Sys Admin truths that will never change | Hacker News Reader

Sys Admin truths that will never change

misterorange.com

25 pointschuhnk11 comments

Tech support truths, this doesn't suond like the realm of what I'd consider system administration.

It's more of a mixed list, but something like "make backups" is definitely a sysadmin thing.

> Linux may be powerful, but the command prompt and configuration files and filesystem obscurity will just as soon get you a pink slip if something goes wrong and no one knows how to fix it but yourself.

Hmm. So, I'm the only one who can fix it. Are they really going to fire me in that situation?

Even in 2005 it wasn't true and these days one can list a whole host of top organisations that use it: BBC, McLaren & Renault Formula 1 Grand Prix teams, the 3000 Munich Local Govt. Linux desktops, French National Police Force....

Retrospect? Mac support? Ah, smell that 2005.

Oh, and regarding #10, from The Practice of System and Network Administration,

"15.1.2 - Fix the Cause, Not the Symptom

A coworker reports that there was a problem and that he fixed it. “What was the problem?” we inquire. “The host needed to be rebooted.” “What was the problem?” “I told you! The host needed to be rebooted.” A day later, the host needs to be rebooted again."

Misterorange (Evan Erwin) has other side of him:

he's an avid Magic:the Gathering card game player, and he has some very good youtube webcasts called "The Magic Show", sponsored by a card game website for those of you interested in that: http://www.starcitygames.com/php/news/archive.php?Article=Ev....

(I know of this because I'm an avid M:TG player myself)

GFischer, if you're interested in helping me playtest my trading card game GeekStack (think Magic but where you're a scientist instead of a wizard), email me at peter @ geekstack . com

Booting should become a thing of the past eventually. But some form of reset from a known state will still happen.

Rebooting is becoming MORE prevalent. Yahoo! reboots IM servers on a schedule to keep them from leaking themselves to death - and trusts in orderly failover to make it nearly painless. Monolithic development environments (think JBOSS) are not perfect when released - rebooting keeps you in production until the next (flawed) release. and so on.

Belief in #7 leads to belief in #10.

True.

And #7 might be true with "at BigCo" at the end but here if you believe in #7 and #10 you're definitely going to prove #7 wrong.

Reply on news.ycombinator.com