Incident Response - The Framework

  • Steve Jones - SSC Editor

    SSC Guru

    Points: 716331

    Comments posted to this topic are about the content posted at http://www.sqlservercentral.com/columnists/sjon

  • Jonr

    SSCrazy

    Points: 2566

    A very relevant article at present Steve. Just a few things to add:

    Firstly, have the contact details for the site services people on the list, along with contact numbers for power and telecoms providers in case of building damage or power loss. If the utilities can provide direct line numbers even better, as the public numbers tend to get hit hard during natural disasters. Also arrange for key people to use alternative entrances to the building - a power failure can often leave electrically operated security doors unusable.

    In a big organisation, it's a good idea to lodge the primary contact sheet with a trusted member of staff, maybe someone from IT security. It should be kept in a sealed envelope, preferably in a fireproof and waterproof safe. This not only protects the contents from disaster, but also prevents people being contacted in non-emergency situations by management.

    Another option to consider is providing emergency laptops with dial-in capability.

    These can be useful if roads are taken out or weather conditions are severe, and also allow people with long commutes to be useful without being present. All laptops should be tested for connectivity from the staff's home, and all security permissions and expected functionality checked before a disaster. Also ensure that the staff are covered for insurance on the laptops, in case of damage/theft from their home.

    Get expenses sheets in with the DR plan, so that people know they're not going to be out of pocket with food and accommodation, and can concentrate on the job. Better still, have someone with a company credit card who's authorised to buy food and accommodation, people feel more comfortable about it than having to reclaim it later.

    If you've got an off-site facility that the business can move to, make sure that staff are aware of how to reach it, have contact numbers for it and ideally have them visit the facility before a disaster occurs. Nothing worse than reading a map at night in a storm trying to find an unmarked building !

    One final thought : When you've finished the framework, how about adding the collective knowledge into the back of The Best of SQL Server Central Volume 3 as a handy, readily accessible checklist for DBAs?


    Jon

  • Andy Warren

    SSC Guru

    Points: 119676

    Not a bad idea at all for Volume 3!

  • sa24

    SSCrazy Eights

    Points: 8027

    Excellent Article.

    Also something which turned out to be very handy for us during the power outage last year was radios. With cell phones running out of signals/battery in the data center, this was a handy tool.

    I'm not sure if you would be covering this in the next article but one thing i thought of mentioning for a big organization is to have the Application Classification and have a list with the order in which the applications would be recovered.

     

  • Steve Jones - SSC Editor

    SSC Guru

    Points: 716331

    Good idea for Vol. 3. I'll keep some stuff handy for that.

    In the next article, I'll look at some of the ways in which we actually responded to disasters and incidents. Not sure about an app order since in a crisis the order shifts. Also it's something that's hard to keep up to date, as I'll discuss in a third article.

  • ckempste

    SSCoach

    Points: 17983

    Hi Steve

    Good article.  My lastest ebook covers this also.  It is important to note that plans need to be crafted in context to the business, its cultural aspects of work and most importantly, be contextually relevent to its readers.  To really kick this sort of initiave off you need management buy in, something that is often overlooked and can be difficult to persist in larger organisations.

    Also be aware of the "greater initiative", meaning, is the business focused on industry standard service delivery models such as ITIL.  What I am saying here is "dont reinvent the wheel"

    Be careful with the work "incident" verses "problem" or "disaster".  All have very different meaning and subsequent actions, especially as you move between support tiers.  Using the team for all types of incidents is an interesting one and would like to hear about the mix of people in your team, did you funnel calls through a help desk and at what times did the group meet etc.., did you break down the support tiers etc?

    Cheers

    Ck

     

     

     


    Chris Kempster
    www.chriskempster.com
    Author of "SQL Server Backup, Recovery & Troubleshooting"
    Author of "SQL Server 2k for the Oracle DBA"

Viewing 6 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic. Login to reply