Storage Pool Warning

Saturday, April 28, 2012 4:15 PM

i have this exact same problem. started last night.

i have two storage groups, one is completely healthy.

the other shows healthy except for the virtual disk, which says "warning".

my disks are in roswill rs8v enclosures. since it started showing the warning, all the disks in both pools show constant active. i'm not running anything else, no virus scan, other software, etc.

ive rebooted, no change

ive tried right-clicking the virtual disk and done repair. the status briefly changes to "repairing" (i think), its displayed so briefly its tough to tell. the option to repair is still there and the virtual disk still shows warning.

like OP, looking for ideas/steps to troubleshoot/resolve the issue.

correction: the status changes to to "running (0%)" and the goes blank.

I tried ptting up a screen shot, but doesnt seem to allow me to do that.

Edited by Broncoaddict Saturday, April 28, 2012 8:59 PM

Saturday, April 28, 2012 11:06 PM

1

I tried copying some files to the pool with the warning and it hung for a minute then "blue screened" .

Said (paraphrasing) "windows was restarting due to a problem, for more information look up REFS_FILE_SYSTEM (REFS.SYS)"

Searching on REFS_FILE_SYSTEM just showed me a bunch of reviews about MS's new feature ReFS (which I am using on these pools).

Basically, it was as helpful as the "warning" health status...which is to say it wasnt (at least not to me).

It would be helpful if a little more info was added into some of these error messages.

Sunday, April 29, 2012 7:46 PM

2

this is weird.

after i rebooted, my other pool (my data pool that did not have a warning or error) now has the warning as well.

i can now not shutdown(says shutting down but just hangs

i can not get the storage spaces manager to refresh, just hangs on getting data...

i can not access the data drive, it causes explorer to hang. the original drive (movies) with the warnings i can access no problem.t

additionally, unless i remove (physically) 2 disks from the data pool, everything 'unhangs' though obviously that 'breaks' the space.

i left it running/hanging overnight and now it appears that the movies space is longer constantly active and shows completely heathy. is it that 'self healing' stuff?

i now have additionaly questions.

both of my spaces have hot spares. how come it did not add it in when i removed a drive.

if it 'self-healing', is there a way to tell that its fixing things or monitor the progress?

Edited by Broncoaddict Sunday, April 29, 2012 7:47 PM

Monday, April 30, 2012 1:44 PM

0

My virtual disk eventually went healthy.

This is with out any intervention by me.

Tuesday, May 01, 2012 4:42 PM

0

I checked this morning, and all my are now in the green as well. I'm guessing it "healed". Would be nice to have some kind of indication of activity. I thought I had lost my data and that storage spaces was hung/on the fritz.

Wednesday, May 02, 2012 4:30 PM

1

I spoke to soon. While it showed healthy it does not appear to in a happy place.

In one folder, I can copy/access some text files, while others I can not. Specifically, it "hangs" and after about 5-10 minutes, the system would reboot.

This time it told me to do a search for DPC_WATCHDOG_VIOLATION.

I did this several times with the same results.

The reboots also took a tremendous amount of time anywhere between 15-30 minutes. Its directly related to storage spaces because if I do not have those arrays connected the system is fine.

I did some digging on DPC_WATCHDOG_VIOLATION and as far as I could tell it had to do with a bad driver or mismatch of some kind.

So having no other ideas, I uninstalled all my drivers and reinstalled them (reboots in between) but no change in my results.

So I figured something was screwed up and maybe a new install would clean/clear things up.

Based on everything I've read, your supposed to be able to move storage spaces from one system to another with out doing anything special.....well it might not be that transparent.

I performed a clean install and now everything boots and reboots cleanly as normal. Except for one little problem. I can see the storage pools and Virtual disks in server manager, but it doesn't show any of my volumes. When I look at the status for the virtual disk it says UNKNOWN.

When I try to "repair virtual disk" it acts the same as before, briefly display's "running 0%" and then nothing.

At this point I don't know if I'm making things better or worse. Any help/guidance would be greatly appreciated.

Thursday, May 03, 2012 9:08 PM

1

Here's some more info I pulled out via powershell, though I'm not sure of exactly what I'm looking at.

The things that standout to me are

IsReadOnly = True for the storage pool,

OperationalStatus = Detached and HeathStatus = Unknown for the virtual disk.

I assume, for whatever reason, I need to manually attach (mount?) the disk. Looking through the online doc's, I can find what to do if manually attaching does not work, but no mention on HOW to manually attach. :-(

Again, any help would be greatly appreciated...

Thanks,

PS C:\Users\Administrator> get-storagepool -FriendlyName "Data - SP" | fl *

Usage : Other
OperationalStatus : OK
HealthStatus : Healthy
ProvisioningTypeDefault : Fixed
SupportedProvisioningTypes : {Thin, Fixed}
ReadOnlyReason : By Policy
ObjectId : {60adcadb-7b36-11e1-b9f3-c860008e3760}
PassThroughClass :
PassThroughIds :
PassThroughNamespace :
PassThroughServer :
UniqueId : {60adcadb-7b36-11e1-b9f3-c860008e3760}
AllocatedSize : 7998839717888
ClearOnDeallocate : False
EnclosureAwareDefault : False
FriendlyName : Data - SP
IsClustered : False
IsPrimordial : False
IsReadOnly : True
LogicalSectorSize : 512
Name :
OtherUsageDescription : Data Stuff
PhysicalSectorSize : 4096
ResiliencySettingNameDefault : Mirror
Size : 11997454270464
SupportsDeduplication : False
ThinProvisioningAlertThresholds : {70}
PSComputerName :
ComputerName : WIN-E6MB4D51MON
ClassName : MSFT_StoragePool
Class : ROOT/Microsoft/Windows/Storage:MSFT_StoragePool
CimClass : ROOT/Microsoft/Windows/Storage:MSFT_StoragePool
Namespace : ROOT/Microsoft/Windows/Storage
Properties : {ObjectId, PassThroughClass, PassThroughIds, PassThroughNamespace...}
CimInstanceProperties : {ObjectId, PassThroughClass, PassThroughIds, PassThroughNamespace...}
CimSystemProperties : Microsoft.Management.Infrastructure.CimSystemProperties

PS C:\Users\Administrator> get-storagepool -FriendlyName "Data - SP" | get-virtualdisk

FriendlyName ResiliencySettingNa OperationalStatus HealthStatus IsManualAttach Size
me
------------ ------------------- ----------------- ------------ -------------- ----
Data - VD Parity Detached Unknown True 5.45 TB

PS C:\Users\Administrator>

Sunday, May 06, 2012 11:52 PM

1

well, i figured out how to manually mount and attach the pools and disk.

now, the computer reboots after about 10 sec after login with the ole 'search online for REFS_FILE_SYSTEM' which is as helpful as it was the first time around.

the only way i can stop it is to reinstall windows and not reattach the filesystem.

is it possible to get some kind of assistance from MS to figure out whats going on/what im doing wrong?

should i/can i open a problem ticket? i couldnt find any option for windows 8.

am i asking for help wrong or in the wrong forum? am i the only person hitting these errors?

i was led to believe this was where we should be coming for help, but so far i'm quite discouraged with the lack of response from anyone.

btw: I almost forgot, when I manually attached my first pool and virtual disk, the volume showed and I thought I was home free. Moments later I received a pop-up from stating "This version of windows doesn't support ReFS".

(These pools were created on windows 8 server and have never come up in contact with anything but windows 8 server)

When I manually attached the second pool is when it started into its reboot, login, reboot, loop.

Edited by Broncoaddict Monday, May 07, 2012 2:36 AM

Wednesday, May 16, 2012 12:52 PM

0

I tried to open a problem with MS on this but they couldn't because its in beta so needed to post in the forums and I would get a response within 24-48 hours. I pointed out I had done that and had yet to receive a single response in 10 days. I provided all the info they requested, a link, screenshot etc. While they did not know why it happened, they assured me my issue would get escalated and they would get a response for me.

That was over a week ago.

A few days ago, I even got a follow up call back saying it had been escalated....but still nothing.

This is very disheartening and not the experience I expected at all....

Thursday, May 17, 2012 5:57 PM

0

I think there might be two possible issues here. One or more disks might have failed and/or one or more disks' capacity is completely used. You can check on these two conditions by looking at the properties of the physical disks in your pool.

The solution for either condition is the same: add more disks. With more capacity among a sufficient number of disks Storage Spaces should be able to repair unhealthy spaces onto those disks. If all that was happening was a low capacity condition, then the additional capacity should be enough to fix the problem.

-Darren

Sunday, May 20, 2012 2:18 PM

0

I think there might be two possible issues here. One or more disks might have failed and/or one or more disks' capacity is completely used. You can check on these two conditions by looking at the properties of the physical disks in your pool.

The solution for either condition is the same: add more disks. With more capacity among a sufficient number of disks Storage Spaces should be able to repair unhealthy spaces onto those disks. If all that was happening was a low capacity condition, then the additional capacity should be enough to fix the problem.

-Darren

I don't see any bad disks and I do have a hot spare for each one.

Looking at the space available on each disk does show some puzzling results.

I have one pool with 5 data X 2 TB + 1 hot @ 2 TB. minus the parity disk that should amount to about a 7 TB virtual disk. Right? Its shows 5.45 TB.

4 of the data disks shows full and one shows empty (I thought it stripped the data?)

What makes it really strange is I only had about 3 TB of data in the pool.

I see the same behavior in my other pool. It has 7 data X 2 TB with parity, they are all full, but I only had 7 TB of data in the pool.

How or what is using all the diskspace. Is there some kind of logging going on?

I will add some more drives and see what happens.

Thanks Darren!!

Sunday, May 20, 2012 4:05 PM

0

i added a disk to each pool, i im back to the previous problem.

when i mount the smaller pool, i get the message that Refs is not supported in this version of windows. (weird as it has only ever been attached to the same version/release of windows server 8 beta)

when i mount the other pool all apears to be good for a moment, then it reboots with message: for more info search online for REFS_FILE_SYSTEM

Sunday, May 20, 2012 9:44 PM

0

Found some errors in the event log for the small storage group (the one that says its not supported..)

and

The file system detected corruption on volume Z: and was not able to correct it.

Volume Z: is formatted as ReFS but ReFS is unable to mount it; ReFS encountered status The volume repair was not successful..

Sunday, May 20, 2012 11:19 PM

0

I misspoke previously when I said I would mount the large volume and it would reboot. To clarify, mounting it did not cause a reboot. It was after accessing the volume it would then reboot. Specifically, I could see the root directory and folders, but once I attempted to view any of the folders, it would show me the contents AND THEN reboot.

I mounted the volume again (after rebooting), but this time instead of accessing via explorer, I opened power shell and used robocopy to try and copy a folder somewhere. So far so good, its copying my files without issue.

I don't know what the difference is between robocopy and explorer, but hopefully it keeps working.

I did not try to navigate to the drive via the prompt to see if that gives the same results. Didnt think about it until after I kicked of the copy. I'll try it when the copying is done, at this point I don't want to take the chance of it rebooting.

Monday, May 21, 2012 12:35 AM

0

I've been experiencing a similar issue with Storage pools in my case the pool completely dissappear and in place of these I'm left with my HDD's in the error state allocated to the primordial pool. When looking at the volumes they have the volume/partiton type of system protection partiton -- I had been using Deduplication on the volume and it was scheduled to dudupe in background and I was actively copying many gigs of data when the server crashed/rebooted itself. After it restarted the pool was gone and there was no way to get it going again. It seems the pool metadat on the HDD's got corrupted in some way. This is the 3rd time something like this has happened to me twice with a 2 drive stripe and once with a simple 1 HDD pool. In all cases I've been using thin and deduplication and the server has been crashing. I've disabled write cache on the HDD's but that did not stop the corruption or loss of the pools.

Monday, May 21, 2012 12:57 AM

0

It ended up rebooting and I was able to verify, that using powershell and navigating the directories from the command line does not initiate a reboot.

Reran robocopy after the reboot and it finished the directory, so while I don't know whats triggering the reboot, I'm hoping I'll be able to continue this until I have everything off.

Anyt,

I'm using fixed with parity, no deduplication, and REFS on the volume group. Are you using NTFS or REFS? Have you looked at the status via powershell?

get-storagepool (shows the storage pools)

get-virtualdisk (shows the virtual disks)

I'm suspecting in my case the problem isn't with storage pools, but with REFS.

I'm thinking you may have a different problem simple because my storage poois have always shown up without issue. Also I've never had any of my disks drop and get added back into the primordial pool, its getting the volume mounted and accessing that I'm having a problem with.

Monday, May 21, 2012 7:21 AM

0

So far I've only used ntfs, but I might try again with ReFs to see if any better. I'm wearing my drives out re-copying data from backups. There is no way this system is stable as it is in the beta now. I've looked at all those PS commands and more and not been able to find any trace of the storage pool when the metadata is lost. There should be some kind of way to recover the data from these drives when the storagepool is lost, but I've not been able to see any diagnostic tools for this that I can use and naturally no info about this scenario available on the web.

Wednesday, May 23, 2012 1:34 AM

0

Right now I don't know what or where the problem is. Did ReFs or Storage Spaces cause the problem or did they save me from something worse?

I got lucky with being able to access my drives via powershell/robocopy and copy files I wanted to keep off, I still don't know whats different between that and using explorer. I imagine its the method they use to access the system.

Maybe my expectations were too high or just out of whack with reality, but I envisioned MS having more involvement in working through these issues/bugs as it would benefit them as well.

I expected to have problems, I did not expect apathy.

Wednesday, May 23, 2012 6:39 AM

0

Hello Broncoaddict, mark

For now i experience same issue. After rebuilding system and attaching storage spaces to it i was able to access all data.
So i decided to continue moving data from one vdisk to another (from simple to parity).
After some time computer is stuck and I rebooted it by pressing reset.
After reboot and playing around with issue i'm able to acess my vdisks configured as simple (stripe), but can't access my main disk named Data (parity cofiguration).
Hard disks doing something (per task manager i see constant reading) but i can't find a way to see a progress or something like that.
mark3grahams and you mentioned that after some time everithing looks green and you able to access data, but how long you wait?

MCSE 2003: Messaging, Security
MCITP: Server Administrator, Enterprise Administrator, Virtualization Administrator
MCITP: Consumer Desktop Support, Enterprise Desktop Support, Enterprise Desktop Support Windows 7, Enterprise Desktop Administrator
SMS/SCCM, ISA 2004/2006, CCNA Security

Wednesday, May 23, 2012 12:37 PM

0

I don't know exactly, but I believe it was between 24-48 hours until if finally went green.

I was never able to find any way to determine what it was doing nor its progress.

Even though both of mine are green,

- One of them I was able to access briefly, for about an hour or two, but never since

(I get a message saying it needs to be formatted or its not supported in this version of windows)

- The other one, if I access via explorer (ie my computer) it reboots, using powershell or robocopy it was "stable", (it rebooted once or twice)

Wednesday, May 23, 2012 1:32 PM

0

Have you tried to run some recovery tool like R-Studio on vdisk which you can't access to try to recover data?

MCSE 2003: Messaging, Security
MCITP: Server Administrator, Enterprise Administrator, Virtualization Administrator
MCITP: Consumer Desktop Support, Enterprise Desktop Support, Enterprise Desktop Support Windows 7, Enterprise Desktop Administrator
SMS/SCCM, ISA 2004/2006, CCNA Security

Wednesday, May 23, 2012 5:33 PM

0

I download the demo version and have it running now. It is accessing the VD and appears to be finding files. Looking at the files they appear to be intact from what I can tell.

It's estimated 2d 18 hrs to completion, but I'm going to let it run to completion. I don't have anything I need to have and am tired of messing with it.

Storage spaces/ReFS seem like really great features, but with all the problems I've had and the lack of response from MS, I don't see a point in continuing.

I may try it again after its RTM's and been out for a year or two. I know it's beta, but it seems to be awfully buggy if they are planning on releasing in the fall.

Another reason for not rushing in is Metro: I've somewhat gotten used to it and there are a couple of nice things about it, but I find it gets in my way more times that not. If there's no way to turn it off, I'm in no hurry to move to windows 8

Wednesday, May 23, 2012 8:38 PM

0

As per my previous experience with this tool - demo version allows you to recover only part of data (some amount of files or Gbs). But you can save search results and reopen it later if you decide to buy program and use it to restore data

MCSE 2003: Messaging, Security
MCITP: Server Administrator, Enterprise Administrator, Virtualization Administrator
MCITP: Consumer Desktop Support, Enterprise Desktop Support, Enterprise Desktop Support Windows 7, Enterprise Desktop Administrator
SMS/SCCM, ISA 2004/2006, CCNA Security

Sunday, May 27, 2012 7:30 AM

0

I've posted question and answer about my issue in following thread - http://social.technet.microsoft.com/Forums/en-US/winserver8gen/thread/0753f0a2-8dbe-44e5-b8ef-1d5db4b0d56a

MCSE 2003: Messaging, Security
MCITP: Server Administrator, Enterprise Administrator, Virtualization Administrator
MCITP: Consumer Desktop Support, Enterprise Desktop Support, Enterprise Desktop Support Windows 7, Enterprise Desktop Administrator
SMS/SCCM, ISA 2004/2006, CCNA Security

Storage Pool Warning

All Replies

Products

Resources

Solutions

Updates

Trials

Related Sites

Training

Certifications

Other resources

Support by product

Other support links

Not an IT pro?