5121 during DPM 2010 backup

[guest]

I'm using DPM 2010 to backup our CSVs which reside on a HP P2000 G3 FC/iSCSI. CSVs on the Hyper-V cluster are accessed by FC only. iSCSI is used to connect the DPM server to the storage.

Log Name: System
Source: Microsoft-Windows-FailoverClustering
Date: 17.4.2011 12:39:47
Event ID: 5121
Task Category: Cluster Shared Volume
Level: Error
Keywords:
User: SYSTEM
Computer: server2.domain.local
Description:
Cluster Shared Volume 'Volume1' ('CSV10k') is no longer directly accessible from this cluster node. I/O access will be redirected to the storage device over the network through the node that owns the volume. This may result in degraded performance. If redirected access is turned on for this volume, please turn it off. If redirected access is turned off, please troubleshoot this node's connectivity to the storage device and I/O will resume to a healthy state once connectivity to the storage device is reestablished.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-FailoverClustering" Guid="{BAF908EA-3421-4CA9-9B84-6689B8C6F85F}" />
<EventID>5121</EventID>
<Version>0</Version>
<Level>2</Level>
<Task>38</Task>
<Opcode>0</Opcode>
<Keywords>0x8000000000000000</Keywords>
<TimeCreated SystemTime="2011-04-17T10:39:47.521872700Z" />
<EventRecordID>1197950</EventRecordID>
<Correlation />
<Execution ProcessID="2136" ThreadID="4360" />
<Channel>System</Channel>
<Computer>server2.domain.local</Computer>
<Security UserID="S-1-5-18" />
</System>
<EventData>
<Data Name="VolumeName">Volume1</Data>
<Data Name="ResourceName">CSV10k</Data>
</EventData>
</Event>


This happens during backup using Microsoft's Data Protection Manager 2010, which utilizes HP's VSS Hardware provider:

C:\Users\administrator>vssadmin list providers
vssadmin 1.1 - Volume Shadow Copy Service administrative command-line tool
(C) Copyright 2001-2005 Microsoft Corp.

Provider name: 'Microsoft Software Shadow Copy provider 1.0'
Provider type: System
Provider Id: {b5946137-7b9f-4925-af80-51abd60b20d5}
Version: 1.0.0.7

Provider name: 'HP StorageWorks P2000/MSA2000 VSS Provider'
Provider type: Hardware
Provider Id: {bd04cbf9-212c-4553-9ea5-c5bfb05ccc8f}
Version: 2.7.0.0



I'm in contact with HP and was told the storage (all latest firmware) has full support for CSV. Why am I even getting this error and is it something I should worry about. What additional steps can I take to resolve this. Is additional information, logs needed. What can I test, ask HP...

[guest]

no worries, normal operation.

you can find some great info here:

http://technet.microsoft.com/en-us/libr ... 34189.aspx

[guest]

Thank you for your reply and reasurance. It appears this is exactly it:

To reduce the impact on VM1, we recommend that you use hardware snapshots, which enable the CSV to resume direct I/O mode as soon as the hardware snapshot has been created. The duration of this process is typically very short, about two minutes.

Should there be an event log entry after resuming direct I/O, so I can double check everything?

[guest]

Hi, I also get this event as well as event ID 5120 and it sometimes causes my VMs to stop during backups.

I have a call open with MS regarding this issue. Apparantly 5121 is normal and can be ignored (then why it is logged as an error I will never know) but Ms have promised to provide a KB that explains further. I will post further details when I get the KB.

How many protection groups do you have for VMs and how many VMs are in each group?


--------------------------------------------------------------------------------

[guest]

Can you tell me how did you managed to get DPM 2010 backups working with MSA 2000 G3? I have the following scenario at one of my customers:

Hyper-V cluster with 4 nodes with 20 VM`s, MSA 2000 G3 FC/iSCSI storage and DPM 2010 on a phisical server. All my nodes are connected via a separate iscsi vlan with jumbo frames, also my DPM 2010 is connected via iSCSI with the storage. I installed capi->VDS-VSS from hp.com website on all hypervisors, i created a vdisk->volume->snap pool for VM backups on the MSA 2000 and mapped it directly to DPM.

After i create protection group for any VM in the cluster it enters OK state after 1 min ( considering that the smallest machine has 30 GB it`s impossible to create a replica under a minute, BTW i dont use 10 GB network) and if i do consistency check it fails with ID 30115 ( http://social.technet.microsoft.com/For ... 2f34ff43b/ )

I checked the storage and it`s licensed for 64 snapshots with default license.

Please help me with this matter.

[guest]

not sure I'll be able to assist you, but let's give it a go. If needed let me know where additional information can be exchanged (e-mail?), I'll gladly provide screenshots, configuration details if that might help.

A snap pool that is created must be owned by the same controller as are the vdisks/volumes you're trying to backup. I didn't map it to DPM, since the purpose of the snap pool is so that the MSA has a space to create hardware snapshoot of the VMs on it. DPM will create the snapshots and delete them afterwards. Make sure that the volumes you're trying to snapshot are master not standard volumes or it won't work (it might be that DPM converts them but I'm not 100% sure).

HP MSA2000 CLI guide if you need help with conversions http://bizsupport1.austin.hp.com/bc/doc ... 520779.pdf

Do you see the snapshots being created in the web interface when DPM starts backing up?