[ale] Stoopid Noobs

Scott Castaline hscast at charter.net
Tue Jun 20 21:26:12 EDT 2006


On Tue, 2006-06-20 at 20:14 -0400, Michael B. Trausch wrote:
> On Tue, June 20 2006 16:59, Sid Lane wrote:
> >
> > I once (early 2000) did a kill -9 -1 on a production sun box running
> > Oracle Financials (as well as the associated database).
> >
> 
> You know, I am kind of surprised that I haven't happened upon that one yet.  
> That's a pretty useful function (the "special" PID of -1, that is).  I will 
> have to keep that in mind for when I want to quickly restart my X session.
> 
> Nice, easy, and cleans everything up.  Perfect!  :-)
> 
> 	- Mike
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://www.ale.org/mailman/listinfo/ale

I was a Unix Admin for a home care company based in S. Fla. They had
regional offices in Boston, Memphis, LA & of course S. Fla. At the Fla
and Boston offices we had approx. 20 wanna be *nix servers, (SCO UNIX)
at each location, 10 more in Memphis and 2 in LA. I was in the process
of distributing a script for backups. Each server's script was unique
and basically would do dd periodically of the system, pack the file and
send it off through the WAN to another server that was the designated
backup server for that night. I wanted to just run a script that would
"blast out" each script to the right machine, but the software dept.
said that I had to do it individually and verify it. This is at 2PM
Friday. My family was expecting me to be home at 5:30, so that we could
head on up to Orlando to Disney World. (By the way this is about 10+
years ago, SCO OpenServer 5.0.0) So I started rlogin to each server. As
time went on I started rushing a little bit and was rloged into 10
systems cascaded and had lost track where I was. Needless to say the
wrong scripts got to the wrong servers. I finished about 5:10, or should
I say I thought I had finished. Mon AM arrives and I retrieved my backup
report from the weekend. I soon discovered that all of the servers had
backed up the wrong info, sent their respective packed files to the
wrong servers, so no tape backups were actually done. To make matters
worse, Boston had gotten hit with a storm and some of the servers were
down. The storms occurred after the backups should have been finished.
Talk about anxiety attacks......... 



More information about the Ale mailing list