Backup and Governance
Showing results for 
Search instead for 
Do you mean 

Data Protection: How long can you live without your backup server?

StephenSpellicy ‎03-12-2014 07:33 AM - edited ‎02-19-2015 12:38 PM

I often hear from customers and partners that it’s all about the restore—trust me I get it. Restore is where it’s at: It’s all about achieving a successful recovery in the shortest amount of time.  When we talk about restores, we think of recovery point objectives (RPOs) and recovery time objectives (RTOs). But without backup, a restore is useless. Moreover, it would be non-existent.

 


 

Backup vs. Restore?

Trust me; I am not here to wage the war of “it’s about backup and not the restore”. In fact, I would argue that the two are made for each other like peanut butter and jelly or lemon and lime. What I want to talk about is how important it is to monitor your backup server and how long you can live without backup.

 

There are lots of stats out there surrounding what companies deem acceptable for an RPO—they can live with data that is current up to 4 hours ago, 12 hours, or maybe 24 hours or more.  Frankly, it depends (don’t you love it when people say that?). What I mean is that it depends on the data in question that you are backing up and why, or more specifically, what level you or your business has classified that data (vital, sensitive, important, critical, etc.). Based on your decision, your RPO should follow.

 

Notwithstanding, you should familiarize yourself with the multitude of options available that can help you achieve a better RPO/RTO for your data center environment. Though not the topic of this post completely, it’s good to know that there are options to meet the most stringent SLAs/SLOs.  We have many solutions available today across both HP Software and HP’s Enterprise Group (servers, storage, and networking) that can easily do the trick!

 

 

Once we decide on criticality or have defined a classification to our data, it’s time to back it up and if the data is really important, we will likely back it up often.  We will use a mix of technologies’ approaches to do so, which may include backing up the file system (OS) and the data (e.g., an application agent specific to a business application like Oracle or SAP), and we might even leverage snapshots on our storage array (such as our snapshotting/replication capabilities on our HP StoreServ, i.e., 3PAR storage platform).

 

But if the backup server is down, well—we can’t really get a good backup to recover from.

 

So, that leaves me with the question to ask…

 

How long can you live without your backup server?

If the answer is never, join the club.  There’s a legion of IT admins out there that likely would agree.  But the vexing question is, “How do you ensure that your backup server is up and running?”

 

Do you cluster the backup server?  Run multiple instances?  Balance your backup jobs across multiple media servers and multiple backup managers?

 

Maybe the answer is yes. I hope—at least—you answered yes to a few of those questions.  How you monitor that infrastructure is now really important.  Do you want to know in real time if a backup job completes or fails?  How about media failures?  After all, it really sucks when you run that restore only to find a month from now that the job never completed, or that the media is bad.  That can make for a tough time during that quarterly disaster recovery exercise…talk about being the odd man out. You might as well have just stayed at home that day! Time to fire up Word and work on the resume…

 

Seriously, though. All of this could be avoided if we just exercised some caution and leveraged an age-old best practice: Try monitoring your backup server’s health, operation and related infrastructure.

 

Here are a few tips and items that you should likely monitor if you want to ensure your backup environment stays healthy:

 

  1. Monitoring solutions should represent the infrastructure topology that relates to backup environment (e.g., what it has under management and what it’s connected to)
  2. Monitoring solutions should monitor the services and processes that are responsible for the critical functions of the backup server (backup job engine, backup repository DB, media server, agent services, etc.)
  3. Agent monitoring and status of all backup jobs, restores and media copy operations should be logged and available for viewing
  4. Any and all events related to service availability or health should be represented graphically or in a tabular format (e.g. for viewing and correlation, real-time, near real-time or post fact)
  5. Providing recommendations on failed conditions should provide guidance and engage the backup admin to correct the problem. Knowing the problem is one thing. Having an intelligent monitoring solution to inform and instruct you as to how to rectify is another…

 

 

 

 

I encourage you to check back with me in a few weeks. In my next few posts, I will explore unique approaches and new technology solutions that will soon be available to address these critical areas of concern!

 

Click here to learn more about HP’s Data Protection Solutions.

 

#HPDPB

0 Kudos
About the Author

StephenSpellicy

Stephen Spellicy is known as a virtualization and data protection subject matter expert with 17+ years of experience in the technology industry. As the Director of Product Management for Data Protection within HP Software's Division, he manages a portfolio of software products that over 60k customers leverage worldwide. His expertise within the backup/recovery, storage, virtualization, and cloud technology areas have also gained him recognition with HP internally and externally, with many contributed articles published technology focused magazines & periodicals.

Events
Each Month in 2016
Online
Software Expert Days - 2016
Join us online to talk directly with our Software experts during online Expert Days. Find information here about past, current, and upcoming Expert Da...
Read more
Sep 30
Seattle, WA
OpenStack Days Seattle
OpenStack Days Seattle, September 30, is the largest gathering of OpenStack users and prospective users in the Pacific Northwest region.
Read more
View all