I'm currently using this script line to find all the log files from a given directory structure and copy them to another directy where I can easily compress them.

find . -name "*.log" -exec cp \{\} /tmp/allLogs/ \;

The problem I have, is, the directory/subdirectory information gets lost because, I'm copying only the file.

For instance I have:


And I end up with:


And I would like to have:


4 Answers 4


What is the reason to copy them to another directory to compress them? The following will create a compressed tar file off all the log files while keeping the directory structure in one step (Assuming it is run from the root directory:

find . -iname '*.log' -print0 | xargs -0 tar zcf /tmp/test.tar.gz

For example:

kbrandt@alpine:~/scrap/tar$ find . *.log
kbrandt@alpine:~/scrap/tar$ find . -iname '*.log' -print0 | xargs -0 tar zcf /tmp/test.tar.gz
#List files in the archive with the -t switch
kbrandt@alpine:~/scrap/tar$ tar -tzf /tmp/test.tar.gz 
  • The reason is ... ignorance, definitely :) .) Thank you Kyle
    – OscarRyz
    Jan 3, 2011 at 18:51

Try using cpio in pass-through mode

find . -name '*.log' | cpio -pdm /tmp/allLogs

find . -name "*.log" | tar -cz --files-from - -f /path/to/file.tgz

What about filtering just the logs into the archive:

sudo tar -zcvf archive.tar.gz /path/**/*.log

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.