I am particularly concerned about sites that have a single administrator
Because that one person has so much responsibility and no-one to back-up
them up in case of:

  sickness,
  holiday,
  training
  doctor's appointment
  bus / train / car wreck

I am thinking of Spring '96 in Adelaide and the "disaster" that happened there.

IF any Administrator who is "alone" at a site were killed / mangled / put out of commission  in a car-wreck, we'd be particularly vulnerable...

SOOOOoooooo....... - as a result - let me suggest some measures that I would like to see implemented at any site that is in this situation.

IT won't ""SAVE THE DAY IN ALL CASES"" but it will certainly considerably reduce our risk(s)::

1.) identify a "temporary backup system admin" in the Engineer community - with the agreement of the Managing Director, and engage in some brief / preliminary training on "easy / simple tasks"...

2.) write your root password on a piece of paper - fold it three or four times, put it in an envelope, seal it, write your name & date on the sealed area,  and tape it to the bottom of your middle desk drawer (or somewhere else where your "tempsysadmin" can find it in an emergency...

  DO THE SAME THING WITH:

   info on:

     office-site storage access codes / passwords / phones / e-mails / 
     Keys for tape & other storage lockers  (make sure someone besides YOU
        can get in to pick-up off-site storage tapes, and knows HOW)
     pager / phone numbers for one & all vendors
     passwords for other "equipment" like routers, PABX config screens, etc.


3.) leave root password on some-other-admin's voice-mail (even in another town) whenever you change it - at least - in a pinch someone might need to telnet into a machine at the site that has lost a system admin.

4.) Make certain single-admin site admins have pager or phone for being contacted during "reasonable" non-office-hours.

5.) MAYBE you want to share password(s) with me too? - just in case..

6.) Make sure there's a section in DRP for the "disaster" that might happen if either one of you disappears

7.) Make sure your manager has a section in his OMDR for you and/or your position, in case you depart - someone is available to back-you-up...

8.) THINK, EVERY DAY - ABOUT HOW someone would pick-up your job if you weren't there tomorrow - and DOCUMENT, DOCUMENT, DOCUMENT, DOCUMENT everything... 

This is really just a "variation on the theme" of "Disaster Recovery Planning". 

But it's a very SPECIAL CASE in sites where S/W Centres have a "single point of failure" (a single System Administrator) - in July '97 - that's Beijing and Nanjing.