpg
How do I go about fixing this problem?
I will try to re-create the LUNs and move the VMs to these new LUNs:

http://i44.tinypic.com/2urpb0z.png

[turkina]

I went full ghetto and just moved my important lab VMs onto local storage then upgraded Starwind to Beta 2 while the rest of the VMs were running. Worked perfectly. Service was only down for about 15 seconds, the running VMs were fine.

Obviously I would NEVER do that in a production environment but being my home lab I didn't care. lol

[mahuan1688]

Why didn't this component backup ah, BETA1 and this part, the official version will join in this function!

[anton (staff)]

We'll do seamless migration V6 -> V8 in a release candidate.

Why didn't this component backup ah, BETA1 and this part, the official version will join in this function!

[Ironwolf]

Beta-2 Starwind service keeps stopping anywhere from 10min to 16 hours, I when I reloaded beta-1 my problems when away.

Log Name: Application
Source: Application Error
Date: 10/29/2013 9:06:21 AM
Event ID: 1000
Task Category: (100)
Level: Error
Keywords: Classic
User: N/A
Computer: Athena.BargeCode.local
Description:
Faulting application name: StarWindService.exe, version: 8.0.0.4130, time stamp: 0x5266a320
Faulting module name: ntdll.dll, version: 6.1.7601.18229, time stamp: 0x51fb164a
Exception code: 0xc0000005
Fault offset: 0x0000000000053377
Faulting process id: 0x77c
Faulting application start time: 0x01ced4c0cb3eb70e
Faulting application path: C:\Program Files\StarWind Software\StarWind\StarWindService.exe
Faulting module path: C:\Windows\SYSTEM32\ntdll.dll
Report Id: 0c91dfca-40b4-11e3-aeb1-0015171d858f
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Application Error" />
<EventID Qualifiers="0">1000</EventID>
<Level>2</Level>
<Task>100</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2013-10-29T16:06:21.000000000Z" />
<EventRecordID>7744</EventRecordID>
<Channel>Application</Channel>
<Computer>Athena.BargeCode.local</Computer>
<Security />
</System>
<EventData>
<Data>StarWindService.exe</Data>
<Data>8.0.0.4130</Data>
<Data>5266a320</Data>
<Data>ntdll.dll</Data>
<Data>6.1.7601.18229</Data>
<Data>51fb164a</Data>
<Data>c0000005</Data>
<Data>0000000000053377</Data>
<Data>77c</Data>
<Data>01ced4c0cb3eb70e</Data>
<Data>C:\Program Files\StarWind Software\StarWind\StarWindService.exe</Data>
<Data>C:\Windows\SYSTEM32\ntdll.dll</Data>
<Data>0c91dfca-40b4-11e3-aeb1-0015171d858f</Data>
</EventData>
</Event>

Log Name: Application
Source: Application Error
Date: 10/29/2013 9:05:41 AM
Event ID: 1000
Task Category: (100)
Level: Error
Keywords: Classic
User: N/A
Computer: Athena.BargeCode.local
Description:
Faulting application name: dllhost.exe, version: 6.1.7600.16385, time stamp: 0x4a5bca54
Faulting module name: hardwareprovider.dll, version: 1.0.0.1, time stamp: 0x52406d28
Exception code: 0xc0000005
Fault offset: 0x000000000000834a
Faulting process id: 0xbf8
Faulting application start time: 0x01ced4c04aa2441e
Faulting application path: C:\Windows\system32\dllhost.exe
Faulting module path: C:\VssHWProviderStarWind\hardwareprovider.dll
Report Id: f4d4697e-40b3-11e3-aeb1-0015171d858f
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Application Error" />
<EventID Qualifiers="0">1000</EventID>
<Level>2</Level>
<Task>100</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2013-10-29T16:05:41.000000000Z" />
<EventRecordID>7739</EventRecordID>
<Channel>Application</Channel>
<Computer>Athena.BargeCode.local</Computer>
<Security />
</System>
<EventData>
<Data>dllhost.exe</Data>
<Data>6.1.7600.16385</Data>
<Data>4a5bca54</Data>
<Data>hardwareprovider.dll</Data>
<Data>1.0.0.1</Data>
<Data>52406d28</Data>
<Data>c0000005</Data>
<Data>000000000000834a</Data>
<Data>bf8</Data>
<Data>01ced4c04aa2441e</Data>
<Data>C:\Windows\system32\dllhost.exe</Data>
<Data>C:\VssHWProviderStarWind\hardwareprovider.dll</Data>
<Data>f4d4697e-40b3-11e3-aeb1-0015171d858f</Data>
</EventData>
</Event>

Log Name: Application
Source: Application Error
Date: 10/29/2013 9:05:18 AM
Event ID: 1000
Task Category: (100)
Level: Error
Keywords: Classic
User: N/A
Computer: Athena.BargeCode.local
Description:
Faulting application name: StarWindService.exe, version: 8.0.0.4130, time stamp: 0x5266a320
Faulting module name: StarWindService.exe, version: 8.0.0.4130, time stamp: 0x5266a320
Exception code: 0xc0000005
Fault offset: 0x000000000015dda4
Faulting process id: 0xb1c
Faulting application start time: 0x01ced4c07f3582fe
Faulting application path: C:\Program Files\StarWind Software\StarWind\StarWindService.exe
Faulting module path: C:\Program Files\StarWind Software\StarWind\StarWindService.exe
Report Id: e736993e-40b3-11e3-aeb1-0015171d858f
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Application Error" />
<EventID Qualifiers="0">1000</EventID>
<Level>2</Level>
<Task>100</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2013-10-29T16:05:18.000000000Z" />
<EventRecordID>7737</EventRecordID>
<Channel>Application</Channel>
<Computer>Athena.BargeCode.local</Computer>
<Security />
</System>
<EventData>
<Data>StarWindService.exe</Data>
<Data>8.0.0.4130</Data>
<Data>5266a320</Data>
<Data>StarWindService.exe</Data>
<Data>8.0.0.4130</Data>
<Data>5266a320</Data>
<Data>c0000005</Data>
<Data>000000000015dda4</Data>
<Data>b1c</Data>
<Data>01ced4c07f3582fe</Data>
<Data>C:\Program Files\StarWind Software\StarWind\StarWindService.exe</Data>
<Data>C:\Program Files\StarWind Software\StarWind\StarWindService.exe</Data>
<Data>e736993e-40b3-11e3-aeb1-0015171d858f</Data>
</EventData>
</Event>

[Bohdan (staff)]

Beta-2 Starwind service keeps stopping anywhere from 10min to 16 hours, I when I reloaded beta-1 my problems when away.

Log Name: Application
Source: Application Error
Date: 10/29/2013 9:06:21 AM
Event ID: 1000
Task Category: (100)
Level: Error
Keywords: Classic
User: N/A
Computer: Athena.BargeCode.local
Description:
Faulting application name: StarWindService.exe, version: 8.0.0.4130, time stamp: 0x5266a320
Faulting module name: ntdll.dll, version: 6.1.7601.18229, time stamp: 0x51fb164a
Exception code: 0xc0000005
Fault offset: 0x0000000000053377

</Event>

Please zip and send us StarWind logs. Thank you!

[fgiroux]

Good day!

I have just installed Starwind 8 beta 2. I had some issues, but I am not sure if it was with Starwind or with vSphere or both.

Lab environment:

• vSphere 5.1 U1c
• 4 LUNs presented to 3 ESXi hosts
• 2 LUNs are using disk bridges and 2 are using Image files
• Permissions are wide-open from the ESXi hosts to the Starwind server
• The Starwind Server is running on top of Windows Core 2008 R2 fully patched
• Hosts hardware is Mac Minis. They are all identical.

The update process was made on top of version 6. The ESXi hosts were powered-off when updating. The update on the server went like a breeze. The service was restarted, but I rebooted the server (just to make sure).

The update of the management console on my workstation was also uneventful. After the management console was updated, I connected to the server and everything looked good. I checked the presence of the configured LUNs, access rights (I am not using CHAP permissions) and everything was in order.

I then started my three ESXi hosts at the same time. None of them recognized their datastores, but they could all connect to the Starwind Server. The datastores view would shows only the local drive.

I then went to the Storage View to “add storage”. On the first host, it showed me the list of 4 configured datastores and I added them one by one. It asked me if I wanted to resignature and I obviously said no and the host was able to see the content. I then started my VMs (including the vCenter).

On the 2 remaining hosts, when I tried to add the storage, nothing appeared. Tried several ESXi debugging steps without luck and I had no error messages on the hosts whatsoever. Out of desperation, I restarted the Starwind service and... I now could see the datastores.

But not so fast... I could see them, but upon trying to add them to the hosts storage, ESXi could not see the signature giving me the only option to format the LUN. I noticed the VMFS label showing a suffix of “(head)”. I then ran the command “esxcfg-volume –M” followed by the name of the volume. The datastores reconnected with success and now, everything is running fine.

Is this an ESXi problem? Starwind? A little of both? I cannot say, but I think it is important you know about this.

Now that it is running, first general impressions are positive. Very happy about VAAI support

I have not noticed an increase in performance yet, but I can say for sure the disks are spinning less than before for a similar load. So, as I increase the load, I should feel a lesser hit.

Are there tweaks applicable to ESXi hosts with Starwind?

I hope this helps

Fred

[Bohdan (staff)]

Thanks guys!
It seems like the issue is related to the product/vendor names and device serial numbers update procedure.
It will be fixed soon and the build will be updated.

[fgiroux]

Hello!

After upgrading Starwind yesterday, I upgraded my vSphere environment from 5.1 U1c to 5.5 with vCenter 5.5 patch 1 that came out today.

After heavily soliciting Starwind storage for the better part of the day (Storage vMotions, guest OS updates, ESXi updates and so on), Starwind did not flinch a single instant. I am all done now (for today anyway) and I can safely say Starwind is running more smoothly than in version 6. It looks like it is dealing better with I/O peaks.

Does VAAI have anything to do with this? Improvement in the code? Anyway... Congrats

The performance graph in the management console seems to be out of whack. The one hour period shows spikes of activities, even if not that much, but the 24 hour graph is flat lined starting about 2.5 hours prior:

Last 24 hours:

Last 24 hours.jpg (71.51 KiB) Viewed 95214 times

Last hour:

Last hour.jpg (71.94 KiB) Viewed 95208 times

Have a nice day!

Fred