Networker

Re: [Networker] Sungard DR testing

2007-07-27 10:12:01
Subject: Re: [Networker] Sungard DR testing
From: Stan Horwitz <stan AT TEMPLE DOT EDU>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Fri, 27 Jul 2007 10:05:32 -0400
                
On Jul 27, 2007, at 8:36 AM, Jon Fraley wrote:

We are currently looking at using Sungard for DR purposes.  I was
wondering if anyone could share their experiences with Sungard in
recovering the Networker server. Since the testing time is allocated in 8 hour slots, how long does it normally take to recover the server? How do you get much more in than the Netwoker recovery? We are currently on
7.2.2 running on Linux.

John, we use Sungard here at Temple for our mainframe DR and also a few other servers such as our LDAP farm, email, and www.temple.edu. Due to the cost involved, we don't keep a spare NetWorker server at Sungard, but we do keep it and the majority of our backup media off- site (a few blocks away from our primary data center). If we lose our primary computer room, we will not lose our NetWorker server unless there is some sort of nuclear holocaust!

There is no way anyone on this list can provide you with specific details without more information. For example, how large are your NetWorker server's media and client file indexes and how many clients would you need to recover at Sungard? The more metadata your NetWorker server manages, the more time recovering it to a standby NetWorker server it will take. How fast is your disk storage and tape drives at Sungard and your Sungard NetWorker server?

Although I can't give you specific numbers, I can tell you one thing that's of critical importance for your employer's business continuity. You need to know how long it takes to deploy critical servers and put them in production if they blow up, flood, are stolen, or a serious hardware problem occurs. Don't go by any numbers you get on this list; you have to do good testing to see for yourself. You need to do a full blown DR rehearsal at least once a year for each critical server. Sungard will also make you pay for this service.

We do DR drills several times a year. In any given week, we recover at least one critical server to a DR server (usually not at Sungard though) and our success rate hovers around 99%! Like many organizations, we get audited for our DR practices, so we are careful to make sure things work during an audit. My trick is to tell those who are about to be audited to do the DR drill a few days before the audit to make sure it works and that they understand the process. Amongst other things, my job is to make sure our NetWorker server and the required resources are available for these audits. Doing practice DR drills isn't always practical, but if possible, encourage it.

In fact, I will be involved with our next mainframe DR rehearsal at Temple's Sungard site in a few weeks to learn how our mainframe DR works; we do this sort of thing a few times a year. Past experience says it takes 12 hours for us to migrate to one of Sungard's mainframes and test the migration, but that may be totally different in your situation and our timing should improve considerably with this next DR rehearsal because we will be testing new backup media.

Depending on your situation, you might be better off with a disk-to- disk replication solution to replicate your NetWorker server's metadata to storage at Sungard instead of relying on recovering the data from tape when a DR situation develops. Even for a small NetWorker server, it can take several hours to migrate the metadata to a different server. A NetWorker server with 20GB worth of metadata will take a lot less time to recover then one with 200GB, but you can also pick and chose which clients' metadata you recover after you recover your media database.

I am working on a project now where we are transitioning Temple's IBM mainframe backups (payroll, student grades, accounts payable, financial management, etc.) from old IBM mainframe tape cartridges to a virtual tape library. That VTL tape library sits in the same room as our mainframe, but its data gets replicated to our Sungard daily. We have not tested this yet, but that test is the one I will be watching in a few weeks. We tried doing this test last month at Sungard and it failed miserably through no fault of our own, so we will be doing it again soon!

My role in this project is to back up our mainframe VTL via NetWorker to LTO-3 media for off-site storage for five years and to know how the entire mainframe backup process works. I doubt we'll need to recover NetWorker tapes at Sungard simply because we rsync the VTL data to Sungard daily, which is also what we do with all the other DR servers we have there.

--
Stan Horwitz
Temple University
Enterprise Systems Group
stan AT temple DOT edu

CONFIDENTIALITY STATEMENT: The information contained in this e-mail, including attachments, is the confidential information of, and/or is the property of, Temple University. The information is intended for use solely by the individual or entity named in the e-mail. If you are not an intended recipient or you received this in error, then any review, printing, copying, or distribution of any such information is prohibited. Please notify the sender immediately by reply e-mail and then delete this e-mail from your system.

To sign off this list, send email to listserv AT listserv.temple DOT edu and type 
"signoff networker" in the body of the email. Please write to networker-request 
AT listserv.temple DOT edu if you have any problems with this list. You can access the 
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER

<Prev in Thread] Current Thread [Next in Thread>