ianw1974

It sure is possible, older hardware would become deprecated and unsupported at some point. Whilst some/most might work, you'll get some that doesn't too :)

I've had problems with some pdf's before, and they could only be viewed correctly in Adobe Reader.

I think the reason you can't see the text is due to the change of resolution meaning you'd have to change the geometry settings on your monitor to reposition how the screen looks and/or change the height/width so that everything is visible. So that isn't a major problem. Try the Xorg -configure as mentioned and let us know if it goes any better. It should start at least with a minimal config and a default driver like vesa or nv if an nvidia card. I wouldn't suggest an ATI ones since they are much more problematic than nvidia ones. So stick with the nvidia one, you're more likely to have this working far easier than an ATI one.

Check to see if xorg.conf exists in /etc/X11 directory. If so, rename it to xorg.conf.old and reboot. Xorg doesn't need an xorg.conf nowadays, so maybe there is one there just a bit strangely configured. Another thing you can do is switch to runlevel 3 and then do something like this: Xorg -configure and then copy the xorg.conf from the /root directory into /etc/X11 and see if it works any better. So in short, first step: su (enter root password when prompted) cd /etc/X11 mv xorg.conf xorg.conf.old reboot, see if any better. If not: su (enter root password when prompted) init 3 Xorg -configure cp /root/xorg.conf.new /etc/X11/xorg.conf and reboot. I'm assuming that the new xorg.conf placed in root will be xorg.conf.new. Let us know how you get on.

Hi, there can be a chance that kdesu or gksu are not installed. I'm not on a Mandriva system right now, but you should be able to do: urpmf --provides gksu urpmf --provides kdesu from a console as root, and it should tell you which package provides it, then you can install accordingly. I believe I noticed this and installed it separately on my Mandriva 2010 system. Although I don't use KDE, so I never checked that, but theoretically it should be the same.

About as far as I've ever gone into "testing" realms is by using RC1 or RC2 candidates. Mandriva's 2009.0 RC1 was actually pretty good. I tend to stay away from beta because of their volatility. I did at one stage use Debian SID effectively "unstable", and it was. One day I updated, and something broke, the next day or two, it was fixed again when updates were released. But I thought what if I had really needed that functionality, I'd be screwed until it was fixed. Ever since then I've never gone that route again.

OK, I'm not really seeing your webcam being recognised, unless of course it is Bus 001 Device 002: ID 1e4e:0100 and it seems that it is because when I check here: http://linux-uvc.berlios.de/#devices your camera is listed, but the note says: Yours is connected to a USB 1.1 hub. If it's a desktop machine, I suggest getting a USB 2.0 card installed to make use of your webcam. Or try a different webcam that is potentially support.

Click Cheese and then the Video option. Any better? If not, then please do the following for us from a console prompt: su (enter root when prompted) lsusb copy and paste the results here so we can see more about your webcam.

I used testdisk and photorec in a much more difficult situation. A hard disk in a laptop had been encrypted. Problems were encountered with the machine, and so they decrypted the hard disk. Problem was, after decryption, entire partition table was lost - similar to what you experienced. They also had more than one partition on disk, which meant it would be hard to re-create it and hope for the best. testdisk did the partition table magic, and I used this to get the majority of the files back. Then I used photorec just to be sure, which gives the random naming because it works at a much deeper level than testdisk. testdisk is good in everyday stuff, but worst case scenario photorec will get you going. And I'm glad it got you sorted out too.

I would now format the card in the camera as usual. Something happened that borked the partition table. Formatting again in the camera should bring it back to a sanitised state. Just for info, testdisk is good at recovering the partition table, and files that existed and their filenames. If you are sure that something is not right, then you can perform data carving, which is what photorec does. Photorec doesn't care about the partition table or even what filesystem was on the disk. It will just look at a sector level (I think) and recover the pictures. Of course, because of this recovery process, it doesn't know about filenames, so they get assigned numeric values. However it does know by file type, hence the extension should stay the same, as jpg, doc, etc, etc.

I would do this: fdisk -l as root, that's a lower case L. Do this with the card inserted, and other than your usual hard disks do you see it? Does it have a partition table? If not, don't worry. I suggest then that you install testdisk, and use this to recover the partition table. It will detect it, and then should recover it. Then after this, you can look at recovering the files. testdisk runs under linux, if it's not available in the repo, you can download the file and just extract it and run it. Available here: http://www.cgsecurity.org/wiki/TestDisk

Did you upgrade your system? If so, perhaps the old config for brasero is the problem. You can easily check/test this by creating a new user on your system temporarily. Then login as this user, and see if brasero runs like it normally should. If so, that would suggest that you need to delete the brasero config from your user profile. Let us know if that helps, as we can always look further if it doesn't.

I've noted a lot of flash problems all when mine upgraded, certain sites stopped working, or crashed firefox.

When booting, press ESC to get rid of the splash screen, and see if you've got the interactive option. Haven't checked this, and didn't want to reboot my system as I'm at work now :) Along time a go there used to be the harddrake service, but it doesn't exist anymore. Else I'd have suggested disabling this. Also, there is a failsafe option from your grub list of kernels, try this. Alternatively press "E" at the grub screen, and edit the kernel line for the kernel you are running, and at the end of the line add "single" without the quotes so that we can boot single mode. One more thing, when you did get to the console window, could you do anything such as login as root or something? Maybe just X locked up, or is it the whole system?

My pleasure, hope that helps :)

You really only need three additional, so swap at 1GB would be fine, it could even be 512MB if you have lots of ram. Then a / partition with 10GB, would be good, and the rest for /home.

My server has one nic, and I have all my servers using this. I have a client installation, whereby I'm using three nics. Two of which are using public IP's because they are using two internet connections and have configured clustering with load balancing to use both internet links at the same time for the service my solution to them is providing, and the third nic is using a private ip address for connection to a NAS for backups of the systems. Depending on how you are connecting this, I'm thinking you want to use both NICS but on the same LAN so that you get better performance but correct me if I'm wrong. Anyway, if that is the case, for the main server running Xen, you'll only want eth0 or whatever configured with an IP address, and just leave the other one without an IP. Then just bind both in the xen config so that you can use both network cards with the virtual machines. Then later, you just use the one that you want in the machines, be it eth0 or eth1 and then you have both nics in use with Xen for balancing out all your machines. Or you can do what I did, and use a separate NIC for a separate network segment.

You can do it with just these two machines, although not ideal. Ideally I would have had four servers in this scenario. Two providing the active/standby cluster machines with the VIP address bound to them. The remaining two being the real servers, which have the httpd and postgresql on them. The first two servers would redirect and load balance between the two back-end (real) servers. If you only have these two servers, then you can run the high-availability daemon on both machines, but then you don't really have a cluster as such. All you are doing is just monitoring either server, and then bringing up the standby server if heartbeat fails on the primary one. So, if you understand what I put, I can help further if I know if you have more than just these two servers or not.

I'm with tux99. My virtualised server environment comprises base system of CentOS 5 x86_64, with four virtual servers. One of which is Red Hat Enterprise Linux with subscription, the remainder are CentOS 5. All virtualised servers are 32 bit versions. All running under Xen.

I use Xen virtualisation and I have four virtual machines under here - Web Server, Mail Server, Proxy Server, Backup Server. I much prefer this, because if someone hacks my web server, then I can just kill the machine and reinstall a new one remotely. If it was installed as a normal server, I would have a problem meaning I would have to visit the site where the server is to reinstall a clean server. So in reality, the virtual server will be less problematic in the event of an attack. Admin is the same, just that you have more servers to connect to but that is minimal.

Glad you got it sorted

No, the logical partition is not the problem. Did you chroot into all your partitions once you booted from the rescue CD?

Sweet updating now.

You can safely remove the old kernels. If you are not booted in that old kernel, then your not using it.