From: Kevin Gillins kgillins@yahoo.com
To: oraapps-dba@cpa.qc.ca
Subject: Re: UNIX Printing Problem
Check your profile option %Copies% at all levels in the application. There may be multiple copies being requested at an application or responsibility level.
Try the same request against a different print queue to verify if it is just one printer or if it is for the specific user.
Kevin
===
Kevin Gillins
Oracle Applications Consulting
WEB PAGE <
Date: Fri, 23 Jul 1999 10:40:40 -0500
From: Michael S Barone Michael.S.Barone@Cummins.com
To: oraapps-dba@cpa.qc.ca
Subject: Re: UNIX Printing Problem
Gagan:
You can FORCE a number-of-print-copies by changing/modifying your
UNIX
"PRT" filename in the $FND_TOP/srw/?????.prt file. You can determine
which UNIX PRT file your report is using via:
System Admin (Responsibility)
Navigate
Install
Printer
Driver
Note your "SRW Driver" ... usually HPL (Landscape HP) HPW (LandWide
HP)
and HPP (Portrait HP). These UNIX Files can be found in the
$FND_TOP/srw
directory with a "prt" extension.
For Example: our HPP.prt file contains the following:
before report "^[&l2X"
between pages control(L) "^[&l2X"
Where the "2X" control-sequence forces 2-Copies to be printed.
Date: Fri, 23 Jul 1999 16:22:42 -0400
From: "Timothy Brewer" tbrewer@tbsolns.com
Subject: Re: UNIX Printing Problem
(I presume that you are selecting 1 copy in the standard report submission
screen)
Continue to isolate this problem.
1. Try a different username
2. Try a different printer TYPE, if possible.
3. Finally, ensure that the lp command in the printer set-up is actually
using PROFILES.COPIES instead of being hardcoded to 2 (someone possibly
changed to lp -d%PROFILES.PRINTER% -n2 for testing......).
Please update the list to the results of the above and maybe we can help
further.
HTH,
Timothy J. Brewer
Senior Oracle Consultant
TB Solutions, Inc.
Script to extract employee information
Date: Mon, 26 Jul 1999 22:31:48 GMT
From: "manu b" manu024@hotmail.com
To: oraapps-dba@cpa.qc.ca
Subject: some advanced shell scripts : Real urgent
Hi Unix gurus,
I am back again..I have a file for employee records. Each employee has 4
lines in it. The first line for each employee record contains the employee
number. In the second line there may or may not be a benefit plan for that
employee.
I am tring to do extract info for each employee from 2nd line for
the benfit plan with use of 'grep' and 'cut'.
I do something like this
--For the employee number
cat file|grep -i emp_num|cut -c10-15 >e_num
--For the employee benefit
cat file|grep -i plan_nm|cut -c10-15 >e_plan
Then I do paste
paste e_num e_plan > new_file
The problem in the above is that it will not have all the employees listed
with their benefit values as some of them may not have the plans. I want to
print "0" in front of the empl num when he does not have the benefit plan.
Will anyone suggest some alternative to this requirement..
Appreciate an early repsonse from anyone..
thanks
Date: Tue, 27 Jul 1999 09:30:17 +0100
From: Philip West Philip.West@exco.co.uk
To: oraapps-dba@cpa.qc.ca
Subject: RE: some advanced shell scripts : Real urgent
Firstly, why not code sql to put 'no-plan' or whatever into flat file in the
first place.
Otherwise: If the format of the line is determinable you could use sed.
example file here
...........................
plan_nm 234
plan_nm 234
plan_nm
plan_nm 888
..........................
*note that here we have the string 'plan_nm ' including a space as minimum
followed by optional employee number
sed "s/plan_nm $/0/" file > e_plan
your new_file will then hold something like
emp1 234
emp2 234
emp3 0
emp4 888
Best Regards
Phil West - Oracle Financials DBA
Telephone: 44 171 9509385
E-mail: philip.west@exco.co.uk
Combining Files - script for
Date: Tue, 27 Jul 1999 11:45:40 -0500
From: Bharat Patel bpatel@DOMINOAMJET.com
Subject: RE: Help : Unix - Shell Scripts
regarding the unix play around. I have following requirement.
1. I am creating a file by using ls -lt|grep 'xxx' > file1.txt
2. cut -c n-n,n-n file1.txt > file2.txt
3. vi and replace the 'xxx' from file2 with 'yyy'
4. run this file to do the work(delete some temp files)
I am looking to create my file2.txt with replace character in a one command.
how can I do that?
Bharat Patel
D.B.A.
bpatel@dominoamjet.com
847-244-2501 ex 1249
Date: Tue, 27 Jul 1999 15:49:47 -0400
From: cwroblewski@orioncapital.com
To: oraapps-dba@cpa.qc.ca
Subject: RE: Help : Unix - Shell Scripts
If I am understanding you correctly do the following
- vi file2
- hit shift :
- type the following 1,$s/xxx/yyy/g
This is a sed command that will do a global replace.
Date: Tue, 27 Jul 1999 16:00:35 -0500
Date: Wed, 28 Jul 1999 10:00:23 +0400
Date: Wed, 28 Jul 1999 07:16:30 -0500
Date: Wed, 28 Jul 1999 10:49:05 -0500
Date: Thu, 29 Jul 1999 10:09:15 +0400
Date: Thu, 29 Jul 1999 11:44:34 +0530 Unix Issues - Picking selective files
From: Singh, Gagandeep [mailto:Gagandeep.Singh@scottishpower.plc.uk]
Date: Wed, 31 Mar 1999 11:14:12 -0500
# Useful shell function for bailing out Modify UNIX Reports with NT
From: pkandra@civicdevelopment.com [SMTP:pkandra@civicdevelopment.com]
Date: Wed, 7 Apr 1999 15:09:06 -0400 UX script to remove files in APPL_TOP
Rudy Wahab rudy@fcm.ed.fujitsu.co.jp wrote:
Date: 4 Apr 99 18:42:02 EDT
Greg Hoover wrote:
Date: Mon, 05 Apr 1999 10:42:42 -0500 FTP and File Management
From: "Bubba New" bubba_new@hotmail.com
Date: Fri, 09 Apr 1999 05:12:39 PDT Monitor Printer queue at Unix level
Date: Wed, 1 Sep 1999 10:25:46 -0400
Date: Wed, 1 Sep 1999 19:28:20 +0100 Control characters in man document
From: Subramaniam Venkat venkat@irs.co.id
From: Disha Dalal[SMTP:dishadalal@hotmail.com]
Date: Wed, 6 Oct 1999 19:02:23 -0400
From: Bharat Patel bpatel@DOMINOAMJET.com
To: "'oraapps-dba@cpa.qc.ca'"
Subject: RE: Help : Unix - Shell Scripts
No, I want to use all in one command line I don't want to use VI editor at
all. I got the reply from other forum users. only thing I am playing around
is to combine ll|grep command and others.
thanks.
Bharat Patel
D.B.A.
bpatel@dominoamjet.com
847-244-2501 ex 1249
From: mzaveri@eppco.co.ae
To: oraapps-dba@cpa.qc.ca
Subject: RE: Help : Unix - Shell Scripts
Dear bharat,
You can just use the same command in
sed too.
sed 's/xxx/yyy/g' src > out
Hope, I understood it ok.
Also you can use -n flag, to just print those replaced lines and stop
non-replaced lines to print..
regds
From: Bharat Patel bpatel@DOMINOAMJET.com
To: "'oraapps-dba@cpa.qc.ca'"
Subject: RE: Help : Unix - Shell Scripts
If I am not wrong ur name is Mehul Zaveri, Right?. Thanks Mehul and all
those who help me out.I got the solutions from Denny. I hope I will get the
same kind of help in future.
Bharat Patel
D.B.A.
bpatel@dominoamjet.com
847-244-2501 ex 1249
From: Gopi_Vakulabharanam@kellwood.com
To: oraapps-dba@cpa.qc.ca
Subject: RE: Help : Unix - Shell Scripts
If you are trying to remove some selective files try this:
ls -lt | grep xxx | awk '{print $9}' | xargs rm -f $1
HTH
From: mzaveri@eppco.co.ae
To: oraapps-dba@cpa.qc.ca
Subject: RE: Help : Unix - Shell Scripts
Just to add on here,
Is it not risky to grep on ls -lt? , it may grep xxx even in other fields
of ls -lt. isn't?
I would advice to exchange grep and awk commands.
No need, If you are confident that in future your string (say 'xxx') will
not be a part of rest of the fields. But this assumption may put you in
big trouble, because you are using it with rm.
Just to highlight the danger point.
From: "Akbar Ibrahim" akbari@wipsys.soft.net
To: oraapps-dba@cpa.qc.ca
Subject: Re: Help : Unix - Shell Scripts
Why not use the find command to find and delete the files. This could be
one good way to do it. grep could be dangerous as has already been
poinged out.
here's the grep command (to find all files with the extension tmp)
find /mydirectory -name "*.tmp" -exec rm {} \;
Note 1 : The find command descend down the specified directory and
performs the exec action on all the files found in all the
subdirectories and their subdirectories.
Note 2 : The *.tmp has to be quoted to prevent the shell from
interpreting it.
Note 3 : The find command has other options such as find files which
have not been modified for over n days or which have been modified
within n days, which are owned by a particular user or group etc.
Hope this is useful
Akbar
Sent: Wednesday, March 31, 1999 9:58 AM
Subject: Unix Question
Hi Unix Gurus!
I need to write a unix script ...which picks up all file beginning
with 9 and extn 'lst' and rename their file extension to 'txt'.
eg a file 9xyz.lst ---> 9xyz.txt
for i in `ls 9*.lst`
do
.......
done
Any ideas would be appreciated.
Thanks
GD
From: "Ellis, Rob"
Subject: RE: Unix Question
This ought to work.
for i in `ls 9*.lst`
do
basename=`echo $i | cut -d'.' -f1`
mv ${basename}.lst ${basename}.txt
done
err_exit() {
echo $* &2
exit 1
}
# Make sure that there are three arguments
[ $# -eq 3 ] || err_exit "usage: $0 directory old_ext new_ext"
start_dir$1
old_ext$2
new_ext$3
# Change to the specified directory
cd $start_dir || err_exit "Cannot change directories to $start_dir"
# Use find to recurse through the directories
find . -type f -name \*$old_ext -print |
while read pathname
do
dir`dirname $pathname`
file`basename $pathname`
newname$dir/${file%$old_ext}$new_ext
mv $pathname $newname
done
exit $?
---------------------------------------------------------------------------
Philip Chee: Tasek Corporation Bhd, P.O.Box 254, 30908 Ipoh, MALAYSIA
e-mail: philip@aleytys.pc.my Voice:+60-5-545-1011 Fax:+60-5-547-3932
From: "James J. Morrow" jmorrow@csac.com
Cc: jmorrow@csac.com
Subject: Re: Unix Question
You can also do this as a simple in-line script (/bin/ksh)
ls -1 9*.lst |while read LINE
do
export BASE`basename ${LINE}|cut -f1 -d"."`
export DIR`dirname ${LINE}`
mv ${LINE} ${DIR}/${BASE}.txt
done
-- James
----------------------------------------------------------------------------
James J. Morrow E-Mail: jmorrow@csac.com
Systems Consultant USSnail: 6380 LBJ Freeway #181
Computer Systems Authority Dallas, Texas 75240
Dallas, Texas (972) 960-0180 x244
Sent: Wednesday, April 07, 1999 1:42 PM
Subject: RPT: How to modify UNIX reports with NT
Can anyone please tell me all of the necessary steps to modify a UNIX
reports 2.5 file with the NT version of reports 2.5 ?
I get the file from Unix, open it with NT and I always get a 'There exist
uncompiled program unit(s)'.
I know I'm missing a step, I just don't know what.
Thanks in advance
Peter Kandra
pkandra@civicdevelopment.com
From: "Vallapareddy, Krishna" kvallapareddy@noblestar.com
Subject: RE: How to modify UNIX reports with NT
This problem usually arises when there are any program units like libraries
attached to the report. The rdf file stores a link (path) to the library
files which are used
in the report. When you ftp the rdf file from unix to NT ,you are not able
to make a connection to the library files still on the Unix system and so
the
program units cannot be retreived as you open the report on the NT side.
What I would suggest is that you open the reports
using any X-Windows software on the unix side. You could use the command
ar25desb at the $ prompt.
If you are uncomfortable with working on X-Windows, then just make whatever
modifications you need to the report on the NT side without
bothering about the error messages and then transfer the file back to the
Unix system. Once on the Unix box, you could
always open the original report which you copied and also open your modified
copy(with X-Windows), then you can simply copy all the program units
from the original to the custom report by drag and drop. You should be able
to compile this way.
Well atleast it has worked for a lot of us.
Hope this helps.
Good Luck,
Krishna Vallapareddy
Oracle Financials Consultant
Noblestar Systems Corporation
(703) 464-4000 ext 2508 or
(800) 470-4668 ext 2508
Hi,
Does anyone have a unix scripts to remove all files in a
particular directory that are 10 days old?
I want to remove all files that are 10 days and above old in
APPL_TOP/common/log
Please help ..thanks in advance
regards,
*************************************************************************
Rudy Wahab Phone: 07-454 2111 ext 211/313/218
System Analyst email: rudy@fcm.ed.fujitsu.co.jp
Information Services Section
Fujitsu Components (M) Sdn. Bhd.
Batu Pahat, Johor.
Malaysia
*************************************************************************
From: Dinesh Chandra dchandra@usa.net
Subject: Re: [removing files in APPL_TOP/common]
Hi Rudy,
Try following
find . -name "*" -mtime +10|while read a; do rm $a;done
Thanks
Dinesh Chandra
Rudy,
You are attempting to remove files from the Concurrent Manager
directory which can
be done programmatically using
an existing AOL program.
Log into the Oracle Application as the System Administrator.
Go into the Concurrent Requests form and submit a request for "Purge
Concurrent
Request and/or Manager Data".
When it asks for the parameters, enter "ALL" for the entity, "AGE" for
the mode,
and "10" for the mode value.
This will force the job to purge all data related to any concurrent
program from
the system and from the concurrent process
database tables.
Also, you will want to read the portions of the Oracle manuals that
deal with
Managing the Concurrent Manager (sorry I
don't recall what manual it is found in).
Greg Hoover
System Engineer
American Trans Air
From: Dan Olson dolson@dacotah.com
Subject: Re: removing files in APPL_TOP/common
I schedule the Purge request to delete once a week after a appl_top
backup. But I do notice that sometimes CM purge will miss some files,
so a do a manual Unix rm, once a month.
Subject: Shell Script for FTP and file management.
Date: Thu, 8 Apr 1999 18:22:50 -0400 (EDT)
Hi Oracle Gurus:
I would appreciate your thoughts on the following task I am trying
to
achieve:
I am trying to write a shell script to automate ftp process from our
server to another (remote) server. What I am trying to do is that
once
a file has been moved to the OUT directory, my shell script should
add
a date and time stamp to it, FTP it to the remote server and then
move
it to the ARCHIVE directory. Similarly on the IN directory, I would
like to move a file to another directory, depending on the name of
the
file (PO or INV subdirectory), rename it (give it a fixed name
depending on the 'Infile' name in SQL*Loader script), and then give
it
a date and time stamp before moving it to an ARCHIVE directory.
First of all, please let me know if it makes any sense at all.
Your comments, piece of advice, sample code, etc. are very welcome.
Thanks in advance.
Mohamed
From: "stephen bryant" sbryant@hotmail.com
Subject: Re: Shell Script for FTP and file management.
I have used korn shell scripts for the following FTP process:
ftp -n IP ADDRESS endFTP
user USER_ID PASSWORD
ascii
put LOCAL DIRECTORY FILE NAME REMOTE DIRECTORY FILE NAME
endFTP
Before and After the FTP process, you can use cp (copy) and/or mv
(rename) commands with date and time stamps.
Any questions, please email me.
Steve
From: "Sankala, Dinesh" Dinesh.Sankala@lgeenergy.com
To: "'oraapps-l@cpa.qc.ca'" oraapps-l@cpa.qc.ca
Subject: RE: printer problem
hi
WE HAVE A TYPICAL REQUIREMENT HERE
WE WANNA MONITOR THE PRINT QUEAE AT unix level
like
is there anyway to know that if any programme hangs in the print que at
Unix level
we need to find that by an alert or by a executing a pl/sql script or ....
any ideas greatly appreciated
thanks in advance
kumar
From: "Steve Spencer" srs@freezone.co.uk
To: oraapps-l@cpa.qc.ca
Subject: RE: printer problem
You can look at the UNIX print queue with unix commands such as lpstat...ask
your DBAs.
If you mean you want to monitor the Concurrent Requests queue from a Unix
level ie outside of the application, then do as you suggest. Write a simple
SQL*Plus script to query the FND_CONCURRENT_REQUESTS and associated linked
tables for those records that have the particular status flag that you
require (Pending or whatever). See your AOL Tech Ref manual for details of
these tables and their flags.
Then get your SQL*Plus script run from inside a unix shell script and away
you go...
Cheers,
Steve
Reply-To: oraapps-l@cpa.qc.ca
To: Multiple recipients of list oraapps-l@cpa.qc.ca
Subject: Help from UNIX gurus
Date: Wed, 29 Sep 1999 02:57:39 -0400 (EDT)
Hello
I made a copy of the man document to text file in unix.I want to
remove the control charecters (^H).How to do it with vi text editor.Can
anybody help me
Thanks in advance
Regards,
Venkat - DBA
PT INDO-RAMA SYNTHETICS Tbk
Purwakarta-41101 Indonesia
Phone 62-264-202311 EXT 2526
Reply To: oraapps-l@cpa.qc.ca
Sent: Wednesday, September 29, 1999 6:44 AM
To: Multiple recipients of list
Subject: Re: Help from UNIX gurus
Hi Venkat,
Open your text file and type in ":1,$ s/^H/ /g" obviously without the
double quotes. To get the control character press control v and then
control
H. All the control characters will be replaced with space at once since a
global command is specified.
Hope this will help you.
Disha.
From: "Annamaneni, Hanumantha" Hanumantha.Annamaneni@alcoa.com
To: "'oraapps-l@cpa.qc.ca'" oraapps-l@cpa.qc.ca
Subject: RE: Help from UNIX gurus
Istead, you could use:
man command | ul -tdumb filename
and frint the file
Hanumanth