VDR - circumvention of Destination of NFS

Similar Questions

• How to make the vDR "redundant"?

  We use a vDR 1.2.1 (just what is the latest version) to save several important small VMs.

  vDR their back up to a network share with its own IP address.

  vDR worked several weeks now, so it is pretty fast now and a lot of incremental backups are accumulated etc..

  How can I make a vDR are redundant, as I have local vDR two backups that don't look alike as possible can be?

  We would like to have two redundant backups locally - ideally one mirror on the other - but any solution is going to do two local vDR backups that are as similar as possible.

  So we want to save one of the redundant copies locally to an off-site NAS - this remains to be tested.

  Overall, we went redundant backups locally vDR, either of which could be used to restore, which can be saved off-site somehow.

  Make the vDR return up to 2 different NFS data warehouses?
  I do rsync the vDR of backup on another NFS data store?

  Ideally, we want to be able to recover from the redundant backup locally vDR.

  Thank you, Tom

  With the help of two destination could be a simple solution...

  But also a rsync can work, if you remember that datarecovery service is stopped during this replication.

  André

• data recovery 1.2 only supports 2 destinations at once

  I started with vdr1.1 and had 3 installation of destinations (due to the limit of 256 GB), 2 were databases shared (from physical disk 1 500gig) and 1 was a network share.

  I installed the unit version 1.2 to version 1.1 of 'upgrade '.

  the mount on the part of network went well and automatically mounted 1 the datstores, but I can climb is no longer the third destination, a warning window

  presents itself. indicating the unit of data recovery is no longer supports more than 2 destinations both...

  Right now I can't do anything while I wait for the integrity check. (looks like it will take hours)

  so

  My questions are:

  can I move those jobs to other databases without creating new jobs?

  Can I set up another new device to use this database?

  Could I "extend" a data to another store

  Any other recommendations would be useful

  VMware has always said not more than 2 destinations, but they did not restrict it.

  now they do looks like.

  However, there is some good news here.  VMWARE now allows up to 10 devices VDR vCenter Server and you can switch between them from the vdr plugin.  so you can set up 2 separate devices vdr with 2 destinations, which allows you to spread your load of deduplication on several hosts.

  now I have 2 setting up devices vdr with 2 vmfs3 vmdk depupe stores and will perform tonight my initial backups.  wish me good luck.

  p.s. I didn't have much luck upgrading my 1.1 restore points, I thought that it's probably just better start fresh... you can keep the vmdk files if you have the room incase you have to perform a restore, just not config in the new vdr, or mount them.

  just a suggestion.

• Full VM data recovery or just snapshots?

  Hi people,

  I'm really in doubt on this subject.

  I started a job to backup with approximately 20 virtual machines, and after a while, when it finished I checked the ocupied backup size. Then I surprised because the size of the entire backup was only 64GB, and the storage space total occupied by all virtual machines is about 1 TB.

  
  So, how can I do a full backup vm?

  Example/Question: If I lost the entire data store where my vm running (one really disaster), I can restore my vm with data recovery (with backups stored in the other data store)?

  Kind regards

  sm4rtins.

  I don't know what help I can be, because my main experience with VDR is full of episodes of frustration. It could have been a reasonable product, but it seems to be deliberately crippled, and although version 2 is much more stable than the 1.x version (we meet only a few points of restore damaged about once a month or more, and not twice a week), it's always make us nervous.

  The only thing I would say is that we use fast SAN storage for destinations. VDR doesn't seem to be sensitive to the speed of disk I/O, and my assumption is that we get a restore point damaged when the disk i/o on the SAN is high (the SAN is shared with about 40 virtual machines and several physical servers that use for cluster storage databases). We do not have the data to back up this hypothesis (still trying to get a handle on how I'd get I/O stats to prove), but I don't see what it might be (other than the product being simply bugged, which might be the case).

  Could you move your store of destination from NFS to a fast SAN, just to see if the error rate?

• remote backups

  I am setting up data recovery, and I want to have a central backup server for all my data centers. Data recovery is set up, based on my options, I can only do backups/restorations with protocols the like SMB, iSCSI and vmdk disks. Is it possible to make backups remotely secure virtual machines with data recovery?

  Remote backups are possible but the bandwidth to the remote location will be the factor limiting. You might want to look at something else that vDR.Veeam is able to replicate to remote locations.

  With regard to the storage protocols do not forget NFS. vDR can use NFS as destination for NFS can be used for a data store.

• backup failed without displaying explicit message

  Hello

  I downloaded the latest version of ghettoVCB and tried to run it several times, but all virtual machines are not saved...

  I have no explicit message about why it didn't... when backup is not the snapshot still in the Snapshot Manager.

  I expect to have a problem with my NFS connection and it did not work when backing up...

  The problem appears if I run manually or with a crontab...

  http://pastebin.com/bjhX6yiV

  ESX 4.0.0 update 2 build 261974

  Destination storage: NFS on NAS with JumboFrame

  I hope you can help me with this problem

  Concerning

  Phil

  First try to save to a local disk if possible to be sure it is related to the NFS problem.

  Second, try this command by hand, (last line of your journal):

  / usr/sbin/vmkfstools-i "/ vmfs/volumes/vmhtoip2_bh06_data3/akspronto-bh06/akspronto-bh06_2.vmdk"-a "free" 'thin' d "/ vmfs/volumes/vmware_backup/akspronto-bh06_2.vmdk"

  Change the target in "BOLD" to have enough space and declare the output and exit code.

  Get the exit code by running

  ECHO $?

  just after the command. Everything is important.

• VDR - addition of network NFS share

  Hello

  I've set up an NFS share on a Dell NAS running Windows Storage Server 2008 and Active Directory Lightweight Directory Services (AD LDS) to provide a UNIX user name mapping to Windows.

  I can add that the NFS share as a data store for my ESXi host with no problems; but for some reason, I can't not add that the same NFS share a device VDR running on an ESXi host, even if this host is the one running the VDR!

  When you add a NFS datastore to a host, the dialog box wonder following:

  • Server: '111.222.333.444.
  • File: 'MyNFSShare. '
  • The data store name: "MyNFSDatastore."

  When you add a destination to the VDR, dialog box asks me what follows:

  • URL: '\\111.222.333.444\MyNFSShare '.
  • Username: "< sharing NFS Username >".
  • Password: '< shared NFS Password >.

  I also tried to enter the vCenter administrator user credentials but it makes no difference, and the newspaper of the events on the NAS server reports a successful security event "Audit success" for the connection of the user anyway.

  Anyone can shed some light on this?

  VDR accepts no NFS as destinations, but it does not accept CIFS.  So, instead of trying to add a NFS target, add a windows share real using windows (or domain) credentials.

  I too seem to get best performance of an instance of NFS Linux than Windows, I think because the implementation of Windows isn't technical compliance.  If I watch the console while doing a vmdk format hosted on a Windows based NFS target the connection seems to constantly go in and out.  I now running on a Windows 2008 R2 server and it works well enough for my purposes, that hosting the vmdk VDR so I can their riffle on tape.

• Can a vDR device cause multiple backup destinations?

  I tried adding the shares network, adding that a different network share seems to remove the network share previously implemented at which the current backup operation is set.

  I want two shares of different network for a device with two jobs vDR, each backup operation goes to a different network share.

  How can I do this? 1.2.1 using latest version vDR.

  Alternatively, if each camera vDR can talk only to a single network share data store, I can set up another camera vDR.

  Thank you, Tom

  CIFS is not a recommended destination and can be much slower. If you want an accessible destination I would add a couple of NFS shares as network data warehouses. Then add a normal VMDK unit vDR. You can do several (more than 2) virtual machine backups but as I've mentioned before that you will incur the trouble to write the cbt file for each additional backup done.

  I'd do more simple a test you can against 2 or 3 small VMs. run multiple backups. Check the speed of the restoration. Decide if you can live with the speed of the restoration. I'm not sure that this is the most appropriate use of the vDR unit.

• Two Destinations - a VDR

  I met an interesting problem...

  I have two network shares added to my camera VDR, it is NFS, the other is CIFS (Windows share). They use different credentials of course.

  I created two backup tasks, using the NFS share, the other uses the CIFS share.

  The first backup job, using NFS Sharing worked well for centuries. The second task of backup and the corresponding CIFS network to share, I added at a later date. When I added the second network (CIFS) share, the second backup task went very well but the first backup task, then failed to run. When I returned in the first backup job, it seemed that the network share was more accessible - the 'Destination' option is grey and when I tried to click on it, I received the following error message:

  "This destination is not available for a backup job. Please make sure that the destination has been formatted and has enough free space.

  I disassembled the part and mounted successfully. This was going well apart from checking the integrity resulting takes too long (over 12 hours).

  Verification of backups, the next day, it turned out that the second backup task now was not to run and of course, when I watched the 'Destination' in the settings of the backup task, it is gray as well!

  So it seems that if I add a network share with different credentials for a backup job, network, share different, with different credentials, will be made invalid which will then affect backup specifying as a destination.

  Is there a 'best practice' for multiple destinations on one device VDR? Is it because they are of different types of share or the device does support only two destinations if they use the same credentials?

  Please note that both of my destinations to target the same NAS (same IP address) - every action is on a separate virtual disk of 500GB.

  Note the two destinations by its IP address to the value 'Server' but I use a different value to the value of 'folder '.

  This is the value of the server that must be unique for different credentials. You can try to plug the CIFS share using the host name (assuming that name resolution works)

  I wonder if the difficulty would be to publish my second network share using NFS which will then use the same credentials.

  It should work.

  André

• VDR 2.0.1 / dev/sdb /-physical disk not rise, destinations

  Hi all

  I have a problem with my VDR 2.0.1 last backup has been the failure due to power and after that problems on VDR destinations where I had destination normally one network, stay another one, physical disk no rise/dev/sdb /. I can not remove from vClient cause it says "destination is in use", but there is not any visible running task. This physical disk is exactly the same size as one of my VM who began with backup when possible failure has occurred. Also, under the VDR icon in vClient I see the same VM hard file.

  That's happened? How can I mage it?

  Concerning

  Kris

  Remove the disc from VDR-VM-mixer under 'Edit Settings' with Remove.

  That make a snapshot of the virtual machine affected and after that delete all the snapshots of this virtual machine.

  Normally this will solve these problems, if VDR was interrupted during backups.

  Paul

• How munt NFS on machine VDR to take backup

  Hi I have the Vsphere 4.1 running on my production facility. Now, I plan to have the backup of Vmware solution. VMware VDR ti I plan to implement. I have 1 GNU / linux redhat spare 4.7 is also which is 1 TB of storage... I went through the document and find that we can mount NFS under

  VDR. Please tell me how can I do this successfully if I use 1 TB of space NFS linus for vmware backup vdr.

  http://planetvm.net/blog/?p=719 must be a help.

• VDR 1.2 write data on FreeNAS via NFS crashes... Suggestions?

  Hello, ALL!

  We have a 1.2 VDR-appliance configured fresh (2 vCPU/4Gig RAM) to use a Datastore of ESX targeting hosted FreeNAS-NFS-Sharing.

  The problem is that the data is written in this VERY SLOW NFS share (i.e. 1-3 Mbps when formatting of the target by VDR).

  Finally, she succeeds formatting. (200 GB targetsize results in 423 MB space used when empty)

  If now a backup task is run on this target, taking snapshots is successful, but then that falls of transfer of the data copy in after about 10 seconds, leaving the VDR to an of a permanent 50% CPU usage (looking at the ESX - CPU 32% constantly instead of generally by 10%).

  Also, there is no way to connect via VDR-plugin.

  A glance at the Vclient console reports the following:! VDR-jerks-20100820 - 1.gif!

  So all I can do is barely powering of the device.

  What is a performance problem or a FreeNAS misconfigurated? (1 vCPU / 256 M RAM, 2 1 NFS share (4 servers), CIFS shares)

  Who is using FreeNAS? Any suggestions?

  Do not hit FreeNas, we use it for other articles, but it gave us fits with VDR, making the switch to Openfiler (you can thank us later), it has NFS or use its iSCSI mode (if NFS mode... use ASYNC).

  http://communities.VMware.com/thread/253718

• VDR: Size of backup destination

  Hi all

  It's maybe a stupid question, but I have to ask. I have test VDR 1.2 and have configured a share as a destination and currently save a few machines with a total of about 450 GB allocated disk. Most of the machines is very similar, c-disk of 20 GB and 2008 R2, but the virtual machine also has a d-300 GB disk that is about half full (with vhd files). If I understand correctly, the first backup is full, then the rest is incremental. I expect the backup destination to about the same size as the actual data used, at least 150-170 GB since that's the size of the unique data for the virtual machine, I backup with VDR, but the size of the backup destination is only 29 GB. I totally hard works of VDR? No explanation will be greatly appreciated.

  Thanks in advance.

  GB

  Hello

  have a different opinion. Data deduplication on all incoming data recovery. Get rid of full/incremental view.

  When you first save a virtual machine all the data of this virtual machine will be read and recorded deduplicated. Second time datarecovery uses the followed the change block. Your vm writes a file (... ctk.vmdk), where he follows the changed blocks. Data recovery simply saves only the blocks changed since the last backup. You can call it incremental if you want, but from the point of view of the transfer data, these are all full backups. There is no dependency between backups.

  If save you a virtual machine with 150 GB and your destination of data recovery shows that 29 GB used - it is possible. It depends on what kind of data your saved vm a. imagine an Oracle database that uses several files that have a size of 10 GB left. If only 1 GB inside this file is used 9 are empty. Who will very well be deduped...

  Maybe inside your virtual machine vhd files are for the most part empty? Or has information that can greatly be deduped?

• VDR - difficult to write to the destination volume, error - 1020 (sharing violation)

  With the help of VMware Data Recovery 1.1, backup on CIFS and get this error

  18/01/2010 08:55:31: progressive run to the top of the disc "[Lun] VM/VM-DB1 - flat hard" using SCSI Hot-Add.

  18/01/2010 08:55:32: trouble of writing on the destination volume, error - 1020 (sharing violation)

  Integrity of the destination checks complete successfully and I tried to reboot the device just in case VDR.

  To resolve the issue I removed the part by the VDR, has made backups to other destinations and renamed the VMware Data Recovery subfolder in the destination, then add the share and makes backups, this of course creates a new backup store.

  Anyone have any ideas why this error occurs, I can't save in this backup more than store means.

  I'm having the same problem. After I spent time on google, I find this is a problem of Linux (VDR is Linux-based)

  I have following the site and do it step by step. Now, the VDR works very well.

  URL: http://www.cyberciti.biz/faq/linux-tcp-tuning/

• VDR 1.0.2 - error message during the destination error-2241 re-indexing invalid/damaged index

  Hi, we get this error message during recataloguer. -2241 destination index invalid/damaged

  Is it possible to recover from and if so - what should I do?

  Hello

  the solution for me was to go in the VDR server, log in as root and delete the lock file.  Then I was able to start the integrity check

  Repeat this step for all data warehouses

  -# cd /SCSI - 0\:1 /.

  # ls

  Lost + found

  VMwareDataRecovery

  # cd

  VMwareDataRecovery /.

  # ls

  BackupStore BackupStore.cat BackupStore.cat.bak

  # cd

  BackupStore

  # ls

  store.lck index - 05.dat 000 bmap-index index - 03.dat - 00.dat

  999

  index of BMAP-index - index - 01.dat 04.dat - 06.dat

  bmap-index - 00.dat

  index of BMAP-index - index - 02.dat 05.dat - 07.dat

  bmap-index - 01.dat

  BMAP-index - 06.dat index - 03.dat journal.dat

  bmap-index - 02.dat

  BMAP-index - 07.dat index - 04.dat map.dat

  [root@VS71

  BackupStore] #.

  # rm - rf store.lck.

  Post edited by: Abid