Sponsored Content
Full Discussion: Archivelog monitoring Script
Top Forums Shell Programming and Scripting Archivelog monitoring Script Post 302310087 by hkansal on Thursday 23rd of April 2009 05:35:04 PM
Old 04-23-2009
Bug Logic only

Hello,

I assume that you are able to get the o/p in a plain file. Here are the steps that you should follow:
  1. Initiate a counter from 1
  2. Loop through the file line-wise
    1. Operate on current line according to value of counter
    2. Increment counter; reset to 1 if value is 6

Actions according to value of counter:
1 this is ur database
2 this is ur filename
3 take col 4 and 5 - use awk or cut
4 same as counter value 3
5 this is the status (I would take "Successful"/"failed" only)

at each step echo these values and use tabs to separate them, eg:
Code:
echo "\t<value>"

use \c to keep output on same line.

adjust the number of tabs as u need, u'll need to tweak a bit Smilie

Regards,
HKansal
 

10 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

monitoring script

hi can someone help me to write a script to monitor the growth of the directory running at 5 hours interval and then pipe it to a file :(? i only know the manual command "du -sk" and the worst i dun know anything about script. (3 Replies)
Discussion started by: legato
3 Replies

2. UNIX for Advanced & Expert Users

monitoring script

i have a unix batch written in c, dealing with really time-consuming database operations i want to write a ksh script to monitor its performance. which items i should monitor? do you have any suggestions? 1)cputime 2)swap area 3)pmem 4) 5) what else? (1 Reply)
Discussion started by: gfhgfnhhn
1 Replies

3. Solaris

Monitoring Script

Hi, I want to write a script that will monitor cpu,mem usage and disk usage for entire day and data will be redirected to file, (4 Replies)
Discussion started by: manoj.solaris
4 Replies

4. Shell Programming and Scripting

IP Monitoring Script

Dear All, I am new to Shell Programming, but I need a script which will monitor the different IP's in the live logs. Condition:- The script will monitor the logs continously and should print only those IP's which exists more than 3 times per day. Please do help me. Thanks and... (1 Reply)
Discussion started by: akhtar.bhat
1 Replies

5. Shell Programming and Scripting

Archivelog monitoring Script

Could anyone please help? I have written a small program that's actually working fine for me and extracting all the details I required. What code does is, it goes to all archivelog directories and see if archivelog backup was failed or successful <<CODE>> ... (1 Reply)
Discussion started by: anjum.suri
1 Replies

6. Shell Programming and Scripting

Help with Monitoring script

Hi Gurus, Currently I am learning UNIX through online forums and unix blogs. I have the below requirement. I need to write a script to monitor server processes. For example, there are 3 processes currently running on the server.(java, pmrepagent, pmserver). If any of the process goes down,... (2 Replies)
Discussion started by: svajhala
2 Replies

7. Shell Programming and Scripting

Monitoring script

Hi, I want to write script that monitors particular ports in a server. I completed the script but.... If the server is restarted i need manually start the script.. Is there any way i can make the script start by it self after the server reboot........ Thanks, Firestar (4 Replies)
Discussion started by: firestar
4 Replies

8. Red Hat

monitoring script

Hi, I ned to monitor the memory for the red-hat machine,when its reaches a threshold value like 2GB then a mail should be sent as soon as it finds the the value is met can you suggest me which is the best way to do this? (2 Replies)
Discussion started by: sriniv666
2 Replies

9. UNIX for Advanced & Expert Users

ldapsearch in monitoring script without bind password written in script

Hi I do a very simple monitoring of our OpenLDAP (runs in cronjob and generate alerts if unsuccessfull) $ ldapsearch -h hostname.domain -D "cn=monitor_user,ou=People,dc=organisation" -w "password" -b "dc=organisation" -x "(&(cn=monitor_user)(ou=People))" dn | grep -v version dn:... (4 Replies)
Discussion started by: slashdotweenie
4 Replies

10. Shell Programming and Scripting

Monitoring Script

Hello All, I have written a script which will grep for error codes in a file and if the count for the same is about 500 it will send and smpp alert. Here is my code. #!/bin/bash #########################################################################################################... (3 Replies)
Discussion started by: Siddheshk
3 Replies
counter(n)						      Counters and Histograms							counter(n)

__________________________________________________________________________________________________________________________________________________

NAME
counter - Procedures for counters and histograms SYNOPSIS
package require Tcl 8 package require counter ?2.0.4? ::counter::init tag args ::counter::count tag ?delta? ?instance? ::counter::start tag instance ::counter::stop tag instance ::counter::get tag args ::counter::exists tag ::counter::names ::counter::histHtmlDisplay tag args ::counter::reset tag args _________________________________________________________________ DESCRIPTION
The counter package provides a counter facility and can compute statistics and histograms over the collected data. ::counter::init tag args This defines a counter with the name tag. The args determines the characteristics of the counter. The args are -group name Keep a grouped counter where the name of the histogram bucket is passed into ::counter::count. -hist bucketsize Accumulate the counter into histogram buckets of size bucketsize. For example, if the samples are millisecond time values and bucketsize is 10, then each histogram bucket represents time values of 0 to 10 msec, 10 to 20 msec, 20 to 30 msec, and so on. -hist2x bucketsize Accumulate the statistic into histogram buckets. The size of the first bucket is bucketsize, each other bucket holds values 2 times the size of the previous bucket. For example, if bucketsize is 10, then each histogram bucket represents time values of 0 to 10 msec, 10 to 20 msec, 20 to 40 msec, 40 to 80 msec, and so on. -hist10x bucketsize Accumulate the statistic into histogram buckets. The size of the first bucket is bucketsize, each other bucket holds values 10 times the size of the previous bucket. For example, if bucketsize is 10, then each histogram bucket represents time val- ues of 0 to 10 msec, 10 to 100 msec, 100 to 1000 msec, and so on. -lastn N Save the last N values of the counter to maintain a "running average" over the last N values. -timehist secsPerMinute Keep a time-based histogram. The counter is summed into a histogram bucket based on the current time. There are 60 per- minute buckets that have a size determined by secsPerMinute, which is normally 60, but for testing purposes can be less. Every "hour" (i.e., 60 "minutes") the contents of the per-minute buckets are summed into the next hourly bucket. Every 24 "hours" the contents of the per-hour buckets are summed into the next daily bucket. The counter package keeps all time-based histograms in sync, so the first secsPerMinute value seen by the package is used for all subsequent time-based histograms. ::counter::count tag ?delta? ?instance? Increment the counter identified by tag. The default increment is 1, although you can increment by any value, integer or real, by specifying delta. You must declare each counter with ::counter::init to define the characteristics of counter before you start to use it. If the counter type is -group, then the counter identified by instance is incremented. ::counter::start tag instance Record the starting time of an interval. The tag is the name of the counter defined as a -hist value-based histogram. The instance is used to distinguish this interval from any other intervals that might be overlapping this one. ::counter::stop tag instance Record the ending time of an interval. The delta time since the corresponding ::counter::start call for instance is recorded in the histogram identified by tag. ::counter::get tag args Return statistics about a counter identified by tag. The args determine what value to return: -total Return the total value of the counter. This is the default if args is not specified. -totalVar Return the name of the total variable. Useful for specifying with -textvariable in a Tk widget. -N Return the number of samples accumulated into the counter. -avg Return the average of samples accumulated into the counter. -avgn Return the average over the last N samples taken. The N value is set in the ::counter::init call. -hist bucket If bucket is specified, then the value in that bucket of the histogram is returned. Otherwise the complete histogram is returned in array get format sorted by bucket. -histVar Return the name of the histogram array variable. -histHour Return the complete hourly histogram in array get format sorted by bucket. -histHourVar Return the name of the hourly histogram array variable. -histDay Return the complete daily histogram in array get format sorted by bucket. -histDayVar Return the name of the daily histogram array variable. -resetDate Return the clock seconds value recorded when the counter was last reset. -all Return an array get of the array used to store the counter. This includes the total, the number of samples (N), and any type-specific information. This does not include the histogram array. ::counter::exists tag Returns 1 if the counter is defined. ::counter::names Returns a list of all counters defined. ::counter::histHtmlDisplay tag args Generate HTML to display a histogram for a counter. The args control the format of the display. They are: -title string Label to display above bar chart -unit unit Specify minutes, hours, or days for the time-base histograms. For value-based histograms, the unit is used in the title. -images url URL of /images directory. -gif filename Image for normal histogram bars. The filename is relative to the -images directory. -ongif filename Image for the active histogram bar. The filename is relative to the -images directory. -max N Maximum number of value-based buckets to display. -height N Pixel height of the highest bar. -width N Pixel width of each bar. -skip N Buckets to skip when labeling value-based histograms. -format string Format used to display labels of buckets. -text boolean If 1, a text version of the histogram is dumped, otherwise a graphical one is generated. ::counter::reset tag args Resets the counter with the name tag to an initial state. The args determine the new characteristics of the counter. They have the same meaning as described for ::counter::init. BUGS, IDEAS, FEEDBACK This document, and the package it describes, will undoubtedly contain bugs and other problems. Please report such in the category counter of the Tcllib SF Trackers [http://sourceforge.net/tracker/?group_id=12883]. Please also report any ideas for enhancements you may have for either package and/or documentation. KEYWORDS
counting, histogram, statistics, tallying CATEGORY
Data structures counter 2.0.4 counter(n)
All times are GMT -4. The time now is 03:42 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy