Operating in Failure

Question

Operating in Failure

Steve Jones - SSC Editor

SSC Guru

Points: 734449
More actions
January 12, 2011 at 8:11 pm

#142169

Comments posted to this topic are about the item Operating in Failure

Viewing 7 posts - 1 through 6 (of 6 total)

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

Russ1of5 SSC-Addicted Points: 490 More actions · Answer 1

Good Article. I started out my career back in the 80’s as a maintenance mainframe programmer and I still work with existing systems both mainframe and web. So I am always thinking of maintenance when I have to dig into code, database design, or the architecture. Sometimes the best way to do something is not always the quickest or most “slick” way. Time and time again I see new applications being created by third parties and then maintained by the “maintenance" staff. Then it takes some time years for that system to become stable (if ever). What a great world it would be if the developers of new systems would have to maintain their work for the first couple of years. Then they could maybe learn about what its like to fix something after it fails, which like you said, it will fail.

TravisDBA SSCoach Points: 15780 More actions · Answer 2

"So plan to fail and you'll be better off." and I would add "and have your notifications processes well in place so that your teams are notified in a timely matter when this does happen.":-D

"Technology is a weird thing. It brings you great gifts with one hand, and it stabs you in the back with the other. ...:-D"

jay-h SSCoach Points: 18816 More actions · Answer 3

TravisDBA (1/13/2011)
"So plan to fail and you'll be better off." and I would add "and have your notifications processes well in place so that your teams are notified in a timely matter when this does happen.":-D

Perhaps it should be 'plan for failure...'. 'Plan to fail' sounds too much like the system is expected to fail. Failures will happen, but the system needs to tolerate failure.

...

-- FORTRAN manual for Xerox Computers --

Alan Vogan SSCrazy Points: 2513 More actions · Answer 4

Good to remember, though I agree with the above, it should be 'Plan FOR failure.' The planning to fail is a funny little catch phrase but enough of those in the work place can get you down (or turn into an enterprise, ie. http://www.despair.com).

I find it valuable when testing to have my manager make a run. He's really good at messing up the intended use of a program, entering wrong data, attempting unintended processes, and sometimes he'll just bang on the keyboard and hit enter. It's also quite entertaining. But even after all that, there are still failures that we hadn't planned for. Maybe the phrase should just be, 'Life happens, what the heck?'

Tom Garth SSCertifiable Points: 6193 More actions · Answer 5

A primary measure of any application should be it's ability for "GRACEFUL DEGRADATION".

Tom Garth
Vertical Solutions[/url]

"There are three kinds of men. The one that learns by reading. The few who learn by observation. The rest of them have to pee on the electric fence for themselves." -- Will Rogers

Steve Jones - SSC Editor SSC Guru Points: 734449 More actions · Answer 6

Tom Garth (1/13/2011)
A primary measure of any application should be it's ability for "GRACEFUL DEGRADATION".

Too true, and too often I see developers that don't ever expect a query to fail or a connection to fail and don't include a reasonable retry, or option to retry.