Recent content by macd68

Please post the final result here or at the Tcler's wiki when you get done. This could be useful for a lot of folks and I haven't googled similar or seen a windows example at the Tcler's wiki.

Semmel, Expect is a recent addition to windows thanks to the very hard work of the Tcl Core group. I'd suggest that you comment out the chmod +x based on an env check for platform type. Please see the documentation for your proposed platform for what tcl's native 'file attributes' supports...

Expect. My advice is to tell him the assignment exceeds your capacity at the current moment and encourage an expert (consultant or otherwise) to come on board while you learn what you need from them.

The bottom line here is that selinux is not a good mandatory addition. I really wish main line distributions would take it out of production. Sysadmins can decide where and when to include it.

On second thought you could use a chained hashtable for this, but the buckets would get big quickly. Quick example: #include <stdio.h> #include <stdlib.h> #include <string.h> #define IMEMBERS 200 #define HASHLEN 211 struct hash_el_ { int pos[IMEMBERS]; struct hash_el_ *nxt; }...

As the previous poster mentioned you need to thoroughly study your application logs. mysql, apache or your milter software. Sar as mentioned is a big help. You should really be running snmpd and trapping on load and application alerts.

Example code. This code is untested but the approach should work fine for you. #!/usr/bin/tclsh package require Expect set timeout 20 set prompt1 ".*asswor.*" set prompt2 ".*>." #customize the above to your system array set optarray {} proc getopt {str lst} { global optarray set cnt 0...

Try a newline...and make sure that the command is received.

I can help you with the expect portion of this but I'm not much of a ksh guy. Have you taken a look at the comp.lang.tcl archives? There may already be an answer to your issue.

Your problem is a design and philosophy problem, not a technical issue.

The idea behind a 'dhcp table that changes all the time' is ridiculous. The dhcp service provides a lease time and many other configurable options and the dhcp client also keeps state. If you continually receive different addresses for the same host then you should check your dhcp service...

This isn't a good question. The linux kernel obviously supports SMP (since 2.2). If a distribution ships a broken kernel than you can always grab source and compile your own.

Are you logging and debugging (man ntpd)? Try this first before asking for help. That's what anyone here would do to fix the issue.

Make sure that lvm support is loaded. Check the output of lsmod. Disregard obviously if other lvm filesystems are already mounted.

ssh -v(vv) is your friend. Try this and running the sshd daemon on the remote host in debug mode also. HTH.