Search results for query: *

...<node second node> <hook>hooky AIX thing </hook> </node> <node third node> <hook>hooky solaris thing </hook> </node>'; $string2 =~ s/\<node.*?Solaris.*?\<\/node\>/replacement /msi; print $string2; Now, according to me this should just take out the third solaris node but the result is...

If there's a mapping between the file group name and the directory it ends up in then use the suid bit on the directory. For example mkdir -m2777 test_dir chown user:test_grp test_dir Any files created in test_dir will have test_grp group ownership. On the internet no one knows you're a dog...

So, I'm working with data lines which end with a return code and an optional non numeric message code. What I have so far is my @bits = split /\s+/; if ( $bits[-1] =~ /^\d+$/ ) { $retcode = $bits[-1]; } else { $retcode = $bits[-2]; } No problem with that, it works but why can't I use my...

...values in entry fields. Press Enter AFTER making all desired changes. [TOP] [Entry Fields] * Installation Target b05301 * Installation TYPE mksysb * SPOT...

LKBrwnDBA I would love to go there directly but firewall rules, yada, yada... and the double jump is my only option. Your code works perfectly Thanks On the internet no one knows you're a dog Columb Healy

I'm writing a procedure in bash which copies files around a number of servers. As a basic check I'm runningssh remote_host 'cd /remote/dir;ls | wc -l' So far all well and good. However some of the servers are only reachable via a double jump. i.e. to run a comand you have to runssh remote1 ssh...

Gotcha - It's sort of First line - store in prev Subsequent lines - print prev then store new line in prev - i.e. print the line before. When completed print reformatted final line Thanks Feherke - you're a star as ever. On the internet no one knows you're a dog Columb Healy

I have a perl script I would like to migrate to awk. It's quite simple - it copies a file and amends the footer to reflect the number of records (line count so far minus one for the header). In perl it looks like#!/usr/bin/perl -w use strict; $#ARGV == 2 or die "Invalid agument count\n"; open...

We've been told to wipe disks to UK DOD standard 5220.22m. Does anyone know of any products we might use on an SP system running AIX 4.3 I keep saying that half an hour with my Black and Decker would be far more effective but you know what security types are like ;-) Thanks On the internet no...

Thanks ZaSter - have a star. There's a typo in your link - it's here but I'd got close enough with this list of up2date <-> yum equivalences - which does not have the answer so using your doc number enabled me to find it. On the internet no one knows you're a dog Columb Healy

command | less On the internet no one knows you're a dog Columb Healy

On my old systems I used to run up2date --update --downloadto retrieve the files andup2date --update to apply the patches. Now I've inherited a RHEL 5 server with yum instead of up2date. I've googled all I can but I can't find the equivalent of up2date --update --download for yum. Any pointers...

You can't do this using ssh in a raw state. There are two alternatives Use public/private key pairs - this negates the needs for passwords - including the risks of having passwords in clear text in scripts! Google "ssh keys" for many sites which will help Use 'expect' - this is a useful utiliy...

...structure. Note that the -exec parameter to find opens a new process for each file found - as does your for loop so, for functions that take multiple files the options are find . -name "*.exe" | xargs do_somethingand do_something *.exe On the internet no one knows you're a dog Columb Healy

Why bother excluding - compress will spot the compressed files and skip them. You will get lots of warning messages but, if you're confident the rest works find dir1 dir2 dir3 -type f -mtime +2|xargs compress 2>/dev/null On the internet no one knows you're a dog Columb Healy

Maybe you should buy your sysadmin a beer and suggest he allows you to use sudo to grep out the entries from sulog. On the internet no one knows you're a dog Columb Healy

...Tuesday ;; #End your case with a double semi colon Thursday) #Do Thursday specific stuff here process_data Thursday ;; *) #Default case - non Tuesday and Thursday stuff if required ;; esac #End of case statement. On the internet no one knows you're a dog Columb Healy

Sorry - addendum The /var/adm/cron/log file exists on AIX but not on RedHat Linux so it looks like this method is dependant on your flavour of Unix/Linux On the internet no one knows you're a dog Columb Healy

Try grep -q Failed /var/adm/cron/log && echo "A cron job has failed" | mail -s "cron job failure" me@myserver Of course you would have to reset the log file after every failure and the script could be nicer but you get the idea. On the internet no one knows you're a dog Columb Healy

Put a wrapper script around it:- #!/bin/ksh /path/to/my/big/script | tee $LOGFILE On the internet no one knows you're a dog Columb Healy