I am using awk and it stops when it encounter line greater then 3000 character. Is there any command which will help me remove line greater then 3000 characters.

You can write those lines which are less than or equal to 3000 characters.

awk ' length($0) <= 3000 ' inputfile

Here is a sed solution for the same

sed '/^.\{3000\}/d' inputfile

Here is what i am trying to do.

awk ' length($0) < 3000 ' access.050531.log |wc -l
awk: record `64.43.232.10 - - [31...' too long
record number 58134
58134

Still it gets to that line and stops parsing log file.

Thanks,
Naren

if you're on Solaris, try using either /usr/bin/nawk OR /usr/xpg4/bin/awk

for awk to be able to count and know that there are less than 3000 characters in the line --- it has to count the characters in the line ... problem is that awk will not stop counting at below 3000 characters if the line actually has more than that ...

maybe you don't want awk in this case ... perl or c might be better ...

How the about the sed solution ? Will that fail ?

sed '/^.\{3000\}/d' inputfile