Wayne D's Blog

Evidently sometimes Windows 10 will add a partition when upgrading which means grub can't find it's config file.

To fix this you can boot using the Centos 7 install DVD into rescue mode. You need to choose the boot option that matches whatever version of Centos you installed, either legacy boot or EFI,

chroot /mnt/sysimage
(for legacy boot)
grub2-mkconfig -o /boot/grub2/grub.cfg
grub2-install /dev/sda (replace the a with whatever drive is your boot drive)

1. Install the EPEL repository. If you are running Centos 7 you can install EPEL by running yum install epel-release. The package is included in the CentOS Extras repository, enabled by default. If your are running RHEL7 or another variant you can run:

rpm - i https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm

2. Install the ntfs support modules with yum install ntfs*

3. Mount the Windows drive using either the filemanager or the disks utility

4. run grub2-mkconfig -o /boot/grub2/grub.cfg

You should see a message mentioning that Windows is one of the boot options

There was an error when compiling the vcglib package, access.not in vcglib/wrap/ply/plystuff. All you have to do is add #include <unitstd.h> on a new line after #include <fcntl.h>

Port forwarding mysteriously stopped working on my Dlink DIR825-C1 running DD-WRT. After trying many different things and almost ditching my ISP I found the following solution on http://www.dd-wrt.com/phpBB2/viewtopic.php?p=673373, thanks to user mojud.

In the Control Panel click on "Administration" then on "Commands". Copy and paste the following inside the Commands box:

iptables -t nat -A POSTROUTING -j MASQUERADE

Click on "Save Firewall" and then "Run Commands"

Before installing CUDA 7.5 on Centos 7 you need to update both your kernel and X using yum update otherwise mayhem will ensue and X won't start. I installed them in the wrong order and ended up with the following error when I tried to start X.

[ 467.272] (EE) NVIDIA(0): Failed to initialize the GLX module; please check in your X
[ 467.272] (EE) NVIDIA(0): log file that the GLX module has been loaded in your X
[ 467.272] (EE) NVIDIA(0): server, and that the module is the NVIDIA GLX module. If
[ 467.272] (EE) NVIDIA(0): you continue to encounter problems, Please try
[ 467.272] (EE) NVIDIA(0): reinstalling the NVIDIA driver.

reinstalling the nvidia drivers didn't solve the problem the key turned out to be this which occurs a bit higher in the Xorg.0.log file

[ 467.240] (II) LoadModule: "glx"
[ 467.240] (II) Loading /usr/lib64/xorg/modules/extensions/libglx.so
[ 467.261] (II) Module glx: vendor="X.Org Foundation"
[ 467.261] compiled for 1.17.2, module version = 1.0.0
[ 467.261] ABI class: X.Org Server Extension, version 9.0
[ 467.261] (==) AIGLX enabled

That's not the glx driver that NVIDIA needs so I went to /usr/lib64/xorg/modules/extensions and removed the libglx.so that was there and made a link to the NVIDIA version.

cd /usr/lib64/xorg/modules/extensions/
rm libglx.so
ln -s /usr/lib64/nvidia/xorg/libglx.so

not pretty but it worked.