[Orca-users] How to exclude filesystems?

Jon A. Tankersley jon.tankersley at eds.com
Thu Jan 27 09:54:32 PST 2005

It never did.... And it dealt with it poorly.  We hacked it to tattle about
the bad lines (and said what it found vs. what it expected) and keep going.
The default action was to bail.
There was a version of orcallator that didn't create a new file whenever it
detected changes or was restarted (like after a reboot).  We rewrote some
preprocessing scripts to split the files apart when there were multiple
headers inside.  This, along with the appropriate edits to the find_files
fixed the problem until the new orcallator came out.

Jon A. Tankersley 
EDS - ITO-SD/US Capacity and Performance 
MS 401 
4000 North Mingo Road 
Tulsa, Ok 74116 
( Phone:+1-918-939-4820 (8-639-4820) 
+ mailto:jon.tankersley at eds.com 



From: orca-users-bounces+jon.tankersley=eds.com at orcaware.com
[mailto:orca-users-bounces+jon.tankersley=eds.com at orcaware.com] On Behalf Of
Duncan Lawie
Sent: Thursday, January 27, 2005 7:59 AM
To: David Michaels; Dmitry Berezin
Cc: orca-users at orcaware.com
Subject: RE: [Orca-users] How to exclude filesystems?

I've never seen a version of orca that could handle changes in column count
- though there was talk of it back in late 1999.

-----Original Message-----
From: orca-users-bounces+tyger=hoopoes.com at orcaware.com
[mailto:orca-users-bounces+tyger=hoopoes.com at orcaware.com] On Behalf Of
David Michaels
Sent: 26 January 2005 23:33
To: Dmitry Berezin
Cc: orca-users at orcaware.com
Subject: Re: [Orca-users] How to exclude filesystems?

Dmitry Berezin wrote: 

Actually no. The number of columns is always the same within one data file.
If it changes, orcallator will start a new data file. AFAIK, orca can not
handle format changes within one data file. Also, something like

Oh?  I could swear I've seen some data files with multiple header lines in
them.  So, every time I mount a new filesystem, or unmount an old one, orca
creates a whole new datafile with a new header?  *looks at his data
directory*  Oh, look at that.  9 data files on Monday.  Hmm.  Was there ever
a point where Orca supported column changes mid-stream?


head -1 data_file | gawk '{for (i=1;i<NF; i++) if
($i~"vxckpt|checkp|verita") print i, $i}' 


will give you the column numbers that you want to exclude. Or this


gawk '{for (i=1;i<NF;i++) {if (NR==1 && $i~"vxckpt|checkp|verita") np[i]=1;
if (np[i]!=1) printf $i" "};print ""}' data_file > new_data_file


will create a new data file without the columns that you don't want.

Indeed, if the columns don't change for a particular file, the gawk becomes
almost trivial. :)  Thanks for the feedback -- the perl regexp did the trick
for future datafiles, and it sounds like the data file cleaning will be
easier than I first thought.




	David P. Michaels

Senior Multi-Disciplined Engineer II W.H.
Platform OS Unix
720.858.5952 fax
720.521.0561 pager

dragon at raytheon.com aka "Dragon" 
"I wonder what news is doing..."

 news at newshost <29> ps -fu news      

 news 18624 12367 2 0:00 makehistory 

"News is making history."


-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/orca-users/attachments/20050127/70d8d961/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Raytheon.gif
Type: image/gif
Size: 481 bytes
Desc: not available
URL: </pipermail/orca-users/attachments/20050127/70d8d961/attachment.gif>

More information about the Orca-users mailing list