Before I start this guide, I need to point out that I'm a linux newbie myself, and I have very little experience with it. I have a passion for computers though, and learning new stuff is something I enjoy very much.
I made my Wireless Network Interface Card (Hereon NIC) work on a HP ZE2247EA laptop. I'm running Fedora Core 4, from a standard, out of the box, installation.
The following guide needs to be performed as root.
The first step, is to identify your NIC, and what chipset it is. When this is done, figure out what drivers WINDOWS would use to make it work. An easy way to do this, if you are running a dualboot, is to check the driver details in windows. When I did that on my computer, I found that my driver is called bcmwl5.inf.
My first attempt at using this driver turned out bad. I had downloaded a 64bit windows driver, but Fedora was kind enough to tell me that I needed a 32 bit, so my search for the correct driver continued. Finally stricking luck, at [/url]http://ndiswrapper.sourceforge.net/mediawiki/index.php/List , I downloaded the correct file and proceeded installing ndiswrapper.
In Fedora, this was a simple task. Something as easy as typing yum install ndiswrapper would be sufficient *if you have a wired internet connection*. Allthough, the kernel needs to support ndiswrapper aswell, so to implement this in the same command, simply type:
Code:
yum install ndiswrapper kernel-module-ndiswrapper
Voila, ndiswrapper is installed, and ready to go.
Next step, unzip the driver you downloaded to a directory you can find it.
I used /lib/windrivers/ . After unzipping the files, I typed out: "ndiswrapper -i bcmwl5.inf". The -i means to install the driver. Of course, bcmwl5.inf needs to be replaced with the name of your own file. To check if the installation and driverusage was successfull, type: "ndiswrapper -l". If this reports "bcmwl5 driver present, hardware present" you are ready for the next step. If it doesnt, make sure you have the correct driver, and that your NIC is active/plugged in.
Since you're still reading, I'm assuming you saw "xxxxxx driver present, hardware present" on the previous step. That means you're ready to write the configuration to modprobe. Type "ndiswrapper -m". This will add the hardware. If that works, type "modprobe ndiswrapper" and/or "ndiswrapper wlan0".
You should now be ready to go. If that doesn't work, you could try something as dramatical as rebooting your system, and that should most likely fix it.
After the reboot or modprobe is done, type "iwconfig". Hopefully, this will now display "wlan0", and not contain any network information.
Set your network SSID by typing "iwconfig wlan0 essid 'Your-Id-Here' [key 'Your key here']". (full details right [url=<___base_url___>/index.php?s=&showtopic=2098&view=findpost&p=7814]here,
Please note that the example here should look like this if your ACCESS POINT ssid=WIRELESS and wep key = aaaaa)
You might also want to set Mode to "managed" (iwconfig wlan0 mode "managed").
Natually, if you're connecting to a static network, you will need to set up ip adress and the likes aswell. I myself connect to a wireless router, and get my IP dynamically adressed by a DHCP server. So, a simple "dhclient wlan0", and I was up and running!
Posted by: rick - 2006-02-16, 10:04 PM - Forum: KDE
- Replies (1)
Hi ,
after trying Mandriva I think Mepis might be more suitable after using the Live CD for 2 days or so.
I have just done a full install off the CD ,which i thought went without a hitch
dual boots with XP with GRUB
after rebooting now when KDE starts loading, after the login window I get this message & cant get past it
"there was an error setting up inter-process communications for KDE the message returned for the system was could not read connection list /root/dcopserver_ricks p4_0 please check that the dcopsever process is running "
as said above I cannot get passed this window ,when OK is clicked the screen goes black for 1 second then back to the same message ,
have i missed something when setting up the install?
I am at the moment building a PHP/MySQL-powered CMS, and recently I have had some trouble when moving the latest versions of the source code to other servers. What seemed to be happening was an error 'No database selected' despite the fact that I always used mysql_select_db() before any database queries. So I did some diagnostics into the problem and I've discovered that MySQL is actually denying access to the database. So on my home system, I tried to recreate the problem (running MySQL as root):
Code:
mysql> revoke all privileges on pkbeta.* from pkbetauser@localhost;
Query OK, 0 rows affected (0.00 sec)
mysql> grant usage on pkbeta.* to pkbetauser@localhost identified by "xxxxxxx";
Query OK, 0 rows affected (0.00 sec)
On my home system, MySQL is now denying privileges to select the database. :( But if I revoke usage this happens:
Code:
mysql> revoke all privileges on pkbeta.* from pkbetauser@localhost;
ERROR 1141 (42000): There is no such grant defined for user 'pkbetauser' on host 'localhost'
Which means the usage privileges were never granted. :(
Adding privilieges by name (e.g. select, insert, update...) works. So why is it not granting usage privilieges on my home system, but most importantly why can't I connect to a database on a different server? Eventually this will not be hosted on a machine I have direct control over (no SSH access or anything), so this is a big issue. :(
go to "Desktop" > "System Settings" > "Add/Remove Applications", enter your root password (or run the "system-config-packages" command as root), then tick the check box for "FTP Server", click "close", click "update" and "continue". You may be asked to insert the Fedora Core 4 installation disk #3 to continue. Go to "Desktop" > "System Settings" > "Security Level". Provide your root password, click "ok" and on the "Firewall Options" tab make sure the checkbox next to "FTP" is checked. On the "SELinux" tab under (expanded) "FTP" make sure all the checkboxes are checked except "Disable SELinux protection for ftpd daemon". Check the checkbox next to "Relabel on next reboot". (You will have to re-configure your iptables setup after your next boot as described in the Azureus section above as the relabel process overwrites your iptables files.) Click "ok". Reboot your machine to enable the new SELinux settings.
* If you are using a router you may have to forward ports 20 and 21 in order for your FTP server to work through your router's firewall. See [/url][url=http://www.portforward.com/routers.htm]http://www.portforward.com/routers.htm for instructions.
* Register a new domain name for your Fedora box, have it mapped to your IP address, and configure your /etc/hosts and /etc/sysconfig/network files as described in the web server section above.
* Open the terminal. Type:
su -
Hit enter. Type:
gedit /etc/vsftpd/vsftpd.conf
Hit enter. In gedit change "anonymous_enable=YES" to "anonymous_enable=NO". Also add a line that reads "chroot_local_user=YES" just under the line that reads "#chroot_list_file=/etc/vsftpd/chroot_list". Also uncomment (remove the "#" from) the ascii_ lines so that they read "ascii_upload_enable=YES" and "ascii_download_enable=YES". Click on the "save" icon in gedit to save your vsftpd.conf file and exit gedit. Close the terminal.
* Go to "Desktop" > "System Settings" > "Server Settings" > "Services". Type in your root password in the dialog box that appears and click on "OK". Scroll down the list and check the check box for "vsftpd". Click on the "save" icon in the Service Configuration window and then close the window. Do this for runlevel 3 as well as for runlevel 5. Reboot your machine. You should now be able to access your home user's directory with an FTP client. The host name on your client will be your registered domain name or the IP address of your Fedora Core box. The user ID will be your Fedora Core non-root user name. The password will be your Fedora Core non-root user's password. (I recommend turning off passive mode on the client to speed up transfers. If you use gftp as a client you should edit /etc/sysconfig/iptables-config on the server and add "ip_nat_ftp" into the "IPTABLES_MODULES="" directive so that it reads "IPTABLES_MODULES="ip_nat_ftp". Then reboot your machine.)
This is my conky :) You can see that this is weighted towards gentoo users
Ok, first the preparation. I will assume that you have conky installed (yum install conky, emerge conky, most systems have it now).
Ok so we need to create a couple of scripts to help create the output. If you aren't a gentoo user, skip these.
If you don't have a local user bin folder then you should make it:
Code:
mkdir ~/bin
~/bin/emerge-current.sh
Code:
#!/bin/bash
# emerge-current.sh by Hellf[i]re
#
# This script is designed to read the name of the last package compiled.
#
# As this script does read the entirety of emerge.log, it will be rather
# heavy on the CPU. It shouldn't be enough to be noticable on newer (2.0Ghz+)
# processors, but it still should not be run more often than every 30 seconds.
#
# Usage:
# .conkyrc: ${execi [time] /path/to/script/emerge-current.sh}
#
# Usage Example
# ${execi 30 /home/youruser/scripts/emerge-current.sh}
tail -n 50 /var/log/emerge.log |\
tac |\
grep 'Compiling' |\
head |\
sed -e 's/.*(//' |\
sed -e 's/::.*)//' |\
head -n 1 |\
cut -d \) -f 1
~/bin/emerge-progress.sh
Code:
#!/bin/bash
# source: Jeremy_Z @ forums.gentoo.org http://forums.gentoo.org/viewtopic-t-351806-postdays-0-pos
torder-asc-start-550.html
#
# This script will report the progress of the last emerge command run. It
# reports the TOTAL percentage complete - not the percentage of the current
# package. For example, if there are 110 packages currently being emerged, and
# it is on the 55th package, it will report 50.
#
# Usage:
# .conkyrc: ${execibar [time] /path/to/script/emerge-progress.sh}
#
# Usage Example
# ${execibar 30 /home/youruser/scripts/emerge-progress.sh}
#!/bin/bash
#
# emerge-status.sh by Hellf[i]re
#
# This script will report the current status of portage.
#
# Usage:
# .conkyrc: ${execi [time] /path/to/script/emerge-current.sh}
#
# Usage Example
# ${execi 30 /home/youruser/scripts/emerge-current.sh}
#
# Known Bugs:
# 1) If there are two emerges running at once, when the first one finishes
# running, the script will report the current status as "Completed".
# 2) If there is a emerge running and you run a search, the script will
# report the current status as "Completed", until the running emerge
# moves to the next package, or itself completes.
# The reasons for this are twofold - one, it's a feature;) and two, there
# would be far too much parsing required to find out the current status of
# every command which is run in parallel.
${color #98c2c7}PROCESSES$color
${color #ffcb48}NAME PID CPU% MEM%$color
${color #007700}${top name 1} ${top pid 1} ${top cpu 1} ${top mem 1}$color
${top name 2} ${top pid 2} ${top cpu 2} ${top mem 2}
${top name 3} ${top pid 3} ${top cpu 3} ${top mem 3}
${color #ffcb48}PROC$color $processes${alignr}${color #ffcb48}RUN$color $running_processes
${color #ffcb48}MEM USAGE$color
${color #007700}${top_mem name 1} ${top_mem pid 1} ${top_mem cpu 1} ${top_mem mem 1}$color
${top_mem name 2} ${top_mem pid 2} ${top_mem cpu 2} ${top_mem mem 2}
${top_mem name 3} ${top_mem pid 3} ${top_mem cpu 3} ${top_mem mem 3}
NOTE if you are not using gentoo remove the lines at the end of the ~/.conkyrc from PORTAGE down to the end. And if you ARE then switch the "/home/znx/bin" for your user. :)
Thats it.. start up conky and it will run. If you are using KDE/Gnome you will need to tamper with them to make it work because conky doesn't work well with them yet.
Conky's Home where I sourced the script and 99% of the idea to the production of this layout.
Weeeeee.. just a quick jaunt.. ask questions if you have any
This line is used to ping another system in my local network, you can either edit the name "mute" to be another machine, or remove the line as it will loose 100% of its packets.
See below for information about securing the genlop command.
Updated to fix mistakes found by dragon_788, many thanks!
