Top 20 Moments in the History of Backup and Disaster Recovery

History

13.7 billion years BC – The universe begins as a singularity; those who believe in the “big bang” theory suggest the disaster is on-going…

3.8 billion years BC – The start of life on Earth. The first cell is thought to have arisen from self-replicating RNA what developed later into DNA. DNA is a store of biological data, the genetic information that allows all modern living things to function, grow and reproduce. Put another way, you are the backup of your parents. Say hi to the therapist for me.

65 million years BC – Dinosaurs, not backed up.

Dino

13.7 billion years BC – The universe begins as a singularity; those who believe in the “big bang” theory suggest the disaster is on-going…

3.8 billion years BC – The start of life on Earth. The first cell is thought to have arisen from self-replicating RNA what developed later into DNA. DNA is a store of biological data, the genetic information that allows all modern living things to function, grow and reproduce. Put another way, you are the backup of your parents. Say hi to the therapist for me.

65 million years BC – Dinosaurs, not backed up.

Cyrus

 

48 BC – The burning of the Library of Alexandria. Among others in your “Top 10 Lost Books of All Time,” the second book of Aristotle’s Poetics went up in smoke and humanity was beginning to realize the fatal flaw in their cunning backup plan; paper is actually quite flammable.

1347 AD – The first known insurance contract is signed in Genoa, Italy. This was great for those buying and selling goods and owning property but information is difficult to value, most people would rather have their data back than receive compensation for its loss.

1436 AD – Johannes Gutenberg, a former goldsmith, created the first printing press in Germany. He used his knowledge of metalwork to fashion letters out of an alloy, pressing these against ink and then paper to create a copy. This made the printing of multiple copies considerably faster, a great step forward in data resilience.

1539 AD – Image based backup, born. Henry VIII, King of England was trying to decide who to marry next, he sent the artist Hans Holbein to make a reliable copy of what his list of European princesses looked like. Based on these images, Henry made his choice and proposed engagement to Anne of Cleeves only to discover she looked nothing like he expected. Corrupt data/bad copy.

Anne

48 BC – The burning of the Library of Alexandria. Among others in your “Top 10 Lost Books of All Time,” the second book of Aristotle’s Poetics went up in smoke and humanity was beginning to realize the fatal flaw in their cunning backup plan; paper is actually quite flammable.

1347 AD – The first known insurance contract is signed in Genoa, Italy. This was great for those buying and selling goods and owning property but information is difficult to value, most people would rather have their data back than receive compensation for its loss.

1436 AD – Johannes Gutenberg, a former goldsmith, created the first printing press in Germany. He used his knowledge of metalwork to fashion letters out of an alloy, pressing these against ink and then paper to create a copy. This made the printing of multiple copies considerably faster, a great step forward in data resilience.

1539 AD – Image based backup, born. Henry VIII, King of England was trying to decide who to marry next, he sent the artist Hans Holbein to make a reliable copy of what his list of European princesses looked like. Based on these images, Henry made his choice and proposed engagement to Anne of Cleeves only to discover she looked nothing like he expected. Corrupt data/bad copy.

Manchester

1964 AD – Mass market computing begins, the Programma 101 was unveiled to the public at the New York World’s fair. One of these computers was used on Apollo 11 and it was pretty much… a calculator. “One small step…” (at a time!)

1972 AD – Mainframe computers deliver applications and data at high speed to hundreds of users, in-built hardware redundancy ensures exceptional RPOs and RTOs. The ancient Sumerians would have just loved this.

1990 AD – Arcserve 1.0 released by Cheyenne software. The age of distributed computing is in full swing and it is all about backing up to these little rectangular things called “tapes.”

1998 AD – VMware founded in Palo Alto, California. Although the concept of a hypervisor originated from 1960s, it was VMware who introduced hardware virtualization to the mass market. Virtualization will go on to revolutionize backup and disaster recovery.

vmware

2006 AD – XOsoft’s WANsync technology is integrated into Arcserve. For the first time mid-market users can perform both backup and full system failover from one solution.

2008 AD – Microsoft releases their competing product to VMware, they call it Hyper V. If you weren’t virtualized before, you are now. Specific software for virtual backup exists but there is little integration with physical servers, tape backups or cross platform Microsoft/Linux.

Hyper V

2006 AD – XOsoft’s WANsync technology is integrated into Arcserve. For the first time mid-market users can perform both backup and full system failover from one solution.

2008 AD – Microsoft releases their competing product to VMware, they call it Hyper V. If you weren’t virtualized before, you are now. Specific software for virtual backup exists but there is little integration with physical servers, tape backups or cross platform Microsoft/Linux.

arcserve award

2016 AD – You are here.

Please register to see a live demo of Arcserve UDP here.

Or download a free copy of Arcserve UDP here.

Would you like to discuss how to get the best pricing for Arcserve or do you have any specific questions about the technology?

Drop me a mail with your contact details and I can help: louis.cadier@arcserve.com

How to Take Hypervisor Snapshot of Lotus Domino VM with Unix/Linux Guest

I previously published a post Online Backup of Lotus Domino with Arcserve UDP, which used custom scripts so that DB’s consistency was guaranteed during the Arcserve snapshot of a virtual Windows Lotus Domino server.

This was, however, only compatible with Domino on Windows guest. I have since collaborated with Daniel Nashed from Nash!Com in Germany to come up with a creative solution to run an Arcserve Hypervisor snapshot of Domino Virtual Machine running on a Unix/Linux Guest.

Interested? Read on…

Daniel Nashed developed a script for Unix/Linux that would stop domino server. Utilising this script will allow Arcserve to take a DB consistent snapshot through the preferred Hypervisor. The script Bundle is available here. (Please note: the use of blogged scripts are at one’s own risk and should be tested with sandbox or lab copy of your production VMs). There is certain risk in shutting down Domino Server services at every backup, however, a shutdown is the only real solution for ensuring consistency of all databases during backup Snapshot pass.

Using virtualized Lotus Domino as a corporate messaging system on a Unix/Linux guest, the database’s consistency is guaranteed during backup by running custom script Rc_domino_script.

Once you have downloaded the script bundle from Nash!Com the bundle should look like this:

bundle

 

 

To start, the VM guest requires the relevant Hypervisor tools to be installed e.g. VMware tools or Host Integration tools. This will allow Arcserve to pass commands through the Hypervisor to the VM guest and initiate the script pre-snapshot and post-snapshot.

Next, the above files need to be copied to their relevant locations:

Rc_domino_script is the main script logic. It needs to be copied to the Unix/Linux guest location: /opt/ibm/domio

Rc_domino is the main entry point file for the service. It needs to be copied to the Unix/Linux guest location: /etc/init.d

Rc_domino_config_notes is the configuration file used. It needs to be copied to the Unix/Linux guest location: /etc/sysconfig

These three files all reference one another and are required for pre/post-snapshot.
If you are required to make changes, such as a different username for Domino, you will need to make changes to the config. file and modify the settings in the Rc_domino script. For more detail, refer to Read me or NashCom.

Once the script has been copied, we can now create an agentless plan in Arcserve UDP under plan > setting. In the Advanced tab, add the following commands to reference scripts:

/etc/init.d/rd_domino stop

snapshot

/etc/init.d/rd_domino start

snapshot

With the above, we are able to successfully snapshot Domino DB on Unix/Linux without an agent and provide application consistency.

Many thanks to Daniel for his Domino expertise!

 

Veeam V9 Vs Arcserve UDP V6 – A Practical Comparison

Comparison

2016 is off to an exciting start with Veeam and Arcserve releasing their latest versions. Veeam releaded the V9 Availability Suite and Arcserve released UDP6 (codenamed “Tungsten”).

Side note: Veeam coined the Phrase “RTPO” which is essentially means “RTO and RPO”. Personally, I disagree with this term as I think that the two terms are completely independent from one another: “Restore Time” and “Restore Point”. This is just my opinion.

Both are excellent products. I have had some time to test both capabilities and have done some research on their features. Below, I have a comparison based on the new features of Veeam V9 vs Arcserve UDP V6 and how each one delivers its’ functionality.

Instant VM Restore

An instant VM restore, or IVM, is about mounting your latest backup point to your Hypervisor host as a temporary data store point and registering that VM into the virtual infrastructure for immediate accessibility. It is no longer necessary to wait for your restore to copy points from backup disk into production – with IVM your RTO is seconds to minutes.

VeeamVeeam has had this functionality in previous versions, but has added some development: vPower Cache. This feature allows for recently accessed backup files to be cached and this will assist in speeding up instant VM restores, unlike Arcserve’s IVM.

 

arcserveUDP6 has just developed this feature in the new release V6. UDP 6 does have vPower functionality and can support instant VM restore cross-hypervisor, can instantly restore a VMware VM onto Hyper-V. In addition, UDP 6 also protects physical machines and allows for instant VM recovery of a physical node “P2V IVM”. This is very powerful stuff, unlike Veeam’s IVM.

 

Replica VM – Virtual Standby (For Disaster Recovery)

Replica VM or Virtual Standby “VSB” is a pre-exported conversion of your latest backup point into a virtual machine ready to power up in the event that production VM is lost. This is not a new feature for Veeam or Arcserve, but I would like to compare the two in any case.

VeeamVeeam has replica VM functionality , which is easy to use. You can create a replica VM off of the latest backup point or live snapshot process on your production VM. Both Veeam and Arcserve have networking and configuration functionality on these tasks. I would like to mention Veeam’s DR failover plan here. Where the plan is created with a boot sequence of multiple VMs, this failover plan allows you to commit to replica VMs or fail back to production VM.

arcserveUDP6 has what is called “Virtual Standby” that very easy to use. You can have a backup plan create virtual instances of the latest backup point. UDP6 VSB supports virtual standby cross-hypervisors and “P2V” (physical to virtual). This would mean you can have a VSB of your Vmware VM on hyper V or vica versa. This also means that you can have a VSB of a physical node onto your hypervisor. In addition, once failed over to VSB from the physical node you are able to restore back to the physical node directly off of the virtual standby to incorporate the latest changes written to the VSB.

 

Instant Bare Metal Recovery “BMR”for Linux

For Linux physical machines, Arcserve UDP introduces Instant Bare Metal Restore (BMR) to support local and remote bare metal recovery of Linux physical machines. Instant BMR provides better restore capability and an improved end user experience by providing users with instant access to a target machine prior to the entire recovery process being finished. This is done without physical interaction with the Linux nodes across the network though PXE boot solution.

VeeamVeeam does not currently support this and has limited support for physical server environments, as it was originally designed for Virtual solutions.

 

arcserveUDP6 provides for exactly what has been described above. Arcserve can instantly regain access into a failed physical Linux node with instant BMR. This is a first of its’ kind and a very powerful feature.

 

Granular Restore Tools “GRT”

GRT are interfaces that the backup vendor develops to facilitate for granular level restore back into applications. Example: an exchange GRT, allowing single mail items to be restored from a DB backup back into a mailbox into the live exchange DB. Most backup products support file level GRT as this is supported by Windows Explorer.

VeeamVeeam refers to a GRT as a “Veeam Explorer” and has in their new release V9 an Oracle Explorer and Veeam has granular VM recovery in Veeam Explorers for Microsoft Active Directory, Exchange, SharePoint and SQL Server.

 

arcserveUDP6 refers to this as “GRT” and has integrated this into the console restore capabilities, rather than having to install a separate feature application. Arcserve must be commended on its exchange GRT as this is very granular, allowing restores of mail down to calendar and tasks back into a mailbox. However, Arcserve currently supports Microsoft Active Directory, Exchange and Linux file/folder on GRT. Granular restores are still possible for Oracle, SharePoint and SQL but rather through its’ own interface.

Corruption Guard or Recovery Point Check

This feature tests the data integrity of the files systems in the backup points to ensure no corruption on backup points and no loss to data or unusable restore points.

VeeamThis is a new feature for Veeam called “Corruption Guard”. This runs a process similar to check disk CHKDSK that tests and repairs file system data integrity issues on your backup restore points.

 

arcserveThis is not a new feature to Arcserve. The feature is called “RPC” or Recovery Point Check. Arcserve will mount the restore point and actively run a CHKDSK on the files system and repair any corruption.

 

 

Backup Data Reduction

This is somewhat a large topic as it covers a variety of features. Data reduction on backup mainly consists of compression and global deduplication in most cases. Deduplication is the comparison to data across a deduplication domain only holding one instance of unique data, thus cancelling out redundant data being held on backup. Compression will compact files during backup to shrink the backup size to as small as possible.

VeeamVeeam has added a few features and further development on their de-duplication feature.The largest setback on Veeam’s global deduplication is that it’s not so global. The deduplication domain is across a single job – only servers within a plan will be de-duped against one another. There has been added development where multiple VMs within a job can be de-duplicated in parallel across both live backup streams. In addition, they have focused on third party storage vendors’ deduplication (such as HP and Netapp) to assist in reducing their backup footprint to storage.

Another feature, defragmentation and compact, will assist on further reducing the backup foot print over longer retention periods. This will access backup points and remove deleted data or VMs without requiring you to create a new active full.

Scale-out backup repositories (SoBR) and bitlooker are further additions to the V9 release.

Bitlooker allows you to exclude files and folders from your backup (e.g. exclude c:\temp). Further to that, it will allow you to exclude blocks that are marked as deleted. When a file is deleted in Windows, the space isn’t actually wiped clean – Windows just removes that file from the master file table, effectively forgetting about it and allowing future data to occupy the space. Unfortunately, because the data is still occupying space, it is getting backed up. BitLooker recognizes this fact and skips over these files.

The scale-out backup repository allows you to create a backup storage pool using multiple physical storage appliances, thus creating a federated storage repository. This increases write performance, as backups are written in multiple streams to multiple devices and this will also reduce storage cost (because you can repurpose storage devices).

arcserveUDP6 has true source side global de-duplication. The deduplication domain is at the backup server (RPS) storage level. Similar to Veeam, storage is presented through Windows or IscsI to Arcserve, but all backups to Arcserve storage repository are de-duplicated against each other, allowing for multiple plans/jobs/policies. This runs at 32k, 16k, 8k or 4k block size levels with a forever incremental strategy utilising CBT change block tracking. This means that only changed data blocks from the source are included into the backup pass. Furthermore, these blocks are compared to the backup storage repository to confirm they are indeed unique, thus massively reducing the storage requirement for backup on arcserve UDP. Both physical and virtual will be included in the same de-duplication domain. UDP6 has included physical Linux nodes into this single de-duplication domain. Impressive lab results have shown that protecting half a Petabyte of storage has left a storage footprint of 13.5TB on backup disk. Whilst the storage features Veeam V9 has developed are beneficial, these seem to be in place to assist their struggling de-duplication protocol or algorithm. Arcserve seems to have hit the nail on the head when it comes to data reduction, leaving backup simple yet efficient.

 

Stand Alone Console

The console would be the management interface for the backup application where reports and logs, jobs, schedules etc. can be configured.

VeeamVeeam has released a standalone console in V9 that allows you to install an application on your workstation that will connect to your Veeam backup servers on the network instead of multiple RDP sessions to your backup servers. The Veeam console is an installed application and so is the standalone server. Performance here is impacted by server or workstation CPU and memory utilization.

arcserveThis has always been a feature for Arcserve (UDP stands for “Unified Data Protection”). Arcserve offers a unified console that is web based and backed by Tomcat. This has a tiny footprint when it comes to compute overhead and is browseable via Http or Https on any workstation or mobile device on your network.

ROBO Support

ROBO (Remote Office, Branch Office) support allows for your main backup infrastructure to communicate and maintain processes or jobs on your remote sites. This should allow for features such as remote backup/restore capabilities and reporting on multiple sites from a central location.

VeeamVeeam previously had issues relaying commands from Veeam B&R console to backup proxies across the WAN sites or VPN tunnels. In V9, Veeam has released Guest Interaction Proxy which allows for a secure SSL connection between sites and proxies back to B&R console. This allows for remote restore across WAN sites and mounting backup points locally. This was a much needed feature for Veeam service providers.

arcserveUDP had a similar issue when it came to ROBO solutions where this had to be done through VPN connection. Even so, this worked and was functional to each remote site. In UDP6, the Remote Management Gateway feature allows for secure SSL connections across WAN links to ROBO sites. It allows for all management out of a single console and the ability to configure and push agents from one console for all sites. This compliments the existing unified console.

 

Cloud Connect (Cloud Backup & Replication)

Cloud backup and replication allows for the backup product console to connect to a service provider service such as storage or compute resource. This will allow the customer to replicate VM copies or backup points offsite into the cloud and pose as a DR (Disaster Recovery) solution.

VeeamVeeam has added development to their cloud connect feature. Previously, this only allowed for copying backup points to a cloud target, but Veeam V9 brings the replication feature too.

Previously, I discussed Veeam’s replica VMs. A customer with an onsite Veeam installation can enter details of their Veeam service provider’s Veeam B&R Gateway and replicate replica VMs to their service provider. A limitation is that if you are replicating from Hyper V, your service provider must provide Hyper V.

 

arcserveThis is not a new feature for Arcserve. Arcserve does this differently, though. The service provider creates a share plan with credentials and a secondary task to export as a virtual standby on either Hyper V or VMware. These credentials are shared with the customer. The customer will add a task to their backup plan to replicate to a remotely managed RPS server and use the credentials provided. The target hypervisor is cross compatible and the source and target hypervisor can differ in this instance. Arcserve supports this on physical servers too at the source, which allows for P2V to the service provider (“Physical to Virtual”).

Hardware Integrated Snapshots

Hardware snapshots allow the backup software to access the hardware array and initiate a snapshot of a VM of the hardware array, utilising the array compute resource to carry out the backup process. This results in a more efficient snapshot.

VeeamVeeam has done a lot of development with various storage vendors and has support on multiple storage arrays, such as NetApp and HP and the new edition in V9 of EMC. Veeam has also released Sandbox for HW snapshots in V9. This creates a clone of a snapshot into an isolated environment for testing purposes directly off the storage appliance, cutting out the unnecessary overhead.

arcserveUDP6 includes a hardware level snapshot integration in the new release for Netapp .

 

Tape

VeeamVeeam has released advanced support for Tape in the V9 release. This includes parallel rocessing, concurrent copy sessions and GFS rotation schemes (“Grand Father, Father, Son “). Tape is still underdeveloped for Veeam .

 

arcserveUDP6 has integration into its father product, Arcserve backup, which has been around for over 20 years. With over 20 years of development around tape features and support, it is far superior to Veeam’s Tape functionality. These features include: Multiplexing (2-32 jobs), Multistreaming, Device Group and Media Pool, GFS Rotation and Synthetic Backup, Append Media, Media Maximization, Media Pool Manager, Tape Library Option and Auto Library Detection and Configuration, Bar Code Support, Auto Inventory, Auto Eject Medi, Monitor Blank Media Qty., Tape Management and Tape Vaulting, Auto Tape Cleaning and Configurable Block Size for Tape.

Both products have their features that stand out. In my opinion, a lot of the new features Veeam has added to their suite were pre-existing with Arcserve since 2014 and before the release of UDP. Veeam seems to be playing catch-up. Veeam also doesn’t have a developed protection solution for your physical server workloads. Arcserve has physical and virtual protection features with  integration into your environment which allows for physical to virtual,  virtual to physical, virtual to virtual, “ Vmware to Hyper V “ restore/conversion functionality.

Arcserve also holds a replication and high availability suite that has been integrated into UDP, which allows for more than just disaster recovery and actually provides what Veeam claims to be: always on, automated fail over and instant replication of your business critical services.

 

 

Arcserve UDP Windows Remote BMR with WDS

 

With the new release UDP6, so comes the functionality of instant Linux BMR (Bare Metal Restore) which allows you to recover physical hardware remotely and instantly. This feature would be also be great for Windows environments not yet available.

A great solution for the remote recovery of physical Windows servers is to use Windows Deployment Service (WDS) integrating Arcserve UDP 6 Restore Capabilities with WDS, allowing for remote physical restore. It is no longer a requirement to have an engineer standing in front of your data centre rack to run system state recoveries on your physical system.

In this post I explain how I have created such an implementation and tested it!

The prerequisites are a Windows server and DHCP server (I used Virtual Servers in my testing but it applies to physical servers too).

The process would be to access your physical server through remote BIOS (e.g. ILO, IPMI or IDRAC or similar) and setting the server to network boot, at which point the server will PXE boot the Arcserve Windows BMR ISO files.

I used one server for the implementation: Server 2012 R2, running Arcserve UDP Console and RPS roles. I added Windows roles WDS and DHCP.

Installation guidance for WDS and DHCP:

How to install and configure Windows WDS

Installing and configuring DHCP

 

This one server had an Arcserve agent, so I created a BMR ISO for X86 and X64 compatible with ADK 8.1. You can create both Windows 8 and Windows 7 compatible boot kit ISOs for Server 2008 and 2012 physical server spreads in your environment.

*One important thing to note: if you run WDS and DHCP on the same server then some properties need to be altered on WDS as they both listen on the same port.

Port

Once your BMR ISOs are created, browse to their location and mount ISOs.

Then open WDS MMC through Server Manager: under “Boot Images” browse, select add boot image and follow the wizard.

How to add boot images

 

Unfortunately WDS can’t use ISO format boot images and requires .WIM.

Browse Image location to:

ISO:\AMD64\SOURCES\BOOT.WIM for X64

ISO:\X86\SOURCES\BOOT.WIM for X86

Name your images as this will be displayed at your boot screen.

Add Image

Once the image has been created and stored, you can begin a network boot. (Log into server remote BIOS interface, initiate boot from network device.)

DHCP will assign an IP address and discover PXE proxy “WDS Server”.

Press F12 to Boot into PXE.

PXE

You can now see available images to boot from.

Boot

arcserve

After boot select you will then see the Arcserve Bare Metal Recovery screen.

This is great for large workstation environments and multiple remote sites.