Geneticus

OK after looking around in some of the network tools, I have discovered that the Public NIC eth1 isn't connected. I go into Monitor connections, turn on the logs and hit connect ethernet. A few seconds later it says the same thing and there is no error in the log. just: net_monitor[8959]:running:/sbin/ifup eth1 daemon I went back into manage connections and the gateway listed on both NICs is the same which is not what I put in originaly, nor what I put in the second time around. I clicked apply and closed then reopened the window, the NIC was reset to the other NIC's Gateway.....

Security level set to High Nothing in hosts.allow Nothing in hosts.deny Not sure if this is important or not.... I have 2 NICs in this box. One is behind a router with a static IP for filesharing, the other one, the problem child is open to the net.

I seem to be having a firewall issue I have outgoing connectivity but nothing incomming. I set up a teamspeak server earlier and it was running for a few hours and connectable. After rebooting it is no longer working. I restarted the server (TS) and can access the web admin from that box but from no where else. I looked at the NIC settings and the gateway field was blank. Updated and restarted the NIC. I checked the firewall and the port is correct. I am not behind a router (static IP). I also alowed ping requests in the firewall settings and cannot ping from another location. Any bright ideas out there?

the /boot/initrd-2.6.17-6mdv.img and the initrd.img symlink was pointing to the .....17-5mdv.img file I am thinking that because my root partition was full it couldn't make the img file.... the config-2.6.17-6mdv and system.map-2.6.17-6mdv were created though No worries though, the 2.6.18.2 kernel is working fine and I managed to get the Nvidia drivers working and detected by cedega. I had 2 problems there. 1. The ati driver was still installed 2. I needed to downgrade driver versions due to a driver bug that wouldn't let opengl/glx function. Now if only I could get moodin installed :(

Solved! from the "other" white meat: Clicky

I'm trying to install on a Dell smartstep 200n laptop and BIOS won't boot to the CD, but I can read from it in dos/Win. Is there a way I can get the installation started via boot floppy or (cringe) LAN boot??

I think this got misssed.... When I was doing the kernel upgrade with urpmi I wasn't able to make a boot img file. Everything else was created except that. I didn't think a manual install of the mdv kernel would go well so....well you know the rest... I got a vanilla kernel compiled and installed and can boot to it for the most part. Urpmi isn't my friend. when I ran it a ton of crap downloaded and filled my root partition. 7gb :( so KDE won't load till I get some stuff cleaned out. I'll probably go and do a total reinstall though to a faster drive with a little more space. I was using a 40gb drive for mandriva, a 60gb mount for /www and a 60gb mount for /ftp. I'll do a reinstall and load up mandriva on the 60, use the 40 for /www, leave ftp alone, and save the other 60 for downloads... BTW this is the guide I followed for the vanilla kernel upgrade, worked very well for me. Quick and Easy Kernel Compile Upgrade, Mandriva 2006

Why is that? how are they different?

I thought I needed a new kernel img which I don't have so I decided to try another route After much searching I figured out how to compile and install a kernel waiting for it to finish building so I can reboot and try the nvidia launcher again with the newer drivers. While not as easy, it's good practice...

So I choose to proceed because it seemed easier that fumbling around with the new one. I got this: Creating symlink /var/lib/dkms/nvidia/8774-4plf2007.0/source -> /usr/src/nvidia-8774-4plf2007.0 DKMS: add Completed. Error! Your kernel source for kernel 2.6.17-5mdv cannot be found at /lib/modules/2.6.17-5mdv/build or /lib/modules/2.6.17-5mdv/source. You can use the --kernelsourcedir option to tell DKMS where it's located. Error! Could not locate nvidia.ko.gz for module nvidia in the DKMS tree. You must run a dkms build for kernel 2.6.17-5mdv (i586) first. So I appear to still be running 5 and have source for 6...

I didn't get the nvidia driver installed I am curently using the VESA driver. The other driver I installed was for the ATI card. urpmi dkms-nvidia ok that worked partway, but It wants to install older drivers now. Current ones are 8776 from nvidia and it wants to do the 8774

I have been having a doozy of a time getting graphics drivers to work I started with an ATI radeon 9800 pro and managed to get the driver installed but I couldn't get cedega to see anything other thatn the stock driver. After 2 days of head banging I gave up and put in an old GF TI4600. I am getting the Kernel module load error, inserting /usr/src/nv/nvidia.ko Invalid Module format After several hours of google I have reached an impass and what I think may be causing the problem. My config: 2007 + free dist. I used the update installer thingy to get the newest source but I don't seem to have the source for what I am actually running. my gcc is 4.1.1 20060724 (prerelease) 4.1.1-3mdk uname -r returns: 2.6.17-5mdv but rpm -qa|grep kernel returns: kernel-2.6.17-5mdv-1-1mdv2007.0 kernel-2.6.17-6mdv-1-1mdv2007.0 kernel-source-2.6.17-6mdv-1-1mdv2007.0 kernel-doc-2.6.17-5mdv-1-1mdv2007.0 Now I have tried to recompile and use the newer one so the source matches but I get an error that it is already installed. If you want to know what my "blah blah blah" is, please give me the commands and locations to run them as I am a console retard.

After I realized that I was doing things the lazy slow way I decided to poke around in the case again. I moved the drives around so that: hda is PRI MAST hdb is SEC MAST CDR is SEC SLAVE was able to boot into the first boot wizard. Wizard? sounds like the OS I'm trying to get away from. Warlock, Witch, Magician even.... ah well thanks for motivating me to solve my own problem :P

[root@rescue /]# grub-install /dev/hda1 I get an error: mkdir: cannot create directory `/boot' : File exists Like I said I'm pretty much a newb when it comes to linux. About all I know how do do is use ls, cd and nano I tried to just run grub. I got into it but when I did setup (hd0) I got: Error 15: File not found I'm guessing it's because there is no grub.conf, but I don't know what text editor is available at the rescue console to create one. Since the rescue gui is trying to reinstall lilo to /dev/hdb when there is no hdb in the system couldn't I edit wherever that path is stored and change it to /dev/hda? Just need to know what editor to use and where the data is stored.....

pulled the raid card and wd drive. rescue console won't let me re assign the bootloader to hda1. Is there a way I can do it from the rescue console? things seem to be in slightly different places than they were in gentoo. or is there a tiny version of linux I can install to a hdd to test with? I'm trying to avoid 1 1/2hr reinstall between TS steps....