Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Question
Sunday, October 16, 2016 11:33 PM
The target cluster are running the final RTM MSDN bits of Nano server datacenter, with latest *.321.* windows update packages, and the member nodes have a mixture of NVMe and SATA storage. The -verbose output of enable-clusters2d shows the text at the bottom of this posting, up to the hang...
The log files on the cluster members show no useful information. The "cannotpoolreason" for the NVMe devices shows text similar to "waiting for verification", after the hang..
I've removed the cluster several times, cleaned the systems, and reinitiated the cluster create and enable s2d with same effect.
How may I further diagnose my problem?
VERBOSE: vms-c: 2016/10/16-16:25:56.110 Setting default fault domain awareness on clustered storage subsystem
VERBOSE: vms-c: 2016/10/16-16:25:56.751 Waiting until physical disks are claimed
VERBOSE: vms-c: 2016/10/16-16:25:59.767 Number of claimed disks on node 'VMS-1': 6/2
VERBOSE: vms-c: 2016/10/16-16:25:59.783 Number of claimed disks on node 'VMS-2': 6/2
VERBOSE: vms-c: 2016/10/16-16:25:59.798 Node 'VMS-1': Waiting until cache reaches desired state (HDD:'ReadWrite'
SSD:'WriteOnly')
VERBOSE: vms-c: 2016/10/16-16:25:59.798 SBL disks initialized in cache on node 'VMS-1': 6 (6 on all nodes)
VERBOSE: vms-c: 2016/10/16-16:25:59.814 SBL disks initialized in cache on node 'VMS-2': 6 (12 on all nodes)
VERBOSE: vms-c: 2016/10/16-16:25:59.814 Cache reached desired state on VMS-1
VERBOSE: vms-c: 2016/10/16-16:25:59.829 Node 'VMS-2': Waiting until cache reaches desired state (HDD:'ReadWrite'
SSD:'WriteOnly')
VERBOSE: vms-c: 2016/10/16-16:25:59.845 Cache reached desired state on VMS-2
VERBOSE: vms-c: 2016/10/16-16:25:59.845 Waiting until SBL disks are surfaced
VERBOSE: vms-c: 2016/10/16-16:26:03.267 Disks surfaced on node 'VMS-1': 12/12
VERBOSE: vms-c: 2016/10/16-16:26:03.298 Disks surfaced on node 'VMS-2': 12/12
VERBOSE: vms-c: 2016/10/16-16:26:06.945 Waiting until all physical disks are reported by clustered storage subsystem
VERBOSE: vms-c: 2016/10/16-16:26:10.188 Physical disks in clustered storage subsystem: 0
All replies (25)
Friday, December 2, 2016 6:12 PM ✅Answered | 1 vote
To follow-up from the offline investigation...
It appears my problem was that I had a pre-existing Storage Spaces volume attached to the system. The Enable-ClusterS2D command hangs in this case, as it doesn't allow a pre-existing Storage Spaces volume in the cluster.
Monday, October 17, 2016 5:16 AM
Hi Danny,
Are all the disks recognized by host?
Have you installed the following update?
https://support.microsoft.com/en-us/kb/3192366
It seems there are not much information about this issue on Server 2016.
Anyway, as Server 2016 was just released, some features may not work properly.
Best Regards,
Leo
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact [email protected].
Monday, October 17, 2016 5:21 AM | 1 vote
All installed disks are recognized and mentioned in the verbose logs (there are 12 disks total, between the two systems).
Yes, that cumulative update you link to seems to be a predecessor to the 14393.321 cumulative update that is installed on the system.
Tuesday, October 18, 2016 5:36 AM
Hi,
I am trying to involve someone familiar with this topic to further look at this issue. There might be some time delay. Appreciate your patience.
Best Regards,
Leo
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact [email protected].
Thursday, October 20, 2016 5:24 AM
Dan,
Then step that it appears to be stuck at is for the disks to show up in the clustered subsystem to be used of a clustered storage pool. The timeout is around 30 minutes, how long did you wait for this to complete?
I haven't seen it hang in this specific place before. Please make sure you have all the windows updates that are available applied to both nodes.
Steven Ekren
Program Manager
Microsoft
This posting is provided "AS IS" with no warranties, and confers no rights.
Thursday, October 20, 2016 6:40 AM
I waited three hours the first time, then five hours the second time, then overnight on the third time. Between each attempt, I removed the cluster, rebooted the member nodes, then performed the new-cluster and enable-clusters2d.
All involved nodes had the latest WU updates per...
(Invoke-CimMethod -InputObject (New-CimInstance -Namespace root/Microsoft/Windows/WindowsUpdate -ClassName MSFT_WUOperationsSession) -MethodName ApplyApplicableUpdates).Update
Here is the update package list:
Package_for_KB3176936~31bf3856ad364e35~amd6410.0.1.210.0.1.0
Package_for_KB3199209~31bf3856ad364e35~amd64
Package_for_RollupFix~31bf3856ad364e35~amd64~~14393.321.1.5
Monday, October 24, 2016 10:12 PM
Regarding Steven Ekren's note above, the enable-clusters2d does NOT timeout on this issue after 30 minutes. I've left it running for hours. Never times out.
If I abort the enable manually, Storage Spaces is left in a non-working but partially activated cluster state... I can see devices across the cluster, but they are non-pool-able.
Subsequently disable-clusters2d will cause the entire storage system to reset and reconnect... Which tends to further the observation that something is being activated, but is non-functional.
I appreciate suggestions on how I may further diagnose this problem..
Sunday, October 30, 2016 7:31 AM
Hi,
Dan, you're not alone. Similar problem here, with slight differences. I'm using core version of Windows Server 2016 RTM, latest updates (the last is KB3197954). Enable-ClusterS2D always hangs on 'Waiting until SBL disks are surfaced', 27% complete. Tried the operation a few times on both nodes of my cluster.
I've noticed that command is always missing (I assume) the last SSD disk, i.e. all the HDD disks and SSDdisks-1 are shown on the list:
Number of claimed disks on node 'A2': 6/6
Number of claimed disks on node 'A1': 6/6, and
Disks surfaced on node 'A2': 11/12
Disks surfaced on node 'A1': 11/12 (partial verbose output of Enable-ClusterS2D, which is stuck on 27%), and
Get-PhysicalDisk in parallel session is missing one SSD of one node. In my configuration, CanPool is true. AND Enable-ClusterS2D does not timeout after 30 minutes, in all attempts.
Dan, I've noticed in your log: Number of claimed disks on node 'VMS-1': 6/2; is that six of two? Does not seem right. Not sure if this is the indicator of the issue.
Would be great if we could break down the entire Enable-ClusterS2D command into smaller functions/tests.
Any suggestions?
Thanks.
GinVai
Monday, October 31, 2016 12:03 PM
Update.
One of the trials produced:
Disks surfaced on node 'A1': 11/12
Disks surfaced on node 'A2': 12/12, but Enable-ClusterS2d stuck on 27%, anyways. Now, CanPool is false and CannotPoolReason is Verification in progress.
GinVai
Tuesday, November 1, 2016 6:08 PM
Just an update, I'm still seeing this with the 321.1.5 update from 10/28.
And, to follow-up on several posts above from GinVai, I am seeing all disks surfaced, as follows:
VERBOSE: vms-c: 2016/10/16-16:26:03.267 Disks surfaced on node 'VMS-1': 12/12
VERBOSE: vms-c: 2016/10/16-16:26:03.298 Disks surfaced on node 'VMS-2': 12/12
The percent complete number is different for me depending on what options I select when invoking Enable-ClusterS2D, so I'm not sure how to map your hang % to the numbers I see.
Clearly there is some bad behavior on the Enable-ClusterS2D.
I wish there were more suggestions from MS on how to further diagnose this problem.
Tuesday, November 1, 2016 8:43 PM
Few things to take a look at:
- Make sure you have the following Update installed on all nodes: https://support.microsoft.com/en-us/kb/3197954
- Run the cluster Validation tool to verify the configuration (or Test-Cluster cmdlet) and look at the C:\Windows\Cluster\Reports directory for the .HTM file
- Run the Get-ClusterLog cmdlet. Look for the section with the following header and please post the results: [=== SBL Disks ===]
- Please send the results of the following cmdlet: Get-PhysicalDisk | Select FriendlyName, SerialNumber, CanPool, Size, MediaType | ft
Thanks!
Elden
Wednesday, November 2, 2016 4:54 AM
1... It appears that link points to an ?older? version of the packages on the nodes:
Package_for_KB3199209~31bf3856ad364e35~amd6410.0.1.010.0.1.0
Package_for_KB3199986~31bf3856ad364e35~amd64
Package_for_RollupFix~31bf3856ad364e35~amd6414393.321.1.514393.351.1.5
Package_for_RollupFix~31bf3856ad364e35~amd64
Either way, I used this to get the packages:
(Invoke-CimMethod -InputObject (New-CimInstance -Namespace root/Microsoft/Windows/WindowsUpdate -ClassName MSFT_WUOperationsSession) -MethodName ApplyApplicableUpdates).Update
I see the packages above were retrieved from WU on 10/28, which is a day later than the KB you link to. Please advise if I am reading the contents of the MSUs wrong. I also see that there are no newer packages from WU, as of the time of this reply.
2... The cluster passed with a warning about there only being two nodes (it recommended four)... The command I ran was:
Test-Cluster –Node vm-1,vm-2 –Include “Storage Spaces Direct”,”Inventory”,”Network”,”System Configuration”
3. The get-clusterlog shows:
[=== SBL Disks ===]
DiskId,DeviceNumber,IsSblCacheDevice,HasSeekPenalty,NumPaths,PathId,CacheDeviceId,DiskState,BindingAttributes,DirtyPages,DirtySlots,IsMaintenanceMode,IsOrphan,Manufacturer,ProductId,Serial,Revision,PoolId,HealthCounters,
{f057cc67-5466-fd0d-9c66-f4a043f928cb},4,false,false,1,18,{00000000-0000-0000-0000-000000000000},CacheDiskStateNonHybrid,0,0,0,false,false,,TOSHIBA-RD400,E83A_9702_0000_0A20.,57CZ4102,{d240ec0e-b38c-46f8-8993-c17aba12e55f},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{1030dd85-fb3d-5fcb-a54b-db71d4b264fd},0,false,true,1,12,{00000000-0000-0000-0000-000000000000},CacheDiskStateNonHybrid,0,0,0,false,false,,ST8000AS0002-1NA17Z, Z8400Z3A,AR13,{00000000-0000-0000-0000-000000000000},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{43d8b197-15e7-2194-a081-d624f6b3a495},2,false,true,1,10,{00000000-0000-0000-0000-000000000000},CacheDiskStateNonHybrid,0,0,0,false,false,,ST8000AS0002-1NA17Z, Z84021DR,AR13,{00000000-0000-0000-0000-000000000000},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{0f7e17a2-c3dd-1c01-f367-4723f3b0e49e},1,false,true,1,16,{00000000-0000-0000-0000-000000000000},CacheDiskStateNonHybrid,0,0,0,false,false,,ST8000AS0002-1NA17Z, Z84020RZ,AR13,{00000000-0000-0000-0000-000000000000},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{b83fffc8-1202-05d7-e92a-b602f957aa0d},3,false,true,1,8,{00000000-0000-0000-0000-000000000000},CacheDiskStateNonHybrid,0,0,0,false,false,,ST8000AS0002-1NA17Z, Z8404F2L,AR13,{00000000-0000-0000-0000-000000000000},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{68e7a9d5-4e93-84a9-c42c-c5d773bd8bd5},5,false,false,1,13,{00000000-0000-0000-0000-000000000000},CacheDiskStateNonHybrid,0,0,0,false,false,,TOSHIBA-RD400,E83A_9702_0000_0A71.,57CZ4102,{d240ec0e-b38c-46f8-8993-c17aba12e55f},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
4. The disk listing:
FriendlyName SerialNumber CanPool Size MediaType
ST8000AS0002-1NA17Z Z84016CS False 8001524072448 HDD
ST8000AS0002-1NA17Z Z8404F2L False 8001524072448 HDD
ST8000AS0002-1NA17Z Z84020RZ False 8001524072448 HDD
Msft Virtual Disk False 4294967296 Unspecified
ST8000AS0002-1NA17Z Z8400Z3A False 8001524072448 HDD
ST8000AS0002-1NA17Z Z8402FQR False 8001524072448 HDD
TOSHIBA-RD400 E83A_9702_0000_0A71. False 1024081264640 SSD
ST8000AS0002-1NA17Z Z84023H5 False 8001524072448 HDD
ST8000AS0002-1NA17Z Z84021DR False 8001524072448 HDD
TOSHIBA-RD400 E83A_9702_0000_0AED. False 1024081264640 SSD
ST8000AS0002-1NA17Z Z84023JC False 8001524072448 HDD
Imation IronKey Wkspace 9 qE False 64023255040 Unspecified
TOSHIBA-RD400 E83A_9702_0000_0A20. False 1024081264640 SSD
TOSHIBA-RD400 E83A_9702_0000_0A84. False 1024081264640 SSD
Wednesday, November 2, 2016 7:00 AM
1. Get-Hotfix:
A2 Update KB3197954 NT AUTHORITY\SYSTEM 10/28/2016 12:00:00 AM
A1 Update KB3197954 NT AUTHORITY\SYSTEM 10/28/2016 12:00:00 AM
2. Test-Cluster -Node A1,A2 -Include "Storage Spaces Direct","Inventory","Network","System Configuration":
There are some warnings about network and this is a red line:
Enclosure connected to node A1 has the same unique identifier with enclosure connected to node A2.
3. SBL
DiskId,DeviceNumber,IsSblCacheDevice,HasSeekPenalty,NumPaths,PathId,CacheDeviceId,DiskState,BindingAttributes,DirtyPages,DirtySlots,IsMaintenanceMode,IsOrphan,Manufacturer,ProductId,Serial,Revision,PoolId,HealthCounters,
{b59a2415-2c86-31c3-6807-78ceef3da270},1,false,false,1,3,{00000000-0000-0000-0000-000000000000},CacheDiskStateDisabled,0,0,0,false,false,ATA ,MK0960GFDKT ,S2DENYAG902453 ,HPG0,{e44fa0be-9d1a-11e6-b049-806e6f6e6963},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{d1bf2b3c-603d-a5a8-e882-912f52d99333},2,false,false,1,5,{00000000-0000-0000-0000-000000000000},CacheDiskStateDisabled,0,0,0,false,false,ATA ,MK0960GFDKT ,S2DENYAG902424 ,HPG0,{e44fa0be-9d1a-11e6-b049-806e6f6e6963},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{97fc9f71-b01b-76ae-036a-20256010f5b4},4,false,true,1,9,{00000000-0000-0000-0000-000000000000},CacheDiskStateDisabled,0,0,0,false,false,HP ,EG1800JEHMD , 08GB3NPA,HPD4,{e44fa0be-9d1a-11e6-b049-806e6f6e6963},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{36d2ba85-fcd4-a4cd-3dbe-e21ee317f65f},6,false,true,1,13,{00000000-0000-0000-0000-000000000000},CacheDiskStateDisabled,0,0,0,false,false,HP ,EG1800JEHMD , 08GB8M6A,HPD4,{e44fa0be-9d1a-11e6-b049-806e6f6e6963},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{73e65396-9389-b2f4-49f4-a61643756ac8},5,false,true,1,11,{00000000-0000-0000-0000-000000000000},CacheDiskStateDisabled,0,0,0,false,false,HP ,EG1800JEHMD , 08GB757A,HPD4,{e44fa0be-9d1a-11e6-b049-806e6f6e6963},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
{906c1eb0-c39a-1f6e-b2e6-c5ec018f6178},3,false,true,1,7,{00000000-0000-0000-0000-000000000000},CacheDiskStateDisabled,0,0,0,false,false,HP ,EG1800JEHMD , 08GB8XJA,HPD4,{e44fa0be-9d1a-11e6-b049-806e6f6e6963},[R/M 0 R/U 0 R/T 0 W/M 0 W/U 0 W/T 0],
4. Overnight there is a change, now all disks can pool (again), although Enable-ClusterS2D still does nothing on 27%; I left the command running for a day. Get-PhysicalDisk:
FriendlyName SerialNumber CanPool Size MediaType
HP EG1800JEHMD 08GB8M6A True 1800360124416 HDD
HP LOGICAL VOLUME False 63989022720 SSD
HP EG1800JEHMD 08GB4J0A True 1800360124416 HDD
HP EG1800JEHMD 08GB757A True 1800360124416 HDD
HP EG1800JEHMD 08GB8XJA True 1800360124416 HDD
HP EG1800JEHMD 08GB7A1A True 1800360124416 HDD
HP EG1800JEHMD 08GB3NPA True 1800360124416 HDD
HP EG1800JEHMD 08GB7TBA True 1800360124416 HDD
HP EG1800JEHMD 08G9K1PA True 1800360124416 HDD
ATA MK0960GFDKT S2DENYAG902453 True 960197124096 SSD
ATA MK0960GFDKT S2DENYAG902424 True 960197124096 SSD
ATA MK0960GFDKT S2DENYAG903077 True 960197124096 SSD
ATA MK0960GFDKT S2DENYAG902565 True 960197124096 SSD
Now, I will try to create storage pool.
Cheers
GinVai
Wednesday, November 2, 2016 11:59 PM
@Dan KB3199986 is the servicing stack update which is a prereq for KB3197954. Looks like that installed... I'm guessing that if you reboot, the update for the servicing stack update will complete... then Windows Update should pick up KB3197954
I'm a little puzzled by the 2-node warning. We announced support for 2-node S2D at Ignite, and made updates to Validate to remove that warning accordingly. I'm guessing once you get KB3197954 installed, that should go away.
In the logs you can see that all 12 of the disks get surfaced, but the cluster storage subsystem is returning 0... so Enable-ClusterS2D is waiting on that.
VERBOSE: vms-c: 2016/10/16-16:26:03.267 Disks surfaced on node 'VMS-1': 12/12
VERBOSE: vms-c: 2016/10/16-16:26:03.298 Disks surfaced on node 'VMS-2': 12/12
VERBOSE: vms-c: 2016/10/16-16:26:06.945 Waiting until all physical disks are reported by clustered storage subsystem
VERBOSE: vms-c: 2016/10/16-16:26:10.188 Physical disks in clustered storage subsystem: 0
Can you run Get-StorageSubSystem... and run it on each node just to verify that the results are consistent.
It's also odd that CanPool is returning False for all the drives. They should be returning True
I also see you have a USB and looks like a VHD local mount, which are returning a MediaType of unknown... that shouldn't cause issues. But might want to remove them when you try again... just me being paranoid.
But get KB3197954 installed first... rebooted...
Thursday, November 3, 2016 12:13 AM
@GinVai The Error in Validate about Enclosure ID's not being unique is a problem. Can you tell me more about your hardware configuration? What are you using for an HBA? Are the drives internal or in an external JBOD?
Can you run the following?
Get-CimInstance -Namespace root/WMI -ClassName ClusPortDeviceInformation | FT Dev*,NumberOfPaths,SerialNumber
Thursday, November 3, 2016 6:54 AM
Hi, Elden,
This is HP Apollo r2800, 4 servers and 1 storage enclosure in one cage. Network adapter is based on QLogic 57810S chipset; HP FlexFabric 10Gb 2-port in each server. RAID controller (P44X series) is set to HBA mode and disks are presented for OS as SAS BusType, and disks are not shared between servers, just packed in one enclosure.
If I run Enable-ClusterS2D, it stops at 27%, but after a while, Get-PhysicalDisk in parallel session, reports CanPool=True.
Get-CimInstance... output is:
DeviceAttribute DeviceGuid DeviceNumber DeviceState DeviceType NumberOfPaths SerialNumber
288 {36d2ba85-fcd4-a4cd-3dbe-e21ee317f65f} 5000 3 0 1 08G...
288 {73e65396-9389-b2f4-49f4-a61643756ac8} 5001 3 0 1 08G...
288 {97fc9f71-b01b-76ae-036a-20256010f5b4} 5002 3 0 1 08G...
288 {906c1eb0-c39a-1f6e-b2e6-c5ec018f6178} 5003 3 0 1 08G...
296 {d1bf2b3c-603d-a5a8-e882-912f52d99333} 5004 3 0 1 S2DENYAG902...
296 {b59a2415-2c86-31c3-6807-78ceef3da270} 5005 3 0 1 S2DENYAG902...
1 {00000000-0000-0000-0000-000000000000} 5006 3 1 1 CZ3601TMV3
0 {00000000-0000-0000-0000-000000000000} 5007 3 1 1
288 {9aaaca60-f8a0-7a0e-c2b5-4d4470b61763} 5008 3 0 1 08G...
288 {9a07221c-1b4e-3e1f-6956-1e286113bc9a} 5009 3 0 1 08G...
288 {6ea8e4b0-4718-8a96-b4e1-72ba48455c22} 5010 3 0 1 08G...
288 {9565fa28-5bda-659e-86af-359e815d8a83} 5011 3 0 1 08G...
296 {f0dc3fad-e069-ad9f-f267-c94b32f8f42a} 5012 3 0 1 S2DENYAG902...
296 {ec109809-6df9-65fd-22dd-1510b3f8753b} 5013 3 0 1 S2DENYAG903...
1 {00000000-0000-0000-0000-000000000000} 5014 3 1 1 CZ3601TMV9
0 {00000000-0000-0000-0000-000000000000} 5015 3 1 1
Can you, please, elaborate more on Enclosure ID uniqueness, why there is a problem other than this reduces availability of disks in case of enclosure's failure, somehow.
Thanks.
GinVai
Friday, November 4, 2016 7:18 AM
Elden, thanks for the suggestions. No joy though... here is what I've found:
1. The warning I received for Test-Cluster does not repro with the most recent RSAT for Win10 tools (I was using TP5 version of the tools). I reran all my steps from a full win2016 box, just to make sure there weren't any other oddnesses from the tp5 RSAT on win10. Same outcome, though.
2. I rechecked that there are NO packages showing "PendingInstall", but rebooted for the umpteenth time, and reissued the WU apply command. Nothing more came down as of 11/4 @ 6:00p PDT. Not a surprise, but a mystery about why the above package isn't coming down.
3. The KB is installed, I have reconfirmed... The manually downloaded KB3197954 x64 MSU file has this:
<assemblyIdentity name="Package_for_RollupFix" version="14393.351.1.5" language="neutral" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35"/>
<source location="%configsetroot%\Windows10.0-KB3197954-x64.CAB" />
And from the nano server's package MUM file for "Package_for_RollupFix~31bf3856ad364e35~amd64~~14393.351.1.5" :
<assembly xmlns="urn:schemas-microsoft-com:asm.v3" manifestVersion="1.0" description="Fix for KB3197954" displayName="default" company="Microsoft Corporation" copyright="Microsoft Corporation" supportInformation="http://support.microsoft.com/?kbid=3197954" creationTimeStamp="2016-10-21T16:02:17Z" lastUpdateTimeStamp="2016-10-21T16:02:17Z">
<assemblyIdentity name="Package_for_RollupFix" version="14393.351.1.5" language="neutral" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35"/>
<package identifier="KB3197954" applicabilityEvaluation="deep" releaseType="Update" restart="possible" psfName="Windows10.0-KB3197954-x64_1.psf">
4. The USB you saw in the device list is the boot device plugged into the chassis MB. This is a Nano hyper-v host, so the boot is tiny and separate from all the eventual high-perf VM storage. No getting rid of that, unfortunately... Since all the other disks in the system will be storage spaces.
5. I see this with GetStorageSubSystem ...
[vm-1]: PS C:\ Get-StorageSubSystem
FriendlyName HealthStatus OperationalStatus
Windows Storage on VM-1 Healthy OK
Clustered Windows Storage on vm-c Healthy OK
[vm-2]: PS C:\ Get-StorageSubSystem
FriendlyName HealthStatus OperationalStatus
Windows Storage on VM-2 Healthy OK
Clustered Windows Storage on vm-c Healthy OK
6. I'm open to suggestions on how to diagnose the CanPool.
Friday, November 4, 2016 4:44 PM
Elden, sorry for being so verbose in the previous posting. The KB3197954 package you're suggesting is installed.
From the meta-data in the KB MSU file:
<assemblyIdentity name="Package_for_RollupFix" version="14393.351.1.5" language="neutral" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35"/>
<source location="%configsetroot%\Windows10.0-KB3197954-x64.CAB" />
And from the nano server's package MUM file:
<assembly xmlns="urn:schemas-microsoft-com:asm.v3" manifestVersion="1.0" description="Fix for KB3197954" displayName="default" company="Microsoft Corporation" copyright="Microsoft Corporation" supportInformation="http://support.microsoft.com/?kbid=3197954" creationTimeStamp="2016-10-21T16:02:17Z" lastUpdateTimeStamp="2016-10-21T16:02:17Z">
<assemblyIdentity name="Package_for_RollupFix" version="14393.351.1.5" language="neutral" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35"/>
<package identifier="KB3197954" applicabilityEvaluation="deep" releaseType="Update" restart="possible" psfName="Windows10.0-KB3197954-x64_1.psf">
This is the package_for_RollupFix~31bf3856ad364e35~amd64~~14393.351.1.5 in the package listings.
So, Now I'm back to square one.
Friday, November 4, 2016 8:15 PM
I think we've reached the extent of what we can do in a simple forum... we need to dig in here, get like a Skype session going. Best would be if you guys could open a support case? Or if you have contacts to Microsoft (TAM, whatever...) just say "Yo, connect me with Elden"! Or just shoot me email eldenc
Thanks!
Elden
Friday, November 4, 2016 10:57 PM
Elden, I have opened a case for this issue through Win2016 support. I have given them your alias.
Monday, November 7, 2016 8:08 AM
Enable-ClusterS2D always hangs on 'Waiting until SBL disks are surfaced', 27% complete. Tried the operation a few times on both nodes of my cluster.
I have the same problem, but I have a test cluster assembled in vmware workstations
The problem was the same serial number of discs that assigns vmware workstations.
The Assembly of the cluster in VirtualBox drives are assigned serial numbers are different.
To check on each node run
Get-PhysicalDisk | ? CanPool -eq $ true
The vmware workstations need to edit configuration files of virtual machines. They should not have the same numbers of disks.
I hope this help you
Monday, November 7, 2016 5:27 PM
Thanks for your suggestion Bogdan.... But it appears your solution won't apply to my problem:
1. My runs hanging later in the sequence... "waiting until all physical disks..." (not SBL)
2. This is not virtualized hardware. There is nano on BARE METAL.
3. The SN's do not match on any of the disks
4. It isn't clear if you ran test-cluster, I'm curious what is reported there.
Friday, November 11, 2016 5:28 PM
Hi Bogdan,
Just curious... did you follow the steps outlined in this blog when you were setting up an S2D guest cluster in VMs on top of ESX hosts?
Thanks!
Elden
Saturday, March 11, 2017 5:14 PM
Hi Dan
how did you find this out and how did you solved this? I think i have same problem. I have 4-node bare metal installation and i know, that I tired StorageSpace (for reference) on one of the nodes before configuring cluster. Now i'm stuck at 27%.
Thank you!
Henri
Sunday, March 12, 2017 2:43 AM
Hi Dan,
What is the make and model of the HBA and servers you are using? The most common reason is incompatible hardware...
Also be sure to run the cluster validation tool, through Failover Cluster Manager or Test-Cluster.
Thanks!
Elden