Microsoft Enterprise Platforms Support: Windows Server Core Team
EPS Team Blogs
Product Team Blogs
Greetings CORE blog fans! It has been awhile so I thought it was time for another blog. In recent weeks, we have seen an issue where the Windows Server 2008 R2 storage validation test List All Disks is failing with a Status 87. Figure 1, is an example of what is displayed in the cluster validation report.
Figure 1: List All Disks failure in Cluster Validation Report.
This error is also reflected in the ValidateStorage log (Figure 2) located in %systemroot%\Cluster\Reports directory.
000016f4.00001714::01:02:06.180 CreateNtFile: Path \Device\HarddiskVolume2, status 87 000016f4.00001714::01:02:06.180 GetNtldrDiskNumbers: Failed to open device \Device\HarddiskVolume2, status 87 000016f4.00001714::01:02:06.180 GetNtldrDiskNumbers: Exit GetNtldrDiskNumbers: status 87 000016f4.00001714::01:02:06.180 CprepPrepareNodePhase2: Failed to obtain boot disk list, status 87 000016f4.00001714::01:02:06.180 CprepPrepareNodePhase2: Exit CprepPrepareNodePhase2: hr 0x80070057, pulNumDisks 0
Figure 2: ValidateStorage log entry
The decode for these errors is shown in figure 3.
# for decimal 87 / hex 0x57 : ERROR_INVALID_PARAMETER winerror.h # The parameter is incorrect.
Figure 3: Error decode
The cause for this failure to this point is unknown. What we do know is the path that is called out as seen in Figure 3: above always points to the 100 megabyte partition that is created at the root of the system drive. This partition is created by default and is in place to support BitLocker. The approved workaround is to assign a drive letter to the 100 megabyte partition and re-run the validation process. The List All Disks storage test should pass at that point. There is no adverse impact to assigning the drive letter to this partition. As a reminder, BitLocker is not supported in a cluster environment. This is documented in KB 947302. If an attempt is made to enable BitLocker in a cluster node, the error in Figure 4 is displayed.
Figure 4: Error when trying to enable BitLocker on a cluster node
I have an ‘ask’ of our readership. If anyone reading this blog can ‘on demand’ repro this issue, we want to hear from you. This goes beyond just telling us, “Yeah, I’ve had that issue myself.” I am interested in hearing from anyone who has perhaps manipulated a setting in their controller card that can either cause validation to fail in this way or make it pass. I am interested in hearing from someone who had this failure, changed a setting of some kind, either in software or hardware, and the error went away. Be sure to provide the details (Make and model of controller, Firmware and driver versioning information, steps to reproduce the issue, etc…)
As always, we hope this has been informative for you.
Chuck Timon Senior Support Escalation Engineer Microsoft Enterprise Platforms Support
I've been able to recreate this problem, although I do not know exactly how I did it. I built up 2 new Server 2008 R2 servers and was attempting to install a SQL 2008 cluster on it. Initially I was able to cluster the two servers with no problem and the validation reports came back fine. I had to destroy the cluster and I'm in the problem of rebuilding it and I'm now having this problem. What I have though is tha same 100MB partition that I initially had (and had no problems with my cluster), but now I have 2 0MB partitions listed in Disk Management. I've seen this before and was able to delete the partition, but this time it's telling me that the media is write protected and I can't delete them. My cluster validation report is giving the same error that you have listed above.
Those 2.0 MB partitions are something you see when you are using snapshots from the storage side. I have seen this issue earlier. Those vary in number, sometimes you may see two partitions and sometimes more. Just wanted to share this information with you.
Currently experiencing this problem using two Dell PowerEdge 2950's with Bios level 2.6.1 and an EqualLogic 4000 iSCSI SAN, the cluster disk will be connecting through iSCSI. We have loaded Server 2008 R2 on both servers and can not seem to get past this error. I am working with a Microsoft Engineer to see if they can help me and so far no luck. I will share any solution that we find later. Let me know if there is any more information that you would like
After trying many things, including giving drive letters to the 100 Megabyte partition, we were still having the same issue. We ended up re-formatting the SAN drive and after that the validation tests started passing. I would say this is a misleading error message.
100% same issue.... thanks for the workaround...
We discussed the topic here as well:
Added a drive letter to the 100MB partition worked for me! I am running a 2008R2/SQL2008R2 cluster with NetApp storage and SnapDrive. Thanks for the tip!