Do you have a DR plan?

Discussion in 'Business & Enterprise Computing' started by Gecko, Jun 18, 2012.

  1. Gecko

    Gecko Member

    Joined:
    Jul 3, 2004
    Messages:
    2,716
    Location:
    Sydney
    I've been absolutely horrified lately by the number of businesses that I know of with absolutely no DR / BCP plans in place. Or the ones who say "our cloud provider handles that for us". :wired:

    Do you have a formal DR plan? An informal plan? Or even a way that you could get something running again in a hurry if you had to?

    Where I am now, we have ~100 servers (mix of physical and virtual) and a semi-formalised DR plan (still a work in progress, but I think we could work with it in a pinch).
     
  2. ra66it

    ra66it Member

    Joined:
    Oct 12, 2007
    Messages:
    807
    Location:
    Melbourne
    Yes I do. I was unable to get any interest from management\executives, and DR equipment wasn't a priority. But I do have a DR document that nobody has bothered to read.

    Yet I keep getting told that uptime is absolutely vital. Work that one out :rolleyes:
     
  3. Iceman

    Iceman Member

    Joined:
    Jun 27, 2001
    Messages:
    6,647
    Location:
    Brisbane (nth), Australia
  4. bsbozzy

    bsbozzy Member

    Joined:
    Nov 11, 2003
    Messages:
    3,916
    Location:
    Sydney
    Yeah, I spent a lot of time setting up DR equipment (servers, routers, switches, SAN), replication, scripts to be run to cut over data, documentation over the last 2/3 years where the people above me severely failed.

    All they did was write a basic document that pretty much said, get latest tapes from offsite, buy some cheap off the shelf hardware, restore.

    The only thing I haven't really bothered with is workstations, our DR site has different rooms with different machines, so rather than fucking around getting and image working for all of them, they already have a basic XP image which is fine, all that needs to be done is to setup a RDP shortcut to the RDP farm and then they are good to go.

    I have done some testing with bringing the site online (CRM, file server, networks), and in an isolated LAB I have done the rest that I couldn't do without affecting the rest of the network (e.g. Exchange). We are yet to do a full test, I have given up trying to get them to do one.
     
  5. -Antiskeptic-

    -Antiskeptic- Member

    Joined:
    Aug 14, 2006
    Messages:
    950
    Location:
    Reservoir, VIC
    Yeah have both here and also a "cold site" ready to go @ a moments notice.

    Do a mock DR twice a year for compliance reasons, get users to go out to the cold site and do UAT testing.

    Usually goes pretty well :thumbup:
     
  6. s4mmy

    s4mmy Member

    Joined:
    May 20, 2004
    Messages:
    1,944
    Location:
    Melbourne
    www.seek.com.au :thumbup:

    haha, but in answer to your question, yes.
    Has it been tested in the last 18months... no.
     
  7. fR33z3

    fR33z3 Member

    Joined:
    Jul 16, 2001
    Messages:
    2,164
    Location:
    Perth
    at my last place only the finance and hrmis end user applications were categorized as requiring disaster recovery. Twas rather interesting that things like email, www servers, phones, doc mgmt were all ignored. That said, i wasn't too upset about that - best to do a few things properly than everything crap.
     
  8. Tinian

    Tinian Member

    Joined:
    Jan 3, 2009
    Messages:
    13,262
    Location:
    15.0° N, 145.63° E
    Yes - a separate DR site.
     
  9. s4mmy

    s4mmy Member

    Joined:
    May 20, 2004
    Messages:
    1,944
    Location:
    Melbourne
    That's not a plan... its just a location. A document on how you are going to fail over to that site... and what you are going to fail over to that site is a plan. :thumbup:
     
  10. Tinian

    Tinian Member

    Joined:
    Jan 3, 2009
    Messages:
    13,262
    Location:
    15.0° N, 145.63° E
    Ye, we've gone to the trouble of replicating an entire data centre but we don't have a plan. :rolleyes:
     
  11. IACSecurity

    IACSecurity Member

    Joined:
    Jul 11, 2008
    Messages:
    738
    Location:
    ork.sg
    You didn't specify anything like that, so of course he his going to comment.

    Lots of places have 2 DC's an no clue in the world from a DR perspective, so hes made a very legitimate point.
     
  12. s4mmy

    s4mmy Member

    Joined:
    May 20, 2004
    Messages:
    1,944
    Location:
    Melbourne
    Exactly, plenty of places also "replicate" but don't necessarily have a plan on how they are going to fail over to that replica.
     
  13. Urbansprawl

    Urbansprawl Member

    Joined:
    May 5, 2003
    Messages:
    531
    Or they back stuff up to an offsite 'cloud' provider without doing the maths on how long it will take to re-download everything or having a contractual delivery time for delivery of those backups.
     
  14. Tinian

    Tinian Member

    Joined:
    Jan 3, 2009
    Messages:
    13,262
    Location:
    15.0° N, 145.63° E
    Obviously my fault for assuming a level of knowledge/ capability in what is really a SME forum. A separate DR site is not just another data centre. So let me be more detailed:
    • Site A with HA on prod. Site A replicates to site B depending on level.
    • If site A goes down, network automatically switches to site B.
    • Site A would be rebuilt from back-ups with delta applied.
     
  15. dave_dave_dave

    dave_dave_dave Member

    Joined:
    Mar 17, 2004
    Messages:
    2,600
    Location:
    Gold Coast
  16. GreenBeret

    GreenBeret Member

    Joined:
    Dec 31, 2001
    Messages:
    19,377
    Location:
    Melbourne
    I have a super awesome DR plan, but no budget for it. Despite all the sabre rattling, nobody gives a shit about their own work around here anyway, so why need disaster recovery? Waste of money that could be spent on other useless but prettier things.

    Best "DR plan" I've come across is some guy telling me theirs is RAID5 with two hot spares. Two!
     
  17. ewok85

    ewok85 Member

    Joined:
    Jul 4, 2002
    Messages:
    8,074
    Location:
    Tokyo, Japan
    Yup - our DR plan is purely on paper. It's pretty much "has A B and/or C happened? Yeah, we're screwed"
     
  18. tensop

    tensop Member

    Joined:
    Mar 26, 2002
    Messages:
    1,194
    this will sound stupid... but how do you failover to a DR site with active directory and not have USN issues, or do you not bring up replicas of the DC's(or dns) that are affected? just switch over dhcp scope for the affected site with the new dns servers to reference?
     
  19. Smokin Whale

    Smokin Whale Member

    Joined:
    Nov 29, 2006
    Messages:
    5,174
    Location:
    Pacific Ocean off SC
    For a lot of people, I am the disaster recovery plan... For myself though, I have servers duplicated across two buildings, which has backup images of every workstation we have. However both these buildings share the same power, so if it's taken out, we're screwed... can't get around that though, for a small business (where most my important work is cloud hosted), it'll do.
     
  20. bugayev

    bugayev Whammy!

    Joined:
    May 15, 2003
    Messages:
    4,096
    Location:
    Melbourne
    My disaster recovery plan is to immediately book leave. As I can approve my own leave, this is a suitable plan for the recovery of any potential or actual disaster.
     

Share This Page