Managing Statistics

Read a good paper on statistics today, by Karen Morton:

Click here to read this paper

format columns in awk with printf

I was struggling with awk to display the diskspace on my Linux host in the most convenient way. One problem was that a df command splits the output on two lines, but this is fixed by df -P.

Formatting the output with awk required using printf instead of print. With printf you can give formatting to the columns. I came on the following which is seems very usefull in my case.

oracle@myserver:~> df -Ph | awk '{ printf "%-12s%8s\n", $6 , $5 }'

Mounted Use%

/ 9%

/dev 1%

/boot 10%

/home 4%

/opt 31%

/tmp 20%

/usr 52%

/var 11%

/data 15%

/backup 83%

Tablespace point in time recovery

I'm glad it was just a test for this TSPITR....

RMAN> run

2> { recover tablespace USERS

3> until logseq 23 thread 1

4> auxiliary destination '/u04/oradata/';

5> }

Starting recover at 02-DEC-09

using target database control file instead of recovery catalog

allocated channel: ORA_DISK_1

channel ORA_DISK_1: sid=146 devtype=DISK

RMAN-00571: ===========================================================

RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============

RMAN-00571: ===========================================================

RMAN-03002: failure of recover command at 12/02/2009 11:41:25

RMAN-06455: point-in-time tablespace recovery requires Enterprise Edition

Ooops.... running SE.

Kill all processes - the rough way

On certain occasions, it might be needed to kill several hunderds of processes on a linux machine. Of course you should NOT do this on a production site, but the use of xargs might be handy:

ps -ef | grep oracle | grep -v bash | awk '{ print $2}' | xargs kill -9

With the grep -v you exclude your own putty session.