<div>
try<br />
&nbsp;<br />
cat /var/www/vhosts/*/statisti<wbr />cs/logs/er<wbr />ror*log | sed '/^$/d' &nbsp;| tail -n 1 &nbsp;&gt; /var/www/vhosts/summary.lo<wbr />g/error-lo<wbr />g.txt
</div>


<div>
That didn't quite it. It seems like it combined all the files and then just gave me the last line of that combination. Basically, I don't want to display the files that don't show anything. Here is a sample result from tail -n 1<br />
<br />
==&gt; /var/www/vhosts/site1.com/<wbr />statistics<wbr />/logs/erro<wbr />r_log &lt;==<br />
[client 50.50.50.50] PHP Warning: Invalid argument supplied for foreach() in /var/www/vhosts/site1.com/<wbr />httpdocs/i<wbr />ndex.php on line 217<br />
<br />
==&gt; /var/www/vhosts/client1025<wbr />.k-a-p.com<wbr />/statistic<wbr />s/logs/err<wbr />or_log &lt;==<br />
<br />
==&gt; /var/www/vhosts/site2.com/<wbr />statistics<wbr />/logs/erro<wbr />r_log &lt;==<br />
[Wed Feb 25 21:13:44 2009] [error] [client 165.165.165.165] File does not exist: /var/www/vhosts/site2.com/<wbr />httpdocs/q<wbr />uark
</div>


<div>
#!/bin/bash<br />
for f in /var/www/vhosts/*/statisti<wbr />cs/logs/er<wbr />ror*log ; do<br />
&nbsp; grep $f | tail -n 1<br />
done &gt;&nbsp;/var/www/vhosts/summary.lo<wbr />g/error-lo<wbr />g.txt<br />

</div>


<div>
#!/bin/bash<br />
for f in /var/www/vhosts/*/statisti<wbr />cs/logs/er<wbr />ror*log ; do<br />
&nbsp; grep . $f | tail -n 1<br />
done &gt;&nbsp;/var/www/vhosts/summary.lo<wbr />g/error-lo<wbr />g.txt
</div>


<div>
then run a loop on them:<br />
<br />
ls /var/www/vhosts/*/statisti<wbr />cs/logs/er<wbr />ror*log | while read file<br />
do<br />
&nbsp; &nbsp;cat $file | sed '/^$/d' &nbsp;| tail -n 1 &nbsp;&gt;&gt; /var/www/vhosts/summary.lo<wbr />g/error-lo<wbr />g.txt<br />
done
</div>


<div>
Another approach<br />
<br />
find /var/www/vhosts -name &quot;error*log&quot; -size +1c | xargs tail -n1 &gt;/var/www/vhosts/summary.l<wbr />og/error-l<wbr />og.txt
</div>


<div>
will<br />
<br />
xargs tail -n1 &gt;/var/www/vhosts/summary.l<wbr />og/error-l<wbr />og.txt<br />
<br />
overwrite the log file or it will be accumulative (concatenated) ?
</div>


<div>
Tintin's solution worked perfectly. The 'find' seems to be a bit slow, I am not sure how to speed that up. It does seem to overwrite the log file, which is correct.
</div>


<div>
Awesome, it is now super fast. That is great. Thanks so much.
</div>


<div>
Did you want to display the last non-blank line of all files, or the last line of all files that are not empty (even if they contain blank lines)?
</div>


<div>
The later. I wanted to display the last line of all non-empty files. Tintin allowed tail's verbose mode (automatic when there are multiple files) to nicely display the result. So my example above became:<br />
<br />
==&gt; /var/www/vhosts/site1.com/<wbr />statistics<wbr />/logs/erro<wbr />r_log &lt;==<br />
[client 50.50.50.50] PHP Warning: Invalid argument supplied for foreach() in /var/www/vhosts/site1.com/<wbr />httpdocs/i<wbr />ndex.php on line 217<br />
<br />
==&gt; /var/www/vhosts/site2.com/<wbr />statistics<wbr />/logs/erro<wbr />r_log &lt;==<br />
[Wed Feb 25 21:13:44 2009] [error] [client 165.165.165.165] File does not exist: /var/www/vhosts/site2.com/<wbr />httpdocs/q<wbr />uark
</div>


<div>
<div class="content wysiwyg-content">
I am currently using the follow command as part of a cron job.<br />
<br />
tail -n 1 /var/www/vhosts/*/statisti<wbr />cs/logs/er<wbr />ror*log &gt;&nbsp;/var/www/vhosts/summary.lo<wbr />g/error-lo<wbr />g.txt<br />
<br />
This works great, but when some of the error*log files are empty, I get empty lines from tail. Is it possible to have tail ignore empty files?
</div>
</div>


I am currently using the follow command as part of a cron job.

tail -n 1 /var/www/vhosts/*/statistics/logs/error*log > /var/www/vhosts/summary.log/error-log.txt

This works great, but when some of the error*log files are empty, I get empty lines from tail. Is it possible to have tail ignore empty files?

Setting tail -n 1 to ignore blank files

Linux is a UNIX-like open source operating system with hundreds of distinct distributions, including: Fedora, openSUSE, Ubuntu, Debian, Slackware, Gentoo, CentOS, and Arch Linux. Linux is generally associated with web and database servers, but has become popular in many niche industries and applications.