Backup to vCOps strategies

I would like to know what success / recommendations people have had with different strategies of backup for vCOps.

Problem: In a large environment your VM vCOps becomes very large (> 2 TB of virtual disk).  In addition, if you use the strategy 'add a virtual disk that you need', then you can added 3 or 4 virtual disks to your Analytics VM vCOps since its inception.  Then you end up with say 4 virtual drives 600 GB each.  Your backup software goes and takes a snapshot of your VM Analytics trying to back it up. Each virtual disk is then a Flash disk created for her.  Your backup software should then remove the snapshot when you are finished.

We have seen our (Avamar) backup software not handle this very well and the situation does not work due to the large size of the virtual machine, virtual disks and need to resolder the snapshots for these large virtual disks later.

I wonder what backup solutions work for you for vCOps in light of the following considerations / issues:

  • Preferably at the image level and no comments-backup is best to facilitate automation and ease of management
  • If virtual disks create more problems, must we create fewer and larger discs to first reduce complications.
  • Which scenario is more prone to errors, a snapshot of a huge vmdk and reconsolidation thereafter, or a snapshot of a large number of small discs of each partner drive delta needing to be reconsolidés in his record parent?
  • What level of surcharge should be reserved in the VM Analytics data store to facilitate a level backup image with its process of decision-making and consolidation of snapshots?
  • Since the VM analytics can be a data store all 2 TB to himself, at the level of the instant group could be on the LUN.  However, there is no utility in vCOps to ensure consistent table snapshots so that the vCOps database empty his info on the disc and the VM data are compatible with the snapshot at the level of the group.  (For example, a tool like NetApp SnapManager for Exchange)  What is the impact / problems lies in restoring vCOps by restoring the LUN of a snapshot on Bay?  Has anyone made this work?

Thanks for your comments

Image-level or VM level backups are preferred to the guest backups.

Personally, I use VMDK ~ 500GB to allow allowed easier maintenance as storage vMotions. Largest VMDK can be a bit of a bear to work if your storage media/protocol is not sized to handle the migration of VMDK 1 TB.

The preference of large VMDK to the smallest vmdk regarding snapshots is just preference... When you post snapshots on the drive they all have the same kick-off time anyway. I'd almost greater VMDK with less than total shots would be easier since less simultaneous tasks to perform on the host and you are still limited by piles of storage even in any case there may be a better answer out there if spend you sifting through the storage mgmt whitepapers.

Lack of air space for the VMDK, it would probably depend on what time of day you run the backup. It will be more read/write activity during DT and cron in the calendar of AM midnight-1-2, so I would say that target just after the opening hours for backups, so they are done ~ midnight. You can check you vC Ops disk activity to know when your busy periods, but they prob will be midnight-early AM. On overhead... Just leave enough for snapshots. If you can, at the SAN level backups are sometimes easier LUN containing the vC unit Ops so you don't have to worry about the free space data store. I can generalize and let say 10-15% free on my data stores, but it depends on the size of your data warehouse, if they are dedicated to the CR Ops, and that your data change rate is on the virtual machines on this data store (just don't run of disk space!)

If you make a snapshot at the level of the group, make sure that the vC Ops LUNS are in the same consistency group or all what he calls your particular storage provider. It's the same concern if you use, let's say, RS to replicate virtual machines to a different storage group. The vC Ops unit is rugged enough and you should not need to worry about comments to the level of rinse agents data (this isn't like SQL)... generally, it is accepted practice to make a snapshot of the device and treat it as a black box.

Tags: VMware

Similar Questions

  • Unable to connect to 2 sites: craigslist and google

    A few months ago, my upload has begun to slow down. Finally, 2 sites: www.craigslist.com and www.google.com would not at all. I've tried 3 browsers and none will have access to these 2 sites.

    Glad it works again and thanks for posting back.

    Unfortunately if the problem should recur is a problem of detention and they are the may be the most difficult kind of problem to understand.

    Now may be a good time to review your backup and recovery strategies.

  • BlackBerry Smartphones Reg: Deactivation of account business

    Hello

    Please help me to disable my business account. My business account is disabled in my company, but also my network provider. The problem is that my mobile content protection is turned on with the password. I can't able to disable. I thought that once my business account is disabled it will be automatically solved. But still I am facing the problem.

    I need the solution for the following

    1. how to disable business account?

    2. how to remove the protection of the content?

    Hello mc_selvendran

    Currently using this device purely for personal purposes?

    mc_selvendran wrote:

    Please help me to disable my business account. My business account is disabled in my company, but also my network provider.

    But... If your company account is already disabled then why you feel that it is actiavted yet? Is it because you are unable to delete the content protection?

    If so refer to this Knowledge Base, and to follow this suggestion, but first make a backup of your device:

    KB31291  :   How to reset a BlackBerry smartphone to factory using BlackBerry Desktop Software Settings

     

    But please note after a factory default reset while restoring your data from your backup to restore only your Contacts, calendar entries, memos, etc. Selective restore but don't restore all of these other backups these COMPUTER strategies which you deleted also restore again and you must follow the KB above again to remove COMPUTER policies.

  • RAID6

    We got an email Dell fom, raid 50 is is more convenient pest for tables with 1 TB SATA disks a NL - SAS.

    You converting RAID 6 arrays or let them run raid 50?

    Michael

    It is entirely up to you.   R50 is always supported by Dell.   With readers, especially SATA getting more broad, rebuild times are getting longer.   Thus, the risk of a double drive failure becomes greater.  RAID6 significantly mitigates this risk.

    What are you backup and DR strategies?   If the faulty table, what would happen to your business?  The revised policy is here to help you make these decisions.  Help you find the balance that suits your situation.

    EMC said the same thing about RAID6.

    Kind regards

  • standby and primary

    On my primary database
    When I archive log list, it gives me

    Database log mode Archive Mode
    Active AutoArchive
    Archive destination/u04/oradata/arch/Prod.
    More old sequence of journal online 29627
    Log sequence archive 29629
    Current log sequence 29629

    On my database of relief, when I archive journal list, it gives me

    SQL > archive logs list;
    Database log mode Archive Mode
    Active AutoArchive
    Archive destination/u04/oradata/arch/STNBY /.
    More old sequence of journal online 29627
    Log sequence to archive 0
    Current log sequence 29629

    But in the main hotback, it appears with these errors

    RMAN-08137: WARNING: log archiving not deleted because it is still necessary
    Archive newspaper filename=/u04/oradata/arch/PROD/PROD_0001_749397572_0000028449.arc thread = 1 sequence = 28449
    RMAN-08137: WARNING: log archiving not deleted because it is still necessary
    Archive newspaper filename=/u04/oradata/arch/PROD/PROD_0001_749397572_0000028450.arc thread = 1 sequence = 28450
    RMAN-08137: WARNING: log archiving not deleted because it is still necessary

    I don't know to sleep is not synchronized with the primary.
    Can someone explain in detail how to solve this problem.
    Any help is greatly appreciated.

    Thank you

    As I said in a previous post, I think you should be able to stop production by referring to the expectation, by disabling the indicated parameters of such Production.

    EG (do on prod)
    The "n" is only for those destinations that are the the eve of production - do not remove check-in local destination own submissions of prod.
    ALTER system set log_archive_dest_state_N = "DAVID";
    ALTER system set log_archive_dest_N = "; (two apostrophes)
    ALTER system set log_archive_dest_state_N = 'enable '.
    ALTER system set fal_client = ";
    ALTER system set fal_server = ";
    ALTER system set log_archive_config = ";
    ALTER system set remote_listener = ";

    alter system switch logfile;

    Confirm that only the remaining archive log destinations (local production) receive an archived redo log.

    Check the log of alerts for error messages.

    Perform an incremental backup of level zero on the primary, or however you make a FULL backup of the production.
    After that is completed successfully, you should be able to cross-check of Rman & remove obsolete. To delete all old backups and archived redo logs, which is no longer necessary.

    There is no need to change the controlfile to Prod at any time during the creation of a facility of DataGuard.
    The controlfile 'sleep' created on the host Prod during standby configuration steps - is NOT used on the production host, but transferred and used on the host of the previous day only.

    The installation manual, maintenance and monitoring of a Dataguard configuration is a very technical proposal.
    You must be very familiar with the machinery and used commands. RMAN backup and configuration strategies are also considerations.
    I maintain my own, 'playDB' primary and physical DataGuard configuration of Eve, linking a Test and Dev hosts, where I practice switch-overs and PSU patch again and again, without affecting the 'development', the test (QAT) or production programs.

    You use spfile - NOT pfile.
    If your production is currently using a pfile file, you will need to have a failure at some point to implement using spfile, before a successful DG configuration is possible. Use this same feature to define primary school "db_unique_name" to something a little supposed EG: name of host or first, ready for a future scenario DG.

    Good luck.

  • Opening backup strategies

    This discussion on another thread (link below), but I thought that it was preferable to a new discussion.

    https://discussions.Apple.com/message/30503051?start=420 & tstart = 0

    I've used several methods to save my Aperture library, but he was told that there may be problems with some of these methods. I summarize the comments made by others here:

    Here are the methods that should be OK:

    -Time Machine

    -Carbon Copy clone

    Here are the methods that may be problematic:

    -Vaults opening (since Aperture 3.3)

    -Possibly a backup online service (BackBlaze & specifically mentioned SugarSync)

    I am currently test-restoring a backup from BackBlaze and will report here results

    David

    I created a small (300 MB) Aperture library, then saved on BackBlaze. Then I restored the BackBlaze test library and downloaded on my computer. I tried to open the restored library and I got the message:

    "Opening has detected an inconsistency in your library. Please click on repair to avoid potential problems. »

    I clicked "Repair" and a few seconds later the open library. All the metadata and adjustments seem to be OK.

    I also sent a request for assistance to BackBlaze indicating that some opening users have had problems with the restoration of files opening and asked them to comment on. They usually take a day to answer. I'll post their comments here when I receive them.

    So far, their service seems to work with Aperture, but I'm skeptical rest since others have reported problems. Here is a link to their recommended procedures for the restoration of an Aperture with BackBlaze library.

    https://help.Backblaze.com/hc/en-US/Articles/217665858-iPhoto-Aperture-Restore-G medium

    Please add your comments about these or other opening strategies/backup problems.

    David

  • HP Pavilion 500 backup strategies

    This PC has hard partitioned, that contains Win 8.1 files and data on the C: partition and a "Recovery" Image on the D: partition.  It's how he's from HP.  I don't know exactly what is contained in the "recovery Image", I assume that it reflects only the system as it left the factory, i.e. no data, no agenda except that was delivered.

    (I use the historical functions of the files and the Windows Image from 8.1 to backup to an external drive).

    When the single drive the PC goes south, because it inevitably will, what should be the strategy?  What is possible with my setup?

    Obviously, the system image on the disk which went to the South is not useful for anything whatsoever.  So assuming I have install one new disk, what else do I need to create/buy/configure to restore my ability?  I don't think, for example, can I ask the bios to boot from the external drive, because I don't think that the windows "system image" is bootable.  Looking for suggestions.

    Hello

    With the help of a third party imaging tool would get the entire contents of the hard drive if you had chosen for the picture of everything. You will probably need two flash drives; one for the startup program that is created by the third-party imaging product and one containing the data.  It would be possible to use just a hold times start programme and data according to the amount of data are being imaged.

    Using the HP recovery media originally created you you you go back to the first day.  Then, you need to restore your data from backup and reinstall all the programs that you have installed.

    The best third-party imaging products are generally excellent. They are able to align advance format of the hard drives and make the boot and restore process.

    Separate the BONES of your data on separate drives is recommended because it reduces your risk by half.

    You want maybe to research; complete and incremental of differential backups backups.  Each has its advantages and disadvantages.

    I usually use another third-party product of data synchronization to synchronize data between an external media (USB key) and hard disks of data separated on my PC.  In the past, I had been sync'ing data on a drive external esata.

  • ASM backup with snapshots of SAN strategies

    Hello people,

    in one of our facilities, we have a database server (11.2) that has only the binaries and config (not the data files) and all the ASM disks residing on the San. The SAN has daily snapshots, so data files are saved. My question is, how should I proceed if the database server fails and needs to be reinstalled? My question is how can I tell the grid that we already have a diskgroup with disks on the SAN. I found md_backup and md_restore, but I don't know how and when do md_restore.
    I searched for a long time, but I'm quite confused with the approaches, please note that RMAN is not an option in this environment.


    Thanks in advance

    Alex

    Hello

    O.k., that changes things. You should have mentioned that in the beginning of your post, this stop the database you before you make the snapshot and you don't not to aid in archivelog mode.
    I would also suggest to disassemble start it before making snapshots.
    However, when was the last time you tried the archivelog mode? 11g has greatly increased performance.

    Note: in this case the snapshot is only a point of view. In case of error, you will lose data. (The last hours of your last backup).

    Now regarding your questions:

    -Online you will install a new system with the GI and the database software. You will need a new diskgroup (e.g. INFRA) to the new OCR and voting disks (they should not be included in other starts as DATA and FRA, so no don't need the old INFRA snapshot).
    => Then to present the DATABASE LUN and FRA diskgroup on the new server. ASM will be able to start it as soon as all the disks are available (and have the authorization of right). No need md_backup or restore. Backup MD_BACKUP the content of ASM headers, but since you have always drives this meta information is still intact.
    -Online record new data and services with the grid Infrastructure (srvctl add the database / srvctl add service etc.). Make sure you just point to the correct spfile (you can find for example with ASMCMD).
    => Database of departure...

    It's your snapshot was working. Only have another forum thread, where a diskgroup has been corrupted due to the snapshot (fortunately only FRA).

    And just as a reminder: a snapshot is not a backup. Depending on how the storage is the snapshot, you should take precautions to move to separate disks and check the TI (is it usable).

    Concerning
    Sebastian

  • for free ESXi backup strategies

    I know there is another recent thread on the same subject, but I don't want to thread Jack.

    I have an ESXi box that I want to put into the production department, but I have very little $ for licences paid, etc. I have a DLT Autoloader connected to the host and I want to know what the best approach would be to make regular backups of the virtual machine. Right now I have a Zimbra server and a pile of Web... databases get saved with the permission of a simple cron job. I'm looking for something simple and low-tech that will allow me to back up virtual machines on tape. Such a thing exists? I started watching GhettoVCB but which began the complicated research.

    I am not opposed to establishing another server to perform this task, but would rather not. I also wonder if I should go with VMWare Server instead!

    It is a project relatively small so it isn't a ton of $ to spend on VSphere Essentials, Veeam backup, etc. Thanks for reading... I enjoy any guidance someone is willing to offer.

    See you soon...

    I would add that I used ghettoVCB for awhile. Despite having and using other backup tools, I always trust the ghettoVCB script. It's a really defined and forget the solution. Spend some time to understand, and you will be rewarded.

  • Strategies of backup for virtual machines 200 +?

    Hello

    I have a question for you guys - if you have more than 100 virtual machines in your environment or your virtual machines are large & gt; 50GB, so what's your backup strategy for them? Are you using the VCB to backup all, or maybe you install the normal backup client (Veritas, CA, etc.) and backup via LAN? Or maybe you have other solutions for backup process?

    If you VCB, you have more than one VCB proxy? Or you start all on one server (a single script of vcb?)

    Thanks for any info.

    I should also mention that we use traditional backup for file servers agents over 200 Gb.  Ranger has a level file restore option, but it can take hours to reopen the file because it is compressed during the backup process.  I would also like to stress that the file formats shown in the screenshot are files zipped after the backup.  the size depends on how much space is used in the VMDK file.  some of the virtual machine listed are 50 to 70 GB vmdk files which are compressed within a 15 GB.  Also, when you are restoring is an exact replica of the original.  Some of my 50 GB VM restore tests were less than 30 minutes.

    If you have found this device or any other answer useful please consider the use of buttons useful or appropriate to award marks.

  • Microsoft have an automatic backup feature and ap?

    By friends, I was told that I should have my hard drive backed up some based on routing.  How can I do this?

    Hello Karen,

    Windows 7 doesn't have a backup application; its found in the Control Panel, backup and restore. It is, however, not a very flexible tool, it saves everything using an approach of 1 to 1 copy, usually requiring several DVDs.

    I use and recommend a 3rd party freeware application EaseUS Todo Backup, which is very flexible and allows you to save as well as clone entire hard drives is fast enough. You can download it here: http://download.cnet.com/EaseUS-Todo-Backup-Free/3000-2242_4-10964460.html?tag=mncol;1 .

    And Yes, you must back up your hard drive on a regular basis. the EaseUS tool has a pretty good help file explaining the different backup strategies. If your computer ever crashes (most people say 'when' it crashes), a backup will be you save hours of work and preserve all of your data from accidental deletion.

    Let me know if this helps you.

    Kind regards

    BearPup

  • Problems with the retention policy that removes a Cumulative incremental backup

    Hello

    I have following two scripts:

    1 RMAN script runs the weekend and made a full backup of level 0.

    2 RMAN script works on a daily basis and a level 1 Cumulative backup.

    at the end of each script I delete noprompt obsolete redundancy 1; (I don't need a single copy of local disk)

    My problem is that it seems that remove obsolete handles only level 0 (full) backups.

    for example, my goal is:

    Saturday = > level 0 + delete obsolete (level 0 + all the incremental)

    Sunday = > obsolete level 1 cumulative + delete

    Monday = > level 1 cumulative + Delete obsolete (backup Sunday)

    etc...

    What happens in reality is rather that remove obsolete Monday, completely ignores Sunday 1 cumulative level.

    and he keeps all the up to the next level 0 backup

    its actually covered in the Oracle Documentation:

    http://docs.Oracle.com/CD/E11882_01/backup.112/e10642/rcmconfb.htm#BRADV8400

    but it's just not make sense.

    in order to have redundancy of 1 copy I only need to level 0 + unique Level 1 cumulative backup.

    I don't have level 0 + all level 1 cumulative backups between full backups.


    so my question is there a way to make the retention policy to keep only the last level 1 cumulative?

    It's true window of 1 day recovery ignores several backups of cumulativ made the same day as being obsolete. Regular retention strategies rman probably cannot provide what you want. You can manually identify the backupset and delete them or change of cumulativ backups to incremental backups.

    That would save the space of several cumulative backups and and would work even for several backups on the same day. Of course with the disadvantage that you need all of the incremental backups and your recovery time may be longer

  • VCOPS 5.7

    After recovering from our analysis of server from running out of space VCOPs is backup place however the reports tab option is now missing...

    If I connect the web option vCOPS and go in Configuration, I have the available default policy in which there are a number of missing features such as planning and reports.

    Please can someone help advise how to retrieve the VCOPs to its former condition.

    Have you checked to make sure that the license is always get applied to the Ops CR? Looks like you lost some of your visibility Caron, who turns on the virtual machine to the user interface and is included in the Standard license and higher. If the license has been removed, it's why she came back to a State of edition of "Foundation".

  • VCOPS: No. puedo ver el Dashboard, activities, reports of analysis of y

    Buenas noches comunidad,

    Solicito su apoyo para what me DIB as puedo hacer!, the Le as the tool of vcops is en no puedo ver edge, Planning, analytical reports of y. Last week's realizó una reintslacion del vcenter derivative has the perdida del equipo, is reinstalo todo y is restauro el respaldo of the database, no is if esto haya tenido algun impacto y por eso no puedo ver todas las del vcops functions.

    Sin mas por el momento agradecere su ayuda.

    Saludos.

    Xaga

    This is because you are running in mode of "foundations".

    If he please ask the correct license, and this should be resolved.

    question - what is the reason for the errors = 100? (you can check on this page of the backup by clicking on the badge of defects)

  • Stopping Services in vCOps

    After the upgrade by-side of a large (> 1 TB) vCOps installation the backup process of the new TIME vCOps takes a long time (> 24 hours) for us.  For more security, should services be disabled in vCOps so that it does not attempt to make his calculations during the process of creation and the snapshot recall making the backup software?  Is there a KB for this?

    See the screenshot of the resource 'vCenter operations analytics.

Maybe you are looking for

  • Ultrabook Envy6-1100sl: install a 2nd OS on the msata SSD - Envy6-1100sl

    Hi, I upgraded my envy6 with 16 GB Crucial RAM and 500 GB samsung SSD evo. After upgrade, I did a clean install of windows 10 and stopped to use the itegrated msata ssd. My question is:Can I install a second operating system on the msata ssd? Can thi

  • Video limitation of 30 minutes Sony A7R II

    UPDATE: thanks to the application of the OpenMemories, the 30 minute limit is no longer a problem on Sony A7 product line. I bought a camera Sony A7R II. Now, I know that in Europe, the camera has a limitation of registration of 30 minutes. However,

  • Satellite L500-1kk and output HDMI for FULL HD LCD TV

    Hi all! I want to connect to my laptop L500-1kk to the TV LCD Full HD Sharp Aquos LC-42A77M via HDMI port, my question is that in the display properties that the maximum resolution is 1366 x 768, 1920 x 1080, so I'm afraid that if I buy a ([http://ww

  • Yoga 15 500 MICI - Dolby digital app does not work

    I improve my SDS HARD drive and reinstall Windows 10 and install it manually download Lenovo driver Realtek but Dolby digital plus app does not work? Model 500 15 DCI Yoga Comment to Moderator: The model is added to the object for clarity.

  • Back button with a focus for 5 d Mark III. How do I put in place?

    How can I configure button back focusing on my new 5 d Mark III.  Can not find the instructions manual supplied Canon or third book.