04-03-2013
Don't do it in real time. Don't use tail -f which seems to me the problem.
Once an hour, or whatever, use tail to save the end part of the log file that has not been examined before. Save the end part to "tail.txt" for example. Use grep -C on tail.txt to get the context for each error line detected. Record total number of lines in the log file. Suppose there are 1000. Next time (in an hour), tail starts with line 1001 (or maybe 991 to make sure you have context).
9 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
as the title, I had try use "wc -l test.txt" but it give me "<many spaces> 384 test.txt" but the result I want is just "384" could any person can help me that?
Thx:( (5 Replies)
Discussion started by: a8111978
5 Replies
2. Shell Programming and Scripting
If I have a file something like as shown below,
ARM*187878*hjhj
BAG*88778*jjjj
COD*7777*kkkk
BAG*87878*kjjhjk
DEF*65656*89989*khjkk
I need the line numbers to be added with a colon when it matches the string "BAG". Here in my case, I need something like
ARM*187878*hjhj... (4 Replies)
Discussion started by: Muthuraj K
4 Replies
3. Shell Programming and Scripting
Hi,
A very Good Evening to All,
I am writing a script for my application. I have a file with 1000 lines. Among that 1000 lines i am searching for a particular string. And from that string i need to pull all the data in to a seperate file.
For example the contents of my file is as below.
... (4 Replies)
Discussion started by: intiraju
4 Replies
4. Shell Programming and Scripting
Hi,
The below code will search a particular string(say false in this case) and return me 10 lines above and below the search string in a file.
" awk 'c-->0;$0~s{if(b)for(c=b+1;c>1;c--)print r;print("***********************************");print;c=a;}b{r=$
0}' b=10 a=10 s="false" "
... (5 Replies)
Discussion started by: vimalm22
5 Replies
5. Shell Programming and Scripting
Hi Guys,
I am trying to write a perl script to search a string "Name" in the file "FILE" and also want to create a new file and push the searched string Name line along with 10 lines following the same.
can anyone of you please let me know how to go about it ? (8 Replies)
Discussion started by: sukrish
8 Replies
6. Shell Programming and Scripting
i have one file, i am doing 'vi Filename' now i want to search for particular string and i want to know how many times that string occurs in whole file (5 Replies)
Discussion started by: sheelsadan
5 Replies
7. Shell Programming and Scripting
Hello,
I want to search 2 lines from the file. I wanted to display all such matches. Example file:
==================
Testfile is test
TEST1
TEST2
testing the file string
to do testing
TEST1
TEST2
sample strings
=================
I wanted to search the file with 2 lines
"
TEST1... (3 Replies)
Discussion started by: balareddy
3 Replies
8. Shell Programming and Scripting
i had requirement like i need to get "error" line of above 3 and below 3 from a file .I tried with the below script.But it's not working.
y='grep -n -i error /home/file.txt|cut -c1'
echo $y
head -$y /home/file.txt| tail -3 >tmp.txt
tail -$y /home/file.txt head -3 >>tmp.txt (4 Replies)
Discussion started by: bhas85
4 Replies
9. Shell Programming and Scripting
Hello
Take this file...
Test01
Ref test
Version 01
Test02
Ref test
Version 02
Test66
Ref test
Version 66
Test99
Ref test
Version 99
I want to substitute every occurrence of Test{2} with a unique random number, so for example, if I was using sed, substitution would be something... (1 Reply)
Discussion started by: funkman
1 Replies
LEARN ABOUT DEBIAN
logtop
LOGTOP(1) General Commands Manual LOGTOP(1)
NAME
logtop - Realtime log line rate analyser
SYNOPSIS
logtop [OPTIONS]
DESCRIPTION
logtop is a System Administrator tool analyzing line rate on stdin.
It reads on stdin and print a constantly updated result
displaying, in columns:
Line number, count, frequency, and the actual line.
$ tail -f FILE | logtop
is the friendly version of:
$ watch 'tail FILE | sort | uniq -c | sort -gr'
OPTIONS
-s, --size=K
Only keep K lines in memory, instead of 10000.
-q, --quiet
Do not display a live view of the data, only display a top at exit.
-l, --line-by-line=K
Print result line by line, in a machine friendly format, K is the number of result to print per line.
Line by line format is : [%d %f %s ]*
%d : Number of occurences
%f : Frequency of apparition
%s : String (Control chars replaced by dots.
-i, --interval=K
Interval between graphical updates, in seconds. Defaults to 1.
-h, --help
Show summary of options.
-v, --version
Show version of program.
EXAMPLES
Here are some logtop usage examples.
tail -f cache.log | grep -o "HIT|MISS" | logtop
Realtime hit / miss ratio on some caching software log file.
tail -f access.log | cut -d' ' -f1 | logtop -s 10000
Realtime most querying IPs on your server, as long as log lines in access.log starts with the client IP.
tail -f access.log | cut -d' ' -f7 | logtop -s 10000
Realtime most requested web pages in a NCSA like log file.
cat auth.log | grep -v "CRON" | grep -o ": .*" | logtop -q -s 100000
Display a one-shot simple analyse of your auth.log.
SEE ALSO
watch(1)
AUTHOR
logtop was written by Julien Palard.
This manual page was written by Julien Palard <julien@palard.fr>, for the Debian project (and may be used by others).
April 16, 2011 LOGTOP(1)