only display df lines that has more fs usage then 80%

Question

root@SERVER ~$ df
Filesystem  512-blocks  Free    %Used   Iused   %Iused  Mounted on
/dev/YXCV   655360      365632  45% 6322    13% /
/dev/ASDF   3801088     670648  83% 41759   32% /usr
/dev/ASR    1048576     500496  53% 5555    9%  /var

how can I pipe the df commands output to only display lines that has more usage then 80%?

e.g.: it would only display:

/dev/ASDF   3801088     670648  83% 41759   32% /usr

Gilles · Accepted Answer · 2011-06-16 14:05:38Z

up vote 5 down vote accepted

Assuming you don't have device names containing spaces (which are a pain when it comes to parsing the output of df):

df -P | awk '+$5 >= 80 {print}'

Adapt the field number if you want to use your implementation's df output format rather than the POSIX format.

edited Jun 16 '11 at 14:05

answered Jun 16 '11 at 8:14

Gilles
140k13160367

df -P | awk '$5 >= 80 {print}' works for me: can you include a note about the benefit of using the substr? Thank you. – jasonwryan Jun 16 '11 at 9:41

@jasonwryan: I didn't remember offhand whether awk ignored trailing non-numeric characters when interpreting a string as a number. It does, so indeed there's no need to remove the trailing % explicitly. – Gilles Jun 16 '11 at 9:56

1

'$5 >= 80 works, ie. it ignores the trailing %, but < > == don't work in the same manner... whereas the (previous edit) subtring version seem to work for most comparison situations, but it too fails when comparing to < 100 ... I guess that is because it is comparing a sub-string.. – Peter.O Jun 16 '11 at 13:16

@fred: Oh, I see. Numeric conversion does ignore a non-numeric suffix, but the comparison operators work on strings, so here the operator doing a string comparison unless both strings are numeric. The best way is to force the string to a number (with the + operator). – Gilles Jun 16 '11 at 14:05

ddeimeke · Answer 2 · 2011-06-16 07:15:16Z

up vote 0 down vote

df | egrep "([89][0-9]|100)%"

will do the trick.

answered Jun 16 '11 at 7:15

ddeimeke
1,100210

2

and what happens when i have 92% free on a mount point? – LanceBaynes Jun 16 '11 at 7:22

1

It works fine for 92%; it works fine for 80 thru 100%... I'm quite puzzled about why someone marked it down.... maybe it is because my df output is different.... Aha! I just checked: my default output only shows Filesystem 1K-blocks Used Available Use% Mounted on ... so it just depends on what your output is (as mentioned in Gilles answer).. If there is a second % column (as in the question), a trailing .*% should work: df | egrep "([89][0-9]|100)%.*%" ...Now I understand why someone marked down... – Peter.O Jun 16 '11 at 14:51

asked	1 year ago
viewed	531 times
active	1 year ago

only display df lines that has more fs usage then 80%

2 Answers

Your Answer

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

only display df lines that has more fs usage then 80%

2 Answers

Your Answer

Sign up or log in

Post as a guest

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

Related