Single controller RAID, LUNs, virtual disk groups question does?

Similar Questions

• PowerEdge 2950 advanced settings recommended virtual disk group

  Hey guys, I was hoping that someone out there has had some information about the nature best practices or the best settings for these two configurations of a Dell 2950 with PERC 5 controller / i.

  Senario 1

  Reference Dell 2950 Windows Server 2003 Enterprise Edition 32-BIT

  RAID 5 (5 TB) hardware RAID & Windows dynamic disk Configuration

  PERC 5 / i with battery

  Group of virtual disk current avancΘs

  Virtual disk 0 - Adaptive Read Ahead and Write Back

  Virtual disk 1 - Adaptive Read Ahead and Write Back

  Senario 2

  Reference Dell 2950 Windows Server 2003 Enterprise Edition 32-BIT

  RAID 5 (5 TB) hardware RAID & Windows dynamic disk Configuration

  PERC 5 / i with battery

  Group of virtual disk current avancΘs

  Virtual disk 0 - Adaptive Read Ahead & direct writing

  Virtual disk 1 - Adaptive Read Ahead & direct writing

  The second server now has a battery that has failed and we are on the way. I want to add that I just found out that a bad or dead battery kills performance on the server reading and the writing speed. After replacing the battery on the first server, I saw performance increase by at least 300% if it is longer. So this lead me to read a lot of posts by others, but I couldn't find a solid peg that indicates what should be the best configuration. So my questions are... What is the best configuration for Virtual Disk Group advanced settings for MY servers above WITH a controller battery? Finally, what happens if I didn't use the battery which is recommended then? Is it better to use a battery or be without one? Bottom line is that I best performance of reading and writing, thank you to you all!

  300% is too much of a performance increase to attribute entirely to the write cache is enabled, even if it makes a significant difference.  When the battery goes bad, the controller disables the write cache, and it is the source of the performance difference.  Unless you have a specific reason to disable the cache (and I see no good reason), it must be enabled.  You can not force the controller to use memory cache without battery, not to mention that it's dangerous for your data to use the cache without battery.

  "Bottom line is that I the highest performance for reading and writing, thank you to you all!

  Bottom line... turn write cache (write back) and replace the batteries when needed (which should not be often).

• RAID 1 / Raid 5 / virtual disks

  I'm working on a PowerEdge T610 with Raid 1 and Raid 5 server using a Perc 6 / i controller.  I am preparing to install a new drive to increase the raid 5 array.  I am aware that I need to install the hot player and use the task reconfigures via storage/Virtual disks.  However, I have a question.  When I "Blink" virtual disk 0 or 1 virtual disk, all five discs blink.  -Can I use the task to "Reconfigure" virtual disk 1 or is it important?

  Thanks for your help!

  Hello

  aforby

  When I "Blink" virtual disk 0 or 1 virtual disk, all five discs blink.  -Can I use the task to "Reconfigure" virtual disk 1 or is it important?

  Yes, it's important. I would not rely on the blink function to determine which virtual drive is that. I suggest that you review the details for the virtual disks and verify that hard disks are part of each virtual disk. Details should also list the virtual disk's RAID level.

  Thank you

• MD 1220 / PERC H800 - how Raid 1 virtual disks?

  Hello I'm trying to configure Raid 12 1 (two pairs disks) in my Inbox MD 1220 with PERC H800. I started with 12 discs as 12 virtual drives Raid 0 and have recently added 12 additional physical disks for a mirror. I am able to mirror 9 virtual disks but when I select my 9th virtual to the mirror, no drives appear as available. The other 3 drives are there in the 'ready' State, but manage to open or the BIOS configuration utility seem to be able to use them as candidates of mirror. My computer is running in a kind of limitation of Raid 1 with Cabinet / card? Operating system: Windows 2008 R2 Datacenter Open Manage 7.4 Latest firmware / drivers for PERC H800 thanks! Dan ouedraogo

  Hi Joey

  I just tried OMCleanup and reinstalled 7.4.  He exhibited the same symptoms showing is not the physical drives in the H800.  I can't find a download 7.2 but would be willing to try the previous versions.  A previous version of OM worked with the same controllers / speakers / computer, but I do not remember the version where I improved.  (I am an individual shop and am not very informed when it comes to managing open, its versions and how to install/uninstall).

  I can still kill the DSM SA connection Service by running a foolish occupation for the H800 report.  The report H700 works fine.

  Dan

• PERC 710 add physical disks for RAID 5 virtual disk and keep data

  I think to set up a physical disk 4 RAID 5 array on my new PowerEdge T320.  That should give me enough space to last long.  I have two empty slots of HD.  If I need more space, later, I'll be able to add one or two additional physical disks without losing the data already on the virtual disk?  If so, how?  I don't see a way to do this in the manual of Perc 710.

  Thank you

  Andrew

  Yes, it is "doable" later.  This can be done in the PERC BIOS, but it can be done in OMSA (OpenManage Server Administrator) within the OS.

  One thing to keep in mind is the limit of 2 TB "floppy" Windows on BIOS systems compatible.  If you plan to 4x4TB disks in a RAID 10, that TB ~ 8 'disc' will be converted to the GPT and BIOS UEFI boot mode value.

• PowerEdge R720 virtual disk question

  Hi, our company has bought a Dell PowerEdge R720. I tried to use the lifecycle controller in RAID configuration. At the start, it has 3 500 GB SAS HARD drive, but I ordered 2 x 320 GB additional SAS being what I intend to do it, use the 2 x 320 GB like OS install as RAID 1 and RAID 5 on the default 3 x 500 GB, but for some reason any, whenever I have configure RAID , either on the two 320 or 3 x 500 GB, it shows just a virtual disk, 2 x 320 GB or 500 GB x 3, I see not two for RAID 1 and RAID 5 virtual disks. Am I missing something here? or I have to perform an extra step?

  Kind regards

  Jeff

  Hello Jeff,.

  If you try to install the operating system through the lifecycle controller method. You must have the latest lifecycle controller OS driver pack installed in your system. For the latest driver pack OS, please download from our support http://support.dell.com/ site by entering you service number and download OS Driver pack drivers.

  You can update the OS driver pack thorugh iDRAC GUI or through the lifecycle controller GUI and once after the OS driver pack update. When you try to deploy the operating system through the life cycle of the controller, you will appear on the list of supported operating systems and you can select the operating system you want to install by mounting the image operating system or DVD media connected to the system.

  Please let me know if you have any question.

  Thanks, Neeraj

• S130 rebuild with 2 virtual disks on the same disk array

  Hi all

  I have 6 drives 750 GB sata in raid 10 with a s130 controller.

  There is a 100 GB virtual disk and a virtual disk of 2 TB.

  One of the drives has failed, so I replaced it with a new drive.

  I have assigned the new disc on the first virtual disk, and this re-built very well. But he'll also let me assign the remaining space on the new disk on the second virtual drive in the table, it is said that there is no available free disk?

  What can I do here?

  Thank you

  Hi NET1K

  It seems that the hard drive that is not in the disk of your virtual disk group 2 (VD).

  You may need to put the reader that you have assigned to rebuild the first VD and assign it as a global hot spare to rebuild your disk, instead of virtual disk group offline.

  Before doing the procedure reversed, please backup you made recent data.

• Reference Dell MD3000 virtual disk

  I have an environment to both nodes of ESX 4.0 and a shared storage (Dell MD3000, no iSCSI).

  On storage, I have 10 at 10K SAS drives.

  Virtual disk how I need to build? A two bigger or smaller?

  Thank you

  Max

  It is not a simple answer, because it depends on:

  • Good storage practices

  • Storage type and the type of disks

  • VM number, size and type of e/s

  • Max VM and e/s per LUN

  • Size Max LUN (less than 2 to - 512 b)

  In your case, with 10 discs and two nodes and storage running in active-passive mode, you can create 2 virtual disks.

  Create 2 different groups of RAID5 (also think about using 1 disk as a global hot spare) and in each virtual disk group build 1 (or more if the size is more than 2 TB - 512b).

  If you have a few VMS with high IO think also to use use 4 disks in RAID10 instead of RAID5 configuration.

  André

• Virtual disk parent has changed since the child was created.

  I recently had one of my virtual machines go down and I couldn't connect to access anything whatsoever, so I rode as an extra hard drive to another virtual machine in order to access the data I needed.

  Once I had finished I he took off the other VM and proceeds to the power on the virtual machine, but I got an error of:

  "The virtual disk parent has been changed since the child was created." The content of the virtual disk ID parent does not match the ID of parent matching the child content. »

  Reason: The virtual disk parent has changed since the child was created. The content of the virtual disk ID parent does not match the ID of parent matching the child content.
  Cannot open disk ' / vmfs/volumes/4e24c6fe-3c2ad61e-a57c-e4115bbbce6c/SNCTrillium/SNCTrillium-000003.vmdk' or one of the snapshot disks it depends on.

  I read upward through the knowledge base of VMWare on the modified files, but I've never had to do that before and I'm not sure about how to go on this subject as I've never used the vsphere CLI either.

  I am currently using ESXi 5.0.0 with vsphere client 5.0.0. I've attached a screenshot of the error.

  According to the snapshot between SNCTrillium - 000001.vmdk and the basic disk SNCTrillium.vmdk is broken, which means that the basic disk has been changed. I have attached an instant replacement for the SNCTrillium - 000001.vmdk with a fixed string that should allow you to open the virtual machine. However, since I do not know what and how much data has changed on the basic disk, I can't tell you how much data corruption, you will see. So you better take an another snapshot to be temporary once loading hard file attached, in order to ensure the current hard data files cannot be changed.

  André

• Please help: the virtual disk parent has changed...

  After a few days of research about this error, I decided that I had to ask for some evidence of your experts to help me recover from this error.

  Thank you!

  It is a small desktop application:

  -Dell T410 esxi Server 5

  -a virtual machine, Windows Server 2003 with 2 discs, no fixed size

  I was help this office by managing services and updates on this Server 2003.  I ran updates on Friday and restarted.  When he came I open vSphere and found the virtual machine does not.  I tried to start it and got the error.  So, it seems I may have the CID issue that needs to be corrected.

  I don't have a direct access to this server so I'm remoting with RDS on 2003 VM server and vSphere on the esxi.  Did I make the right diagnosis and is it possible for me to make the necessary corrections to distance or I need direct access to the server?  Everything I've read has files of people editing CID numbers and it seems that they directly access the esxi server.

  Shows the data store:

  hard 'server '.

  hard 'server_1.

  'server' - 000001.vmdk

  "server_1" - 000001.vmdk

  'server' - snapshot1.vmsn

  I stuck a few newspapers below...

  DISC: Scsi0:0 [OPEN ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001.vmdk' persistent R]
  2012-08 - 05T 20: 00:06.022Z | VMX | DISKLIB-VMFS: ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001-delta.vmdk ': open with success (10) size = 5620404224, hd = 213826. Type 8
  2012-08 - 05T 20: 00:06.022Z | VMX | DISKLIB-DSCPTR: open [0]: "Server-000001 - delta.vmdk ' (0xa)
  2012-08 - 05T 20: 00:06.022Z | VMX | DISKLIB-LINK: open ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001.vmdk' (0xa): vmfsSparse, 35526656 sectors / 16.9 GB.
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-VMFS: ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-flat.vmdk ': open with success (14) size = 18189647872, hd = 181059. Type 3
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-DSCPTR: open [0]: 'server - flat hard' (0xe)
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-LINK: open ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server.vmdk' (0xe): vmfs, 35526656 sectors / 16.9 GB.
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB LINK: DiskLinkIsAttachPossible: incompatibility of the content ID (parentCID f72e60c7! = f05537b3) /vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001.vmdk vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server.vmdk.
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-STRING: ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server.vmdk ': cannot open (the virtual disk parent has changed since the child was created.) The content of the virtual disk ID parent does not match the ID of parent for children's content).
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-VMFS: ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001-delta.vmdk ': closed.
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-VMFS: ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-flat.vmdk ': closed.
  2012-08 - 05T 20: 00:06.035Z | VMX | DISKLIB-LIB: can't open ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001.vmdk' with the virtual disk 0xa flags parent has changed since the child was created. The content of the virtual disk ID parent does not match the ID of parent for children (18) content.
  2012-08 - 05T 20: 00:06.035Z | VMX | DRIVE: Cannot open disk ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001.vmdk ': the virtual disk parent has changed since the child was created. The content of the virtual disk ID parent does not match the ID of parent for children (18) content.
  2012-08 - 05T 20: 00:06.035Z | VMX | Msg_Post: error
  2012-08 - 05T 20: 00:06.035Z | VMX | [msg.disk.noBackEnd] unable to open drive ' / vmfs/volumes/4f4781a4-15d3bbb8-2681-782bcb6a6964/server/server-000001.vmdk' or one of the snapshot disks it depends on.
  2012-08 - 05T 20: 00:06.035Z | VMX | [msg.disk.configureDiskError] reason: the virtual disk parent has changed since the child was created. The content of the virtual disk ID parent does not match the ID of parent matching the child content.
  2012-08 - 05T 20: 00:06.035Z | VMX |

  Thanks again for any direction you can provide!

  / r

  John

  The parentCID of the snapshot must match the CID from its parent, as shown in the log file

  "... The content ID mismatch (parentCID f72e60c7! = f05537b3)... »

  Server.VMDK

  CID = f05537b3
  parentCID = ffffffff

  Server - 000001.vmdk

  CID = a2b9ba8b
  parentCID = f72e60c7---> f05537b3

  Make sure you change the file on the host using the vi editor or - if you want to change the file in Windows - use an editor who is aware of the line breaks Unix (do not use Notepad!)

  Once this is done, I highly recommend that you take another cliché before you on the virtual computer. This prevents existing virtual disk files to be modified, so that you can return to this State where you need. After that everything works as expected, consider deleting all snapshots.

  André

• Question on the addition of LUNS to the disk group

  The GI version: 11.2.0.3.0
  Platform: OEL
  
  I have to add 2 TB for the following disk group. Each logic unit number is 100 GB in size.
  
  100 x 20 GB is 2 terabytes. Which means that I have to run 20 orders as below
  

  SQL> ALTER DISKGROUP DATA_DG1 ADD DISK '/dev/sdc1' rebalance power 4;

  Each ALTER DISKGROUP... Rebalancing of the command Add DISK takes about 30 minutes to complete.
  Thus, this operation will take 10 hours to complete.

  ie. 20 x 30 = 600 minutes = 10 hours !!!

  I have a 3-node RAC. Can I run DISKGROUP ALTER... Orders of DISK to be added in parallel with each node? Will ASM instance allows
  several operations of rebalancing in a single disk group?

  http://docs.Oracle.com/CD/E11882_01/server.112/e18951/asmdiskgrps.htm

  Oracle ASM can rebalance of Group of disk at once on a given instance. If you launched balances multiple groups of different drives on a single node, Oracle then processes these operations simultaneously on additional nodes if necessary; otherwise, the rebalancing is performed in series on the single node. You can explicitly run balances on groups of different disks on different nodes at the same time.

  Why do you need to run 20 of these commands?

  You can not only add 20 records at one time:
  ALTER DISKGROUP DATADG ADD DISK
  "diska5/devices / ' NAME disk1.
  "diska6/devices / ' NAME disk2.
  "diska7/devices / ' NAME disk3.
  ...
  .. .etc

  You will need to should not manually rebalance as Oracle will automatically do the rebalancing when configuration changes unless you want to use a default value of such power as defined by the ASM_POWER_LIMIT parameter. You can run the command above with an option of rebalancing so if you you need to do.

• Reconfigure the virtual disk using OpenManage of RAID 0, RAID-5

  I apologize for the spacing.  I'm having a little wrong with this forum edit my post.

  My small business has had a T610 run for a few years with the following configuration:

  T610 Server

  PERC 6 / i controller

  4 physical disk 1 TB

  Virtual disk 0 size 1 800 GB RAID - 5

  Size of the virtual disk 1 992 GB RAID - 5

  Virtual Disk 1 was running Exchange and everything else on VD 0.

  Last weekend, I had a breakdown of table flat due to the failure of a damaged drive on Windows SBS 2008, who couldn't recover. When I contacted support Pro and got a new drive, I was told not to recreate the same virtual drives and use one.

  The local tech and I have been quite underlined and used the Perc 6 / I BIOS to recreate the exact configuration. Then, we had to use Backup Exec 2010 to restore the data. Somehow, we have selected by mistake instead of RAID 5 RAID-0 on recreation of the two virtual disks.

  So now, I have the following configuration in OpenManage:

  Virtual disk 0 size 1 800 GB RAID - 0

  Size of the virtual disk 1 992 GB RAID - 0

  I called Dell and gave instructions in OpenManage on how to reconfigure a virtual disk by adding a physical drive and conversion to RAID-5. My problem is reconfigure is not an option in the dropdown menu, I am told, because I have two virtual disks that have stripes.

  I have the ability to drive to go to 1 single virtual disk, given only Exchange is on the second and I can move, then unmold Virtual Disk 1. So my plan is:

  Full backup tape.

  Move Exchange off the drive sitting in virtual disk 1 to disk on virtual disk 0.

  Remove the virtual disk 1.

  Add a 5th physical HD.

  I hope at this point OpenManage will then give me the option in the drop-down list to reconfigure Virtual Disk 0.

  The option to Reconfigure in OpenManage will appear after that he sees only 1 virtual disk?

  Hello

  KRoy66

  The option to Reconfigure in OpenManage will appear after that he sees only 1 virtual disk?

  Yes, the ability to reconfigure should be available as soon as you remove the second table off readers. Technicians are you talked with correct you cannot reconfigure a table in slices.

  I know that you went through the entire process of relocation and restoration of backup, but it is preferable that do you it again. Let me explain why your table has been configured in this way, and why you may need to reinstall again.

  There are limits on the partition. To create a partition more of 2 TB, it must be TPG. In order to boot the system from a partition GPT Board must be set to UEFI mode. The problem you will encounter when trying to create a large RAID 5 is that you will not be able to use more than 2 TB of it. You create a new table in slices to use the extra space.

  If the system card is already set to UEFI mode then you should be able to convert the GPT partition and expand to necessary to hold account space. If you have installed the operating system with the system of administration established on the MBR can change in UEFI and conversion of GPT likely to cause BONE to be unstable if you are able to boot on it at all.

  I think you need to save yourself a lot of hassle and do this upgraded correctly. I know it will be a lot more time to restart, but you should set the UEFI boot mode, remove the two RAID 0, create the new large RAID 5, reinstall the OS and then restore the backup.

  Do it right now probably will save you a lot of time.

  Thank you

• two flash disks in a single disk group vsan

  Every morning.

  We had our second ssd flash not to shoot the entire disk group in vsan, not a problem all the resynced and recovered ok.

  I know that we can make groups of multiple discs for each host with its own flash ssd by group, my question, do we put two SSD flash in a single disk group? If so, if one of the two breaks down the entire disk group go offline? or will she continue to work ok until these two flash drives don't?

  Thanks in advance

  Nathan

  Band 2 will improve performance, the FTT2 will lower write back-end IOPS / s (perhaps a problem, perhaps not according to the back pressure and filling of the cache). In theory, it could improve read performance cache miss, but between the CRA, the system cache read-ahead for a absences from the cache and all the rest I do not expect much.

  Small groups of disks, etc. more flash. I would recommend you start observer VSAN and watch your hit rate Read Cache and see if another groups of disks (and Flash) would be a good idea.  You tend to get diminishing returns, trying to push beyond the 90% cache hit rate

  To keep in mind that, unless you have a balancing really bad object (it would take much effort to be honest attention creating VM and deletion and then affecting the PAF policies to create) you should see a heavy a lot to rebuild many (IE when a disk or group of disks is lost the pairs of mirror for these objects will be scattered across the cluster and they will rebuild in) free space of all other disks until mirroring policy is armed).  Keep in mind that the objects is already memorized block in new components never 256 GB (by default). Because of this design as the cluster becomes larger and more areas of failure (host, disk groups, readers) ads you expect actually rebuild performance better!

  Have you experienced problems of performance on your reconstruction, and if so what are your flash/disc types and make the model of your server HBA etc.?

• Cloning of a virtual disk before developing a RAID 5 system

  Hi all

  My apologies if this has been answered before, or there is a simple solution that I have not yet met.

  I have a Dell Poweredge 2900 with a Perc 6 / i RAID controller. I'll use OMSA to reconfigure a virtual disk RAID 5 I want to increase the storage of data on this drive, but I would like to the VD to backup before you go ahead and commit to reconfigure the disk (data stored on the VD is critical and I don't want anything to go wrong when the reconfiguration and lose all data).

  I have the RAID 5 VD on 4 discs and filled the other empty bays 4 identical drives and created a new VD on these new readers. Is it possible that I can use the BIOS RAID configuration or the OMSA to everything from the original to the new VD VD clone, or do I need a third party software?

  No, you need some sort of backup and imaging software.

• HP EVA virtual disk LUN ID

  I posted this is 3.0, but I wanted to try here also.

  We are in the process of upgrading of 4 guests isolated from 3.02 to 3.5. All network configuration is identical on all hosts, but the client has decided to present a separate HP EVA virtual virtual disk (LUN) for each host. Isolation was a contractual obligation. The question I have is on shared disk and LUN ID.

  I want to use VMotion to minimize downtime. Since they do not share currently disc, I need first to present all the LUNS (HP EVA virtual disks) on all servers.

  My question concerns the LUNS ESX ID. EVAs HP do not allow you to set the LUN ID the first (HP EVA virtual disk (LUN) presented to a host receives a LUN ID of next 1, 2, 3, etc..) In my case, the ID of the LUN on current drive of each host is set to 1. When I introduce all disks to all hosts now have a situation where A ESXHOST will see the LUN ID 1 for a 'DISK 1' and 2 ID LUN "disc 2", while ESXHOST-B could show exactly the opposite.

  It will be temporary because as soon as I can VMotion servers out, I will redo the presentation so all match however; There will be a short period where ID LUN will differ from a drive on different hosts.

  Is that going to be a problem?

  Hello

  You can set the LUN ID on the EVA only when he introduced a LUN to just one host at a time. It's a bit annoying, especially when you have a lot of guests. But the fact in this way you can ensure that all LUNS are presented with the same ID for the guests.

  If you select multiple hosts at the same time the CommandView chooses the next available on its own LUN number.

  Thorsten