Quantcast
Channel: All System Administration posts
Viewing all articles
Browse latest Browse all 1179

Re: tar taking 2 hrs for filesystem with deep directory structure

$
0
0

> We need  /backup of 6.7GB to be copied to /temp_backup on same VM
> [...]

   Do you want a giant "tar" file at the destination, or do you want a
file tree like the source file tree?

> cd /backup
> tar -cvf /temp_backup/backup.tar  *

   That will give you the giant "tar" file.  Again, I'd omit "v".  Also,
"*" will not catch "hidden" files (".login", and so on).  If I wanted a
giant "tar" file, I'd do:

      ( cd /backup ; tar cf /temp_backup/backup.tar . )

   If the bottleneck in your scheme is the disk-writing speed, and your
CPU is speedy. then you might save some time by compressing the data.
For example:

      ( cd /backup ; tar cf - . | gzip -c > /temp_backup/backup.tgz )

 

   On a modern GNU/Linx system, GNU "tar" probably has built-in
compression options, but using an explicit compression program in the
pipeline (as shown above) works on systems with older and/or less
capable "tar" programs.

      http://www.gnu.org/software/tar/manual/html_node/gzip.html#SEC134

> Tar pipeline ? Can you please share the same

   Your Forum/Web search found nothing useful?  One example (of many):

      http://h30499.www3.hp.com/t5/x/x/m-p/3966579

   Substituting "&&" for ";" in those example pipelines can improve
behavior when errors occur.


Viewing all articles
Browse latest Browse all 1179


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>