Search Results

Having preview on every thread makes the forum display pretty cluttered, imo.

So, you could extract the allowed groups from sshd.conf, and query LDAP to get the users in those groups.

On our systems this would be something like:
for group in $(awk '/AllowGroups/...

Without knowing how access is controlled on the hosts, there's no way to say what you need to search for...

Is t1 always going to be after p1 & g1? If so, you could do something like:

$ awk '$1~/"g1"/ {g=$3} $1~/"p1"/ {p=$3} /"t1" : "Reading C \(bytes\)",/ {printf "%s %s %s\n", g, p, (p < g ? "OK" :...

-S none /web/Transfer_Files/data/
-S doesn't take an argument (at least on Ubuntu Linux), so I'd guess it's interpreting none as a source (relative to cwd, which is presumably /web/Transfer_files)

df -Bg -xtmpfs -xdevtmpfs! Getting a tad unwieldy, but I guess you could alias it.

df -Bg should work

Shouldn't df -P (or df -hP) force one line per filesystem?

Using ftp?

Why do you want to do them 10 at a time? Bandwidth issues, or something else?

It looks ok, except LC_ALL=C date --utc --iso-8601=seconds -d"$begindatum"+1minute returns something that will never match einddatum (at least, on my machine).

Run it with bash -x script.sh and...

Counting the number of lines isn't much of a check that the files are really the same - use an actual checksum (e.g. cksum or md5sum).

Something like:

scp -i $HOME/.ssh/id_dsa...

As Don said, there's no way to guess the year (unless you know the date is within a specific time window e.g. no more than 300 days in the past).

There are some date scripts in the FAQ and Tips...

Mark threads "solved" by adding a "solved" tag to the thread tags (above); and always take a few seconds to properly tag all threads with appropriate technical terms and keywords.

If you have GNU awk and assuming your input is just a simple csv (i.e. no ',' characters within a field), try something like:

$ awk '{$3=$5=""}1' FS=, OFS=, file.csv
Col1,col2,,col4,,col100...

I don't think awk supports indirect referencing for variables.

In this case it looks like you could just use an array anyway.

You could try using file and checking if each file found appears to be a cert store of some kind, then using openssl to actually read it.

Cron may be setting a default PATH.

e.g. from Linux 2.6 man pages (https://www.unix.com/man-page/all/5/crontab/)


(I've no idea if that's the case on Raspberry though. Or on reboot, for that...

You've left in the esac for the case statement which you've deleted.

-vCOMP='20150430'

You can use SetGID, but SetUID is ignored for directories on must Unix & Linux systems, as far as I know. (Except BSD, apparently. It definitely doesn't work on AIX.)

Can't you run the script as...

Which is it? Do you want the attribute value, or the tag value where the nil attribute is true?

Have you tried greping for NVV in $HOME/.* and /etc/*?

What have you tried?
awk 'FNR==8 || FNR==10 || (FNR>=13 && FNR<=250)' sample.txt
should work.

The first place too look would be the man pages. System calls are in section 2 - e.g. man 2 alarm (or look on this web site - https://www.unix.com/man-page/linux/2/alarm/).

You could use the array indices in the for loop:
for i in ${!arrVERSION }
do
echo "${arrVERSION[$i]}"
done