Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 5456
  • Last Modified:

How to truncate log files in solaris

Hello,

I'm wondering how to truncate the log files (like /var/adm/messages or utmpx) in solaris? Is there a way to keep the log file size to a particular size limit?

Thanks,
Ashok
0
rdashokraj
Asked:
rdashokraj
  • 2
  • 2
  • 2
  • +3
2 Solutions
 
TintinCommented:
man logadm

and

cat /etc/logadm.conf
0
 
arthurjbCommented:
If it is a log file you don't care about and never need it back, here is the simple way;

> name-of-file.log

(Yes it is just a greater than sign, the logfile name, and press return.)

It is important that you do not just delete the file, since the logging programming could continue to write in a way that you cannot easily see the file that the info is going to, and it will eventually fill up the disk...
0
 
Brian UtterbackPrinciple Software EngineerCommented:
The problem with truncating the file with the redirect is that not all programs have logfiles open with
append. It is not uncommon that they just open it for writing and that the file pointer is left as is after
each write. Thus, after you truncate the file, the next write will write to the same position it would have
before, creating a sparse file with a huge number of nulls at the beginning.
0
Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
vikaskhoriaCommented:
You can do this I think:
From the program which is writing the logs, where ever it is writing to the log, first check the no. of lines it has in it. And if the number of lines is greater than a given value of maximum lines, replace that file else append to that file.

For example if your logfile is LOG.TXT then we can write something like this before writing to the file:

lines=`wc -l LOG.TXT | awk '{ print $1 }'`
if [[ $ lines -gt $maxLines ]]
then
 echo "Lines for log" > LOG.TXT
else
 echo "Lines for log" >> LOG.TXT
fi

And if you don't have access to the file generating the output (means you cannot change that) then you can write a script which will periodically check the size and reset the file whenever required:
lines=`wc -l LOG.TXT | awk '{ print $1 }'`
if [[ $ lines -gt $maxLines ]]
then
echo > LOG.TXT
fi

0
 
vikaskhoriaCommented:
Also do you don't want to change the file to zero size everytime??
0
 
TintinCommented:
UUOA ;-)

lines=`wc -l LOG.TXT | awk '{ print $1 }'`

lines=`wc -l <LOG.TXT`

0
 
TalmashCommented:
split -l 10000 large_file <output_file_new_extension>

example:
abc has 20005 lines,
xaa will hold lines 1-10000
xab will hold lines 10001-20000
xac .. 20001-20005
0
 
rdashokrajAuthor Commented:
Thanks for all your solutions. However the solution given by Tintin seems to be appropriate one and it satisfies my requirement. Thanking you all once again.
0
 
rdashokrajAuthor Commented:
Thanks for all your solutions. However the solution given by Tintin seems to be appropriate one and it satisfies my requirement. Thanking you all once again.
0

Featured Post

Hire Technology Freelancers with Gigs

Work with freelancers specializing in everything from database administration to programming, who have proven themselves as experts in their field. Hire the best, collaborate easily, pay securely, and get projects done right.

  • 2
  • 2
  • 2
  • +3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now