SurlyJake Blog

ubuntu cloud images in libvirt and virt-manager

2020-10-09T20:53:00-07:00

I stopped using IOS's to create VM's. I started using 'cloud images' I might expand on the resons to do this here... lets hope this feature request in the GUI virt-manager gets some traction: https://github.com/virt-manager/virt-manager/issues/143.

NOTE

i'm going to show examples using virt-install and running it with my user account, which in turn, uses the qemu session connection. this gets around filesystem permission issues, etc. there is a difference between qemu:///session and qemu:///system. Just understand that there is a distingction. If you want to use something like virt-manager with this vm afterwards,you can simply connect virt-manager to qemu://session in addition to the default qemu:///system by selecting File-> Add Connection. You can read more about this subject at: https://blog.wikichoon.com/2016/01/qemusystem-vs-qemusession.html

NOTE

if trying to connect to a bridge as i did below, you need to create a file to let qemu allow users (using the aforementioned qemu:///session) to connect to the system-level bridge. Edit the file /etc/qemu/bridge.conf and add a line that corresponds to tbe bridge you'd like to use. In my case, all it needs is "allow virbr0". edit to suit your machine.

grab the image. Start here: http://cloud-images.ubuntu.com/. Navigate their complicated folder structure to download a specific version. I used https://cloud-images.ubuntu.com/daily/server/focal/20210720/focal-server-cloudimg-amd64.img.

I cant find documentation on what each of the different formats are other than the rather brief description column and making an inference on the file extention. For example, lets look at these two examples:

https://cloud-images.ubuntu.com/daily/server/focal/20210720/focal-server-cloudimg-amd64.img
https://cloud-images.ubuntu.com/daily/server/focal/20210720/focal-server-cloudimg-amd64-disk-kvm.img

so whats the difference? if you're going to run the image in kvm, you might be tempted by the -kvm image, but from what i can see, by looking at the .manifest files https://cloud-images.ubuntu.com/daily/server/focal/20210720/focal-server-cloudimg-amd64.manifest and https://cloud-images.ubuntu.com/daily/server/focal/20210720/focal-server-cloudimg-amd64-disk-kvm.manifest, it looks like the -kvm image has a kernel with kvm enabled, which would only be needed for a kvm hypervisor, not a guest.\</rant>

    wget https://cloud-images.ubuntu.com/daily/server/focal/20210720/focal-server-cloudimg-amd64.img

Now create some cloud-init configurations documentation here: https://cloudinit.readthedocs.io/en/latest/. its complicated. in our simple environment, we're going to use 'datasource' of 'NoCloud'https://cloudinit.readthedocs.io/en/latest/topics/datasources/nocloud.html. which is just a .iso file with two files: meta-data and user-data. those two files define all of the configuration that will be dropped onto the vm at first boot.
meta-data has the machine specific data like hostname and network configurations. For this use-case we're gonna let it default to dhcp, so lets keep it basic and only set and instance-id and local-hostname.
```
instance-id: foo-tester
local-hostname: foo-tester
```

user-data is documented here: https://cloudinit.readthedocs.io/en/latest/topics/format.html# . here we'll set up a few things like our user and ssh key. I'm going to use docker on these hosts, so i'll install and enable docker and docker-compose.

users:
  - name: foobar
    ssh_authorized_keys:
      - ssh-rsa <<copy your ssh pubkey from ~/.ssh/id_rsa.pub>>
    sudo: ['ALL=(ALL) NOPASSWD:ALL']
    groups: sudo

# runs apt-get update
package_update: true

# runs apt-get upgrade
package_upgrade: true

# installing additional packages
packages:
  - docker
  - docker-compose

# run some commands on first boot
runcmd:
  - systemctl daemon-reload
  - systemctl enable docker
  - systemctl start --no-block docker

# after system comes up first time. 
final_message: "The system is finally up, after $UPTIME seconds"

use virt-install to pull in the user and meta data files for first boot, start the vm, configure it.

virt-install \
  --name foo-tester \
  --memory 4096 \
  --vcpus 4 \
  --os-type linux \
  --os-variant ubuntu20.04 \
  --network bridge=virbr0 \
  --nographics --disk size=10,backing_store=$PWD"/focal-server-cloudimg-amd64.img",bus=virtio \
  --cloud-init user-data=$PWD"/user-data",meta-data=${PWD}"/meta-data"

Your terminal session will attach to the VM. let it boot for a little bit so that you can see the dhcp'ed address print out to the console. When you want, detach from the serial console hold down control, then press seven key followed by the ] key. ( ctrl+^])
if anything goes wrong, you'll want to get rid of the machine that you created to start over:

virsh destroy foo-tester || virsh undefine foo-tester --remove-all-storage

you can scroll up in the output to find out the dhcp address your VM picked up:

[    8.499751] cloud-init[652]: ci-info: +--------+------+----------------------------+---------------+--------+-------------------+
[    8.501392] cloud-init[652]: ci-info: | Device |  Up  |          Address           |      Mask     | Scope  |     Hw-Address    |
[    8.503118] cloud-init[652]: ci-info: +--------+------+----------------------------+---------------+--------+-------------------+
[    8.504659] cloud-init[652]: ci-info: | enp1s0 | True |      192.168.122.212       | 255.255.255.0 | global | 52:54:00:xx:xx:xx |

ssh foobar@192.168.122.212

QT segfault when accessing ui objects

2013-10-10T20:53:00-07:00

I'm starting with QT for a hobby project. My design called for a public method inside of my QMainWindow main window. the method will update the text in the display. When calling anything using the ''ui->'' pointer, the app would segfault. The QT Creator debugger showed the ui pointer as ''. That was a clue. To confirm, I wanted to see gdb's output.

To enable debug symbols for GDB, I had to edit the '.pro' file and add the line ''#CONFIG += debug''. After the edit is made, change directories to the project folder, run qmake, make clean, and make. Once the build is completed, 'gdb ./terminal' (terminal is the name of my program).

$ gdb ./terminal
(gdb) run
Starting program: /home/jacob/program/./terminal 
Program received signal SIGSEGV, Segmentation fault.
0x0804f8f3 in MainWindow::updateStatus (this=0x2c74c085, status=0xbfffee4c) at mainwindow.cpp:233
233     qDebug() << ui->valueLabelLocusStatus->text();
(gdb) print ui
Cannot access memory at address 0x2c74c09d

After the segfault, I tried to get it to print the value of the 'ui' pointer, it said ''cannot access memorry at address...''. That confirmed it, Something is wrong with the ui pointer.

The problem was threading. Because my application was in a thread other than that of the main window, access to it's widgets isn't allowed. To resolve the issue, I followed the QT documentation at http://qt-project.org/doc/qt-5.0/qtcore/signalsandslots.html . I changed the public method on the mainwindow class into a QT slot. I made my class extend QObject and added a signal to my class instead of calling the method directly. Once the signal is connected to the slot, QT handled the threading issue.

Easily Test Packages From Arch Linux Testing Repositories

2012-03-29T12:18:05-07:00

Sometimes, when a new kernel or application that I use a lot comes out, I want to quickly download and test it on my system. It is simple enough to edit /etc/pacman.conf and enable the testing repositories, then update your system. Problem is, enabling the testing and/or community-testing repos at their default position will update all packages which have versions in testing.

If you put the [testing] and [community-testing] repos after [core] and [community] accordingly. This will tell pacman to prefer the stable versions of packages. Running pacman -Syu will result in the same set of packages as before.

Once pacman is configured with the testing repositories enabled as described above, you can selectively run versions of applications that are in testing with all of their testing dependencies without messing about. If you want to install a package from the testing repository, you simply prefix it with the repository's name. Here is an example:

Linus releases a new kernel and you have read that it fixes some bug that has been plaguing you for months. It's in [testing], but you just can't wait. You can:

Go to the Arch packages site and manually download the .pkg.xz files of the kernel and all it's dependencies, then install them with pacman -U.
Enable the [testing] repo and run an update. This might give you a lot more than you bargained for.
If you configured pacman as I described above, you can simply run pacman -Syu testing/linux

If you have already installed a package which requires an older package than the one you are instaling, pacman will prompt you with the problem like this

error: failed to prepare transaction (could not satisfy dependencies) :: nvidia: requires linux

You can usually sove this by installing a newer version of that dependency. The solution in this case is to also install the nvidia package from testing:

pacman -Syu testing/linux testing/nvidia

Now It should go through.

Disclaimer: Obviously running something from testing is done at your own risk. Be sure to read up on the repositories you enable here: https://wiki.archlinux.org/index.php/Official_Repositories

New ArchLinux mirror

2012-03-15T15:36:05-07:00

I love Arch linux and have been using it for a few months now. When I started, many of the mirrors I used were slow or would go offline and/or out of date. When I saw an offer for a VPS which should allow me to host a mirror of the archlinux package archives, I snatched it up. If you run arch, you can point pacman at my server by adding this line in /etc/pacman.d/mirrorlist:

Server = http://archlinux.surlyjake.com/archlinux/$repo/os/$arch

I threw together a quick page showing some general stats for the server which you can view at http://archlinux.surlyjake.com/.The server has been added as an official Arch Linux mirror, but is not in the pacman-mirrorlist package (I'm not sure what the process for that is).

Review After One Year of QuickWeb VPS Hosting

2012-02-19T11:05:42-08:00

One year ago, I purchased a VPS from QuickWeb. I found the deal at my one of my favorite sites for cheap hosting: www.lowendbox.com. The deal was:

OpenVZ VPS
256MB RAM/386 Burst
10G disk space
250GB bandwidth/ Month on 1 IP
$35.88/ 1 Year

I ordered without much of a problem and my Debian server was provisioned within a few minutes. I was a happy camper. $2.99/month is a great deal for root access to a server, even one with such meager specs. I've had a decent year of hosting with QuickWeb, but I decided to take my business elsewhere when my plan came up for renewal. To help explain why, I want to share the three support tickets I had to open with them. This will give you the clearest idea of what you get for the money. I Would have included the full text of these exchanges, but they insert a privacy notice in the signature of their support emails, so i'm worried that I may have agreed to something in their TOS.

The server that QuickWeb provisioned for me was fully loaded. For me, that wasn't such a good thing. The server already had web, DNS, and mail server software installed and running (apache, named, and postfix IIRC). I wasn't comfortable with that because I prefer to start from scratch, planned on using nginx and didn't need to run a DNS and public mail server. I opened a ticket asking if i could get a server provisioned that only includes the base Debian install. The response from QuickWeb was that they don't have a minimal template 'yet', but that I could uninstall packages that i don't need using apt-get. Thanks.
In July of last year, there was a service interruption. This site was down. Not just me, I checked stuff like connecting from a few other servers that I have access too and, of course, http://www.downforeveryoneorjustme.com/. I opened a ticket. An hour and a half later, a response told me that I had sent the ticket to sales instead of the helpdesk and next time, to use the helpdesk for faster response. Another hour and a half later, they replied and told me that the VPS was actually up the whole time, but there was a brief network interruption in their management network, so I wan't able to connect to check the status of my VPS. I had to reiterrate that whether or not the server was UP, It was unreachable over the network for an hour and a half. The last reply from them came 8 hours in. They asked me to do stuff like turn off IPtables and give them a traceroute. I gave them a ticket without a lot of information, somehow sent it into their sales queue, but it took them 8 hours to get me a semi-intelligent response. They never acknowleged a problem with public connectivity to my server.
The isue that sealed the deal for me happened this December when QuickWeb changed the IP of my server. Despite their claims to have notified me, I never got an email from them. I used a gmail account to handle all their communication with me. Yes, even checked the spam folder. No notice. I submitted a ticket for this one too and all they did was tell me that they sent me emails about it while I told them that they didn't.

The server this site is runing on now is a KVM VPS from bitcable. So far the support there is night and day compared with QuickWeb. Preetam helped me set up the virtIO drivers on FreeBSD. More on that later.

Create a custom application launcher in GNOME3

2011-12-30T09:26:51-08:00

On Arch, the system-wide shortcuts are stored in /usr/share/applications. Each one is a .desktop file with a few parameters. user-specific icons are stored in ~/.local/share/applications. I downloaded eclipse and extracted it in my home directory, so I'll use eclipse as the example application.

Create the .desktop file

nano ~/.local/share/applications/Eclipse.desktop

Paste this in:

[Desktop Entry] 
Encoding=UTF-8 
Version=1.0 
Type=Application 
NoDisplay=false 
Exec=/path/to/eclipse/eclipse 
Name=Eclipse Comment=Launcher for Eclipse

Each line is pretty self-explanatory. Edit to suit your needs. As soon as you save the file, you will be able to see your new shortcut in the list.

(Optional) Give the shortcut a nice-looking icon: You probably noticed that your icon is an ugly diamond. To pretty it up, you simply have to add one more line to the desktop file.
```
Icon=/path/to/eclipse/eclipse.svg
```
Your application may have included a png icon, but to make it really look nice, you should grab an SVG file of the logo from somewhere. Arch has a very simple packaging process which lets you examine all of the package building elements through their webpage at http://www.archlinux.org/packages/. I was able to grab the SVG which the package uses like this:
```
wget -O eclipse.svg http://projects.archlinux.org/svntogit/packages.git/plain/trunk/eclipse.svg\?h\=packages/eclipse
```

Wordpress HTTPS secure login

2011-12-20T21:36:11-08:00

Back when I started with Wordpress, I had to create server rewrites to force my browser to use HTTPS. That changed back in 2008, but this is the first i've heard of it.

Wordpress 2.6 (released in July, 2008) and newer include the tools do do this out of the box! Using this documentation, you can get rid of some old .htaccess or nginx rewrite rules: http://codex.wordpress.org/Administration_Over_SSL. All you have to do is add the declaration in wp-config.php:

define('FORCE_SSL_ADMIN', true);

This forces logins and administrative sessions to use SSL.

Now, you can remove the old nginx rewrites. They aren't needed anymore:

rewrite ^/wp-login.php(.*) https://www.surlyjake.com/wp-login.php$1 permanent;
rewrite ^/wp-admin(.*) https://www.surlyjake.com/wp-admin$1 permanent;

WordPress Redirection Plugin

2011-09-09T15:10:17-07:00

This blog is undergoing some changes.There is some content (like the zabbix pages) that I can no longer maintain since I no longer use them in my day-job. To keep everything working seamlessly, I wanted to create some 302 (permanently moved) redirectors for the content that I was taking down. In the past, I used a plugin called "smart 404" to intelligently redirect visitors to a good page instead of the dummy standard 404 page. Unfortunately, It looks like smart 404 is having some trouble in newer WordPress versions and wouldn't create a smart suggestions on my 404 page.

Enter Redirection. Redirection gives you a menu in the admin interface which allows you to easily create and maintain 301 and 302 redirections (among other features). Here is a screenshot of what it looks like once activated:

{% img http://www.surlyjake.com/images/blog/wordpress-redirection-plugin.jpg %}

From the screenshot, you can see that I was able to create a simple 301 redirect from a page on my WordPress instance to an external website where the content will live from now on. The second rule allowed me to match the URL using a regular expression and redirect it to another page on my site. This is a great piece of software.

One issue I ran into when testing the plugin was that edits I made did not seem to be taking effect. This turned out to simply be my browser caching the 301 response. I recommend that you either manually clear the cache or download the web developer toolbar and disable cache. I should also note that my server runs nginx and not the Apache web server. I simply deleted the redirection 'module' for Apache which attempts to create .htaccess rules to perform the redirections. Without the Apache module, WordPress handles the redirections.

Redirection Plugin documentation: http://urbangiraffe.com/plugins/redirection/

Prevent samba from creating home directory for machine accounts.

2011-08-05T17:34:29-07:00

Since this setup automatically creates a home directory for a user as soon as they access the machine (via samba, ssh, console), I noticed some directories being created for computer accounts. To prevent this, I have added a line that only "Domain Users" are allowed to authenticate. To add this:

wbinfo -n "Domain Users"

It will spit out the SID for the Domain Users group. Something like this:

S-1-5-21-((some number))-((some number))-((some number)) Domain Group (2)

Take that number and change the Pam_winbind.so line in /etc/pam.d/common-session to look like this:

session sufficient pam_winbind.so require_membership_of=S-1-5-21-((some number))-((some number))-((some number)

Create a single standalone .exe from a Python program

2011-03-28T08:53:00-07:00

I have been working on a small windows command line tool that I wanted to distribute it as a single executable file on windows. I tried cx_freeze and py2exe. Both of these tools worked well, but I couldn't find an easy way to compress make the whole program into a .exe file. py2exe and cx_freeze both create working programs, but there are always some dependent .zip archive or .dll's somewhere that need to be distributed with it. Pyinstaller, I found, actually compresses everything into a single .exe. This makes a pretty big executable (my small command line utility created a 5MB .exe file), but it's simple and it works.

To use pyinstaller:

grab pyinstaller 1.5rc (1.4 doesn't work with python 2.7). extract the zip file anywhere.
change directories to the pyinstaller folder you just created.
Before you create your first executable, you will have to run this once.
```
python configure.py
```
Now, pyinstall needs to scan through your program and create what they call a spec file.
```
python makespec.py --onefile path\to\program\program.py
```
Now, run this command to generate the executable.
```
python build.py program\program.spec
```

Once the command has finished, the standalone executable will be available in the program\dist folder inside of pyinstaller. Instructions for how to do this for a linux executable on ubuntu linux can be found here: http://excid3.com/blog/2009/12/pyinstaller-a-simple-tutorial/. You can find more info on pyinstaller at their website: http://www.pyinstaller.org/.

Install Open-VM-tools on Debian 6 squeeze

2011-02-08T15:51:53-08:00

Installing Open Virtual Machine Tools is a nice alternative to managing and installing the vmware tools package that comes with Vmware's hypervisor products. It allows you to use APT to manage all the dependencies and updates. Module assistant will recompile the kernel modules for you automatically when you update kernels too.

I have verified that this configuration works under ESXi 4.1.0, 260247 and a new Squeeze install. All of the essential functions like being able to see guest OS information on the Vsphere summary page and performing "Shut Down Guest" and "Restart Guest" all work as expected. If your experiences vary, please leave a comment

This should all be performed by the root user (otherwise prepend 'sudo' to the commands if you must use sudo). _ The first thing you need to do is add the 'contrib' archive to your apt sources config:

nano /etc/apt/sources.list

Add the 'contrib' to the end of these source lines. I use the debian.uchicago.edu repository, so mine look like this on a new install:

deb http://debian.uchicago.edu/debian/ squeeze main contrib
deb-src http://debian.uchicago.edu/debian/ squeeze main contrib

After you save changes to the file, simply update your package list and perform the install:

apt-get update
apt-get install open-vm-tools open-vm-source

You might see an error that looks like this:

Loading open-vm-tools modules: vmhgfsFATAL: Module vmhgfs not found.
 vmmemctlFATAL: Module vmmemctl not found.
 vmsyncFATAL: Module vmsync not found.

Don't worry, this is because we have not installed the kernel modules for some of the more advanced VMware functions like the vmhgfs file system, Memory Balloon (vmmemctl) Driver, and vmsync, which helps freeze the VM while taking snapshots. To install these kernel modules, we're going to use module-assistant. It will compile the modules for your kernel:

module-assistant auto-install open-vm -i

The '-i' puts module assistant into non-interactive mode. It wont ask you to confirm any additional package installs necesarry to set up the build environment. If you're running a server without a Graphical environment, you're finished. If you want features such as automatically adjusting the VM's monitor resolution or allowing your mouse to seamlessly move between your host system and the VM, you need to install one more package:

apt-get install open-vm-toolbox

Now you won't need to use CTRL+ALT to switch back and forth.

Tar extract destination folder

2011-02-08T11:51:10-08:00

Tar can redirect extracted files to a different folder. This can be handy if you want to unpack a tar archive stored on a cd or other read-only media and want to eliminate the step of first copying the archive, extracting it, then deleting the extra tar. All you have to do is add a '-C ' to the tar command. So here's an example using the VMware Tools tarball:

# tar -C /root -zxvf /media/cdrom0/VMwareTools-8.3.2-257589.tar.gz

This will redirect the extracted files into the /root folder. simple

Brasero, no option for creating ISO

2011-01-19T13:37:20-08:00

I love Brasero. It's simple, lightweight, and comes with Ubuntu. Recently, when I wanted to back up a disk, ISO was not on the list of formats. As is usual, the issue was a missing package: brasero-cdrkit.

apt-get install brasero-cdrkit

Now, .ISO (ISO9960) will be an available disk format for you under Brasero.

<a href="http://www.surlyjake.com/wp-content/uploads/2011/01/Brasero-iso.jpg"><img src="http://www.surlyjake.com/wp-content/uploads/2011/01/Brasero-iso.jpg" title="Brasero-iso" height="520" width="552" alt="" class="alignnone size-full wp-image-690"></img></a>

Tandberg endpoints cannot establish calls over Juniper Routers

2010-11-19T10:52:08-08:00

This is a nasty little issue. On my private network, suddenly some endpoints would refuse to establish calls made over SIP or H323. Error codes would be:

channel unacceptable
487 / Request Terminated
480 / Temporarily Not Available

Not particularly helpful information. After all the packet sniffing and diagnostic checks, it appeared packets were being rewritten or dropped between the endpoint and our VCS. The culprit here is Juniper's 'ALG' for SIP and H323 traffic. I don't have the time not to figure out exactly why or how it was breaking my calls, but i do know that in my case, it wasn't needed. Disable the SIP and H323 ALG's (in the SRX web UI under Configure -> security -> ALG. The endpoints will start working again after your reboot them and allow them to re-register to your gatekeeper.

Monitoring Flexlm usage with Zabbix

2010-11-18T14:16:10-08:00

Flexlm license usage can be a hard thing to accurately measure and monitor. Luckily, the output from lmutil can be used to display license usage. First thing to do is locate the lmutil binary and try to get it to run. I use Flexlm for autodesk on Windows and ESRI on linux(centOS).

Windows: This server seemed to want the license file specified. You may need to launch the lmtools GUI tool and watch the status bar for the path to it. This is what mine looked like:

"c:\program files\autodesk network license manager\lmutil" lmstat -c "c:\program files\autodesk network license manager\license\license.lic" -a

Linux:

/home/esri/arcgis/license10.0/bin/lmutil lmstat -a

That will spit out the current license usage for all your products. To isolate the numbers we want to monitor, we will be piping the output into some other commands like find and cut. I recommend using the GNU tools for windows http://gnuwin32.sourceforge.net/. These examples use the built-in 'find' utility, and a 'cut.exe' tool i found googling around. I had overlooked the gnutools when I first set this up. My mistake can be your gain. With gnutools, it will be easy to isolate the output you want using 'grep' and 'cut'. First, isolate the line you are looking for with grep. You will have to figure out what feature code you want to monitor. mine looks like this:

Windows:

"c:\program files\autodesk network license manager\lmutil" lmstat -c "c:\program files\autodesk network license manager\license\license.lic" -a | find /i "64300acd_f:"

Linux:

/home/esri/arcgis/license10.0/bin/lmutil lmstat -a | grep ARC/INFO

now that you have the right line, you can trim the extra characters with 'cut'. It will take some experimentation to get it right.:

Windows:

"c:\program files\autodesk network license manager\lmutil" lmstat -c "c:\program files\autodesk network license manager\license\license.lic" -a | find /i "64300acd_f:" | cut -c 62-64

Linux:

/home/esri/arcgis/license10.0/bin/lmutil lmstat -a | grep ARC/INFO | cut -c 59-61

The output now should be only the number of licenses being used.

To allow zabbix to monitor this value, we need to create a 'UserParameter' read up on it here: http://www.zabbix.com/documentation/1.8/manual/config/user_parameters. This is what the parameter looks like on my servers:

Windows:

UserParameter=licenses.autocad.used,"c:\program files\autodesk network license manager\lmutil" lmstat -c "c:\program files\autodesk network license manager\license\license.lic" -a | find /i "64300acd_f:" | cut -c 62-64

Linux:

UserParameter=licenses.arcinfo.used,/home/esri/arcgis/license10.0/bin/lmutil lmstat -a | grep ARC/INFO | cut -c 59-61

Once you have the userparameter created on the license server, restart the agent and add an item in zabbix to begin collecting values.

View string including special characters like newlines in Python

2010-10-19T15:28:04-07:00

After spending a frustrating hour or two trying to format a string, i stumbled on python's repr() funcion. repr() allows you to inspect a string object when you're troubleshooting your code. For example:

I was trying to understand a string that was shown in the terminal like this:

print mystring




Node - 192.168.1.104
ERROR:
Description = Generic failure

Seems pretty simple to remove the extra newline characters right? no. doing a string.replace("\n", " ") did not fix it, but actually mangled the string. So here comes repr() to the rescue...

print repr(mystring)




'Node - 192.168.1.104\r\r\nERROR:\r\r\nDescription = Generic failure\r\r\n'

\r\r\n... i've never heard of it before, but knowing that it's there, i could easily change my replace to:

print mystring.replace("\r\r\n", " ")




Node - 192.168.1.104 ERROR: Description = Generic failure

'Pstools: Access Denied in a Domain Environment'

2010-09-15T14:53:10-07:00

After upgrading to a windows 7 VM at work, I was having trouble getting pstools commands to authenticate on remote machines. After much trial and error, I realized some curious behaviour with psexec. Obviously, when connecting to a remote machine, I would try to use the '-u' switch to specify my administrative account, but would always get 'access is denied'. Of course, all of the normal things should be checked: simple sharing turned off, $ADMIN share working... you know...

The issue was apparently that if I log into my workstation as a non-administrative user, but try to issue pstools commands as an administrator, it fails because Microsoft wants me to log into my workstation and work logged with my domain admin account.

Take a look at this example using psexec: On windows 7, running 'cmd' as your non-admin user, if you type in the command:

{% codeblock lang:text %} c:\Program Files (x86)\PsTools>psexec -u domain\domainadmin \targetmachine cmd

Password: Could not start PsExec service on targetmachine: Access is denied.

If you:

Hit start
Type "cmd"
Hold down 'shift' and right-click on the 'cmd' in the start menu
Select 'run as different user'.
Type in your administrative credentials. Use the same ones you will use in the psexec command.

Now you should have your command line window open. If you run the same command as earlier:

{% codeblock lang:text %} c:\Program Files (x86)\PsTools>psexec -u domain\domainadmin \targetmachine cmd

Password:

Now you're in!

ntpq timed out on freeBSD

2010-09-07T13:42:38-07:00

I was running ntpd locally on a server and wanted to check in on ntpd's status. ntpq -p was not producing any output even though ntpd was running.

# ntpq -p
localhost: timed out, nothing received
***Request timed out

This is because I had the option "restrict default ignore" set in /etc/ntp.conf. ntpq. This makes ntpd ignore EVERYTHING, even queries to the loopback interface. Ntpq queries ntpd over the loopback interface at 127.0.0.1. To allow these local queries, add:

 restrict 127.0.0.1

to /etc/ntp.conf, then also add restrict lines for your other upstream ntp servers.

restart ntpd:

/etc/rc.d/ntpd restart

Now ntpq -p will show you status of it's peers

# ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 dns3.untangle.c .INIT.          16 u    -   64    0    0.000    0.000   0.000
 mirror          .INIT.          16 u    -   64    0    0.000    0.000   0.000
 153.16.4.133    .INIT.          16 u    -   64    0    0.000    0.000   0.000

Cannot delete ZFS snapshot

2010-08-23T15:14:09-07:00

While Deleting to delete a snapshot, i typed:

zpool destroy zfsdiskpool@Sat

and got:

cannot open 'zfsdiskpool@Sat': invalid character '@' in pool name

This is because to delete snapshots in ZFS, you should use 'zfs destroy' instead of 'zpool destroy'. DOH!

zfs destroy zfsdiskpool@Sat

Script to run handbrake recursively through a folder tree

2010-08-10T10:22:29-07:00

Handbrake is a fantastic tool for (among other uses), converting videos into mobile formats. It is extremely easy to use and can usually get a video properly converted in just a few clicks. To get Videos encoded for use on my phone, I use the "iPhone & iPod Touch" preset. It will convert your video to .mp4 as well as scale it down to a consumable size. This format will work on any smartphone I've seen.

My issue is that when converting a folder full of videos, I don't want to have to use the gui to add a bunch of videos to the queue one by one. Its a very clicky process. This simple bash script will walk handbrake through all the files in a selected folder (and its subfolders). This uses the find command to traverse recursively through a directory structure. It will place the transcoded file in the same folder as its source and change its extension to ".mp4".

Usage: handbrakefolder.sh [FOLDER] Run handbrake on all the files contained in [FOLDER]. (the current directory by default)

#!/bin/bash
#
# Change this to specify a different handbrake preset. You can list them by running: "HandBrakeCLI --preset-list"
#
PRESET="iPhone & iPod Touch"
if [ -z "$1" ] ; then
    TRANSCODEDIR="."
else
    TRANSCODEDIR="$1"
fi
    find "$TRANSCODEDIR"/* -type f -exec bash -c 'HandBrakeCLI -i "$1" -o "${1%\.*}".mp4 --preset="$PRESET"' __ {} \;

Save that into a .sh file like "handbrakefolder.sh" and grant it the execute permission (chmod +x handbrakefolder.sh).

Thanks to Vinnie and http://mywiki.wooledge.org/UsingFind for their help in this.

Now you can simply execute this script against a folder containing your video files. like this:

For a folder structure like this:
Videos
-> show1
    vid1.avi
-> show2
    -> season 1
        ep1.avi
        ep2.avi
    -> season 2
        ep1.avi
        ep2.avi
-> show3
    vid1.avi
    vid2.avi

hostname% handbrakefolder.sh Videos
...
....
...
When done, it will look like this:
Videos
-> show1
    vid1.avi
    vid1.mp4
-> show2
    -> season 1
        ep1.avi
        ep1.mp4
        ep2.avi
        ep2.mp4
    -> season 2
        ep1.avi
        ep1.mp4
        ep2.avi
        ep2.mp4
-> show3
    vid1.avi
    vid1.mp4
    vid2.avi
    vid2.mp4

Write zero's to disk Infinite loop

2010-07-30T17:05:49-07:00

I was bored... So this is a little bash script i wrote that will write zero's to a hard disk infinately until you press CTRL+C. It uses dcfldd instead of plain dd because i like the progress output it provides.

#!/bin/bash
echo -e "Which disk would you like to wipe out? (sda, sdb, sdc)?   \c"
read DISK
read -p "You picked "$DISK", are you sure? (y/n)" -n 1
if [[ $REPLY =~ ^[Yy]$ ]]
then
C=1
for (( ; ; ))
do
        echo -e "\nStarting zero Sweep number "$C" ..."
        dcfldd if=/dev/zero of=/dev/$DISK
        echo "Zero's written to Disk "$C" time(s)"
        echo "[ hit CTRL+C to stop ]"
        let C=C+1
done
fi

Convert High-Def MKV to play on xbox 360 using Linux

2010-06-29T16:23:32-07:00

There are a million different tutorials out there on how to convert a MKV file into a format that an xbox360 will play. You may have found, like i did, that most of them use 50 different questionable pieces of software to manipulate individual tracks and separate the mkv, etc, etc... I think those are ridiculous.

Please read the FAQ regarding xbox360 file format compatibility. It may help you pick better options for your particular files than the general ones i offer below.

The solution is simple: Use Avidemux.

Install avidemux. To install it is simple. It's in the Ubuntu repositories and I imagine you can also find it in other distros quite easily.
Open avidemux and open the mkv you want to convert. If prompted some garbage about 264 and safemode, just use safe mode and dont worry about it.
Select File -> Properties. This will tell you some info on the formats in use in your video file. Take this opportunity to identify what the xbox doesn't like. When you are finished, click OK. If you want a second opinion, open up the folder with your mkv in it. Right click -> properties -> Audio/Video tab. This will also tell you the video and audio formats.
We obviously know that it wont play a video in a MKV container, so first thing to do is change the "Format" dropdown to say "MP4" (you can, of course use AVI, but the majority of files I run into are h.264 and aac audio. For this combo, you want mp4...)
From the Properties menu, recall the video codec. H264 files show as "AVC1" inside of avidemux. I'm sure theres a technical reason for this, but do you want to talk about it or watch your video? Most of the time, you can leave the video droptown in avidemux on "Copy" this is nice because it means that your processor wont be re-encoding the video. This saves you quality and time.
From the properties menu, recall the audio codec. If you have a video with AAC stereo audio, leave the dropdown on "copy". This is where most of my files need some love. Many MKV's have 5.1 surround audio tracks. This is great, but not for an xbox360. To mix the audio down to stereo, select AAC on the audio dropdown, then click "filters". In the mixer dropdown, select "stereo".
Click "save". Avidemux will prompt you for a filename for the converted file. It does not default a file extension, so do yourself and your xbox a favor and add one yourself like ".mp4".

Once avidemux is finished with your file, it's ready to go.

http://avidemux.sourceforge.net/

Script to run handbrake on an entire folder

2010-06-10T16:33:09-07:00

UPDATED 8-10-10: This script has been superceeded by its newer version at http://www.surlyjake.com/blog/2010/08/10/script-to-run-handbrake-recursively-through-a-folder-tree/. The new version features all of the previous features, but can also traverse recursively through a folder structure.

{% codeblock handbrake.sh %}

!/bin/bash

if [ -z "$1" ] ; then TRANSCODEDIR="." else TRANSCODEDIR="$1" fi for file in "$TRANSCODEDIR"/* do HandBrakeCLI -i "${file}" -o "${file}.mp4" --preset="iPhone & iPod Touch""" done

Save that into a .sh file like "handbrakefolder.sh"

Updating FreeBSD ports nicely using nice

2010-06-08T10:34:22-07:00

Thanks to: http://scottspare.com/bsdfun/?p=75 for pointing me in the right direction. Compiling port updates can take a while and slow down your server. What you can do is use the 'nice' utility to force the processes to a lower priority. This will help your server to run almost normally during an update.

When you use the 'nice' command inside of csh or tcsh, you need to mind that you give the full path to the binary so you dont use the built-in 'nice' command.

# /usr/bin/nice -n 10 {your update command}

What i use is:

# /usr/bin/nice -n 10 portupgrade -aRrP

Man page for 'nice': http://www.freebsd.org/cgi/man.cgi?query=nice&apropos=0&sektion=0&manpath=FreeBSD+8.0-RELEASE&format=html

Freebsd-update fetch interupted

2010-06-07T15:38:35-07:00

I lost power while freebsd-update fetch was downloading patches up to 8.0-Release-p3. When the server came back online, I ran a freebsd-update fetch again and it still found two files that needed updating. after running freebsd-update install, got this error:

# freebsd-update install
Installing updates...gunzip: (stdin): unexpected end of file
gunzip: (stdin): unexpected end of file
 done.

I guess a patch archive was corrupted. no problem though, just remove the download:

# rm -r /var/db/freebsd-update/files

now you can run

# freebsd-update fetch
Looking up update.FreeBSD.org mirrors... 3 mirrors found.
Fetching metadata signature for 8.0-RELEASE from update5.freebsd.org... done.
Fetching metadata index... done.
Fetching 2 metadata files... done.
Inspecting system... done.
Preparing to download files... done.
Fetching 2 files... done.

The following files will be updated as part of updating to 8.0-RELEASE-p3:
/boot/kernel/kernel.symbols
/boot/kernel/nfsclient.ko.symbols
# freebsd-update install
Installing updates... done.

no problems

'ZFS Cant rm: No space left on device'

2010-01-19T17:52:42-08:00

If you completely fill up a zfs pool, it wont let you delete files on it. What you CAN do, is pick a scapegoat file to wipe out or remove a snapshot. Then you will be able to use the rm command. what I did:

# df -h
Filesystem     Size    Used   Avail Capacity  Mounted on
themirror       39G     39G      0B   100%    /home/jacob/themirror
# rm 3gfile
rm: 3gfile: No space left on device
# dd if=/dev/null of=3gfile
0+0 records in
0+0 records out
0 bytes transferred in 0.000046 secs (0 bytes/sec)
# rm 3gfile

FreeBSD error in /boot/loader.conf

2009-12-23T16:29:41-08:00

While experimenting with ZFS in FreeBSD, I made some tweaks to the vk.kmem_size variable in /boot/loader.conf. when setting it like this:

#/boot/loader.conf
vm.kmem_size="1024"

everything worked, but I wanted to see what would happen if i doubled it. Unfortunately, setting vm.kmem_size to 2048 kept the FreeBSD kernel from booting. At startup it would just do this:

panic: kmem_suballoc: bad status return of 3
cpuid = 0

To fix the erroneous variable setting, I had to :

Reboot.
Wait for the FreeBSD Boot menu. (the screen that lists Default, ACPI disabled, safe, and single user modes)
Press 6 to select "Escape to loader prompt"
At the loader prompt, type "show". This will provide all the default variable settings. press the spacebar to page down. In my case, at the end, the incorrect variable was: vm.kmem_size="2048".
to switch it back and allow the system to boot, type

set (variable)=(correct value)

in my case this was:

set vm.kmem_size=1024M

type

boot

When you are finished with all that fun, you should edit the /boot/loader.conf file back so you don't have to do this again. Thanks to "crsd" from the FreeBSD IRC channel for the help.

Debian eth0, eth1, eth2, in Virtualbox or VMware Virtual machines when copying

2009-11-13T14:44:17-08:00

Debian uses udev. Udev handles mapping MAC's to the appropriate /dev/eth(X) file. If you copy a Virtual machine, Udev will remember the MAC address of the old NIC. When you copy the machine, the virtual host usually generates a new MAC address for the VM. Udev will assign the new Device to eth1, eth2, and so on. If you want to change your NIC assignments make Udev forget the old MAC. In Debian 5 (lenny) it is in this file:

/etc/udev/rules.d/70-persistent-net.rules

In Debian 4 (etch) it is in this file:

/etc/udev/rules.d/z25_persistent-net.rules

To apply changes in Lenny: "udevadm trigger" or "udevtrigger" (in Etch)

Karmic install cannot login (gdm freezes) Nvidia

2009-11-03T12:13:46-08:00

On a fresh install of karmic, using an Nvidia 7800 GT. When I went to click on my username to log in, some artifacts would appear on the screen, then the system would freeze. hard. Couldnt even ctrl+alt to another console. Installing the Nvidia drivers fixed the issue. To fix it:

Do a hard reset on the machine
When you arrive at the GDM screen, DO NOT CLICK ON ANYTHING.
press CTRL + ALT + F5. Your screen will switch over to a text console.
Log in.
Install the Nvidia glx driver
```
sudo aptitude install nvidia-glx-new
```
Reboot the machine.
```
sudo reboot
```

Windows 32 (x86) or 64 (AMD64) detection in batch files

2009-10-20T16:53:04-07:00

While there are a lot of ways to detect for a 64 bit version of windows. you can test for %programFiles(x86)%, but handling the output and writing the IF comparisons is messy.

In Batch files, you can easily check for architecture by using the "processor_architecture" variable. x86 versions of windows will have this set to "x86", and x64 versions "x64". Here's an easy example:

@echo off
IF %processor_architecture% == AMD64 echo This is a 64-bit version of windows
IF %processor_architecture% == x86 echo this is a 32-bit version of windows.
pause

Uninstall ALL Versions of WinZip Batch Script

2009-10-06T15:43:34-07:00

I have been struggling with this for quite a while, and ran into a lot of issues with all of the different versions on our network. There are people with 100 line .vbs scripts to do this, and I think this is much simpler:

@echo off
REM
REM
REM    Detects winzip and silently uninstalls
REM    There is a un-stoppable pop-up window. so this will kill
REM    all instances of  IE after the UN-installation.
REM
REM
REM
taskkill /F /IM wzqkpick.exe
wmic product where "name like '%%winzip%%'" call Uninstall
REM
REM This section is for WinZip <12
REM
If Exist "%programfiles(x86)%\Winzip\winzip32.exe" GOTO 64
If Exist "%programfiles%\Winzip\winzip32.exe" "%programfiles%\WinZip\Winzip32.exe" /Uninstallx
taskkill /F /IM iexplore.exe
GOTO :END
:64
"%programfiles(x86)%\WinZip\Winzip32.exe" /Uninstallx
taskkill /F /IM iexplore.exe
:END

For getting rid of WinZip 12, the magic happens in the 'wmic' line. It searches all of the installed product's names for "winzip" and will uninstall anything it finds. This only works for products that use MSI's. If you ran this outside of a batch file, you will have to use a single % percentage sign instead of the %% double percentage sign... Windows scripting sucks, and that's just the way it is. Same thing that happens in "FOR" loops.

I never knew anything about the Windows management instrumentation control (wmic) until now, but I will be sure to exploit it's features.

Tethered Blackberry Modem on Ubuntu Karmic

2009-09-11T17:54:43-07:00

Using your blackberry as a 3G (in my case) modem proved to be Incredibly simple. There are five million different methods to do this and many have you compiling stuff and editing text files. I love those sorts of things, but this is much easier. I Wish that I had tried this sooner.

Confirmed using Blackberry Tour (9630) on Verizon.

What we will do is use a program called 'barry'. http://www.netdirect.ca/software/packages/barry All we need to do is install some packages from the repository below. These commands are from the terminal, so launch it from "Applications -> Accessories -> Terminal" in the Menu.

Edit /etc/apt/sources.list

sudo nano /etc/apt/sources.list

add these lines:

deb http://ppa.launchpad.net/doctormo/barry-snapshot/ubuntu karmic main
deb-src http://ppa.launchpad.net/doctormo/barry-snapshot/ubuntu karmic main

install the repositories key.

sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 113659DF

Now update your package list

sudo aptitude update

and install the programs:

sudo aptitude install libbarry0 barry-util barrybackup-gui

Thats it! now to create the connection, the command will vary depending on your provider. The ones they have preconfigured are: ATT Cingular, China Mobile, KPN Nederland, O2 Ireland, Rogers, Sprint, T-Mobile US, Verizon. the syntax is : sudo pppd call barry-{provider}

for Verizon:

sudo pppd call barry-verizon

You'll see a bunch of output, and it'll tell you that the connection is up. to close it down, press (CTRL+C). Make it pretty by creating an application launcher on your desktop.

Compiz desktop in Debian Squeeze

2009-08-19T11:32:00-07:00

It has come to my attention that this post is out of date. Since Squeeze's release, I have not been able to update it. Your best bet to get compiz working correctly would be to follow to the directions in the Debian wiki: http://wiki.debian.org/Compiz

Getting compiz running on squeeze is surprising easy. Debian has included a document to help set it up, but there are a few little things necessary to get it to be usable.

Get the packages installed.

apt-get install compiz compizconfig-settings-manager compiz-fusion-plugins-main

Follow the instructions here: http://wiki.debian.org/Compiz to customize your /etc/X11/xorg.conf file
Restart gdm
```
/etc/init.d/gdm restart
```
open up a terminal and type
```
compiz --replace
```
If a went well, you should see everything flash around as it redraws your desktop.
My window decorations (title bars) dissapeared when I enabled compiz. Additionally, i couldn't drag windows by clicking on the title bar. To fix this: go to
1. System -> Preferences -> Compizconfig Settings Manager
2. Scroll down to the "Effects" section and enable "Window decoration"
3. Scroll down to "Window Management" and enable "Move Window"
If you want to use the desktop cube and/or workspaces, you will need more than the single or dual workspaces provided by default.
1. System -> Preferences -> Compizconfig Settings Manager
2. Click "General options"
3. Select the "Desktop size" tab.
4. Set the "Horizontal Virtual size" to whatever you want. For a cube... 4
If you want compiz to start automatically with GNOME:

Add "compiz --replace" to "Desktop -> Preferences -> Sessions -> Startup Programs"

'Vmware - Unable to change virtual machine power state: Internal error.'

2009-08-14T08:43:23-07:00

Ran into this while running Vmware Workstation under Ubuntu Jaunty. I got an error while shutting down the machine through an NX session.

This is a result of a zombie 'vmware-vmx' process running. All you need to do is kill the process. This command sends 'signal 9' to the process. When sent to a program, SIGKILL causes it to terminate immediately. In contrast to SIGTERM and SIGINT, this signal cannot be caught or ignored. For more information: more sigkill info.

killall -s9 vmware-vmx

After that, I was able to start up the virtual machine without issue.

Linux command line bandwidth monitor

2009-07-29T16:50:48-07:00

You can find out how much bandwidth your linux machine is using with a simple tool called "bwm-ng". In Debian, install it with

aptitude install bwm-ng

Then, just type 'bwm-ng' in the command line. It will give you something like this:

bwm-ng v0.6 (probing every 5.000s), press 'h' for help
input: /proc/net/dev type: rate
-         iface                   Rx                   Tx                Total
==============================================================================
             lo:           0.00 KB/s            0.00 KB/s            0.00 KB/s
           eth0:        2221.47 KB/s           48.13 KB/s         2269.60 KB/s
------------------------------------------------------------------------------
          total:        2221.47 KB/s           48.13 KB/s         2269.60 KB/s

Pressing the "h" key while it is running wil actually pull up a nice menu to change some of the options you are looking at.

bwm-ng is very basic, "iptraf" is another tool that provides some more functionality if you want to drill further into what is moving in and out of your box.

Batch extract a directory full of split RAR archives "part1", "part2"

2009-07-26T18:18:10-07:00

If for some reason you have a directory full of RAR archives, and each is split into parts, extracting them individually can be a pain. If you just type "unrar *" into the terminal, you will extract each file multiple times depending on how many "parts" it has. First, install the "unrar" package:

sudo aptitude install unrar

then cd to the directory with all of your RAR archives. Then type:

unrar e '*.part1.rar'

make sure to keep the single quotes. If the rar is password protected, add the "-p" switch followed by the password without any spaces.

unrar e -p<password> '*.part1.rar'

In Ubuntu Jaunty, shift, ctrl,alt, and caps lock buttons stop working

2009-07-07T16:45:03-07:00

This appears to be related to vmware, to correct: open up the terminal, and type:

setxkbmap.

Thats it.

Pidgin OSD pop ups in Ubuntu Jaunty

2009-07-07T10:17:52-07:00

You can remove or configure the Pidgin pop up's that you see from pidgin. Buddies logging on/off, and in my case, new messages. I didn't want every new message i receive to be blasted up on the desktop OSD. To control what you see there, Pull up your Pidgin window (buddy window, not your conversation box), then: Tools -> Plugins -> Libnotify Popups. When you have selected the libnotify pop-ups, click on "configure plug-in". That will give you some check boxes to customize the behaviour

Xbox Media Center on original XBox

2009-06-08T21:23:13-07:00

I picked up a refurbished original XBox specifically so i could install XBMC on it. Turns out it is a GREAT project, and through some trickery, was able to get it installed and running thanks to this post: http://www.boards.ie/vbulletin/showthread.php?t=2055256404 by 'SuperCell' at boards.ie. I highly recommend using this 'hotswap' method if you don't already have the gear necesarry to mod an xbox with a chip, or by using a savegame exploit. It'll save you some cash, plus who doesn't like yanking an IDE cable from a spinning hard drive?

Best of all, all of the tools used in the process run perfectly in WINE, or directly on my Linux desktop.

Installing Skype on eee 1000 running Ubuntu 9.04 Jaunty

2009-05-23T11:21:44-07:00

I have an eee 1000 netbook running Ubuntu Jaunty netbook remix. Installing and configuring Skype was pretty simple:

Add the Medibuntu repository:

sudo wget http://www.medibuntu.org/sources.list.d/jaunty.list --output-document=/etc/apt/sources.list.d/medibuntu.list
sudo apt-get update
sudo apt-get install medibuntu-keyring
sudo apt-get update

Install skype:

sudo aptitude install skype

Video worked out of the box. To configure sound, got to Options -> Sound Devices. Set: Sound In: HDA Intel (hw:Intel,0) Sound Out: Pulse Ringing: Pulse

Install backupexec 12.5 agent on Debian Lenny

2009-05-12T15:24:16-07:00

The path to it is

BEWS_12.5.2213_LINUX-UNIX-MAC-SAP_AGENTS.tar.gz\BEWS_12.5.2213_LINUX-UNIX-MAC-SAP_AGENTS.tar\pkgs\Linux\VRTSralus.tar.gz\VRTSralus.tar\VRTSralus-12.5.2213-0.i386.deb

Unfortunately, the Deb package doesn't specify It's dependencies. You need to manually install the libstdc++5 package.

aptitude install libstdc++5

You can follow the rest of the install here: http://insanelabs.com/debian/linux-symantec-backup-exec-12-and-debian-etch-3264/

Group share for a Active Directory domain group with Samba

2009-05-06T15:38:36-07:00

Once you have joined your samba server to your domain and have set up user authentication, creating a share that will allow a group of users to access files is pretty simple. We need to create a folder that is owned by the root user and the domain group. The samba share will inherit permissions from the parent folder. Here is an example for creating a folder for a domain group called "corporate_HR". The domain is called "acme". There is administrative support group called "admins". Setting the folder's permissions to "2771" means that the owner (root) and the group (corporate_HR) have full access. The "2" is a setgid bit. It will force all new files created under the HR folder to take the group parameter from it's parent. We use the "force group" parameter to ensure that permissions are set properly when a member of the "ACME\admins" group accesses the files.

to set up the filesystem:

cd /home
mkdir HR
chgrp corporate_HR HR
chmod 2771 HR

set up the share:

[HR]
comment     = share for corporate HR group
readonly    = no
inherrit owner  = yes
inherit permissions = yes
authorized users    = @ACME\corporate_HR @ACME\admins
force group = ACME\corporate_HR

Join Debian Lenny to Active Directory using Samba

2009-05-01T15:48:45-07:00

If you install GNOME on Debian and intend to use it as a server, first thing you should do is remove the network-manager and network-manager-gnome packages. This tool can be helpful on a desktop, but since we are tweaking DNS, IP addresses, and perhaps even NIC bonding, the network-manager can screw up your settings.

aptitude purge network-manager-gnome network-manager

If this is a VMware Virtual machine, install VMware tools. The prerequisites for Debian Lenny are:

aptitude install build-essential openssl libssl-dev linux-headers-$(uname -r) psmiscapt

You may have to manually create a DNS entry for your server. Just to be safe, do that now.

If you plan to support NTFS-like ACL's, install support for it now.

aptitude install acl

now, you need to mount the partition with ACL's enabled, to do this, edit your /etc/fstab.

nano /etc/fstab

The line I needed looked like this:

/dev/sda9 / ext3 acl,defaults 0 1

The important part is to get "acl" into the options list. Dont forget the comma. I recommend rebooting just to make sure your fstab is set before you continue. It's easy to screw something up here and make your system unbootable, so if you reboot, and cant get your system back up and running, get into single user mode, and use this command. (You will have to adjust /dev/sda1 to your root partition. In my experience, it is usually /dev/sda1 or /dev/hda1. Use fdisk -l to list the partitions on the disk.

mount -o remount,rw /dev/sda1

Now lets get the samba and kerberos software packages installed.

aptitude install samba ntpdate smbclient winbind krb5-config krb5-user

Put in the Workgroup/domain info when prompted if you like. They will set up some .conf files for you, but we're not going to use them anyways. We wont worry about WINS for now. Now lets stop samba and winbind:

/etc/init.d/samba stop
/etc/init.d/winbind stop

move the /etc/samba files somewhere

mkdir /etc/samba/vanillaconf
mv /etc/samba/* /etc/samba/vanillaconf/

create your own /etc/samba/smb.conf: Make sure that the domain and realm fields are all caps.

[global]
workgroup = ((DOMAIN)) 
server string = %h server
wins support = no 
security = ads 
realm = ((DOMAIN)).COM 
encrypt passwords = yes 
obey pam restrictions = yes
winbind use default domain = yes
winbind enum users = yes
winbind enum groups = yes
template shell = /bin/bash
idmap uid = 10000-20000
idmap gid = 10000-20000

sync time

ntpdate ((domain controller))

now that samba knows what domain its part of, and we have a basic config for it, we need to get the authentication part working. Kerberos is the softare that communicates with your Active Directory domain to authenticate users.

nano /etc/krb5.conf

configure domain in /etc/krb5.conf. important parts:

[libdefaults]
default_realm = ((domain)).COM
[realms]
((DOMAIN.COM)) = { 
    kdc = ((domain controller)) 
    kdc = ((backup DC)) 
    admin_server = (domain controller)) 
}
[domain_realm]
.(domain).com = (domain controller).(domain).COM
{% endcodeblock %}

do a test by typing:

kinit [username]

then type:

klist

if you get some output looking like this, you are ready to go.

Ticket cache: FILE:/tmp/krb5cc_0 Default principal: (username)@(domain).COM Valid starting Expires Service principal 04/27/09 13:54:23 04/27/09 23:54:26 krbtgt/(domain).COM@(domain).COM renew until 04/27/09 23:54:23

To get Debian to recognize your Active Directory users, you need to update /etc/nsswitch.conf

nano /etc/nsswitch.conf

add "winbind" to the passwd and group lines. if you have "compat" in the line, put in like this:

passwd: compat winbind
passwd_compat: winbind
group: compat winbind
group_compat: winbind

Now start samba and winbind back up

/etc/init.d/winbind start
/etc/init.d/samba start

Join your machine to the domain.

net ads join -U (administrative user)

You should see

Joined '(server name)' to realm '(domain).com'

If you see

No DNS domain configured for '(servername)'. Unable to perform DNS Update. DNS update failed!

Don't worry. Just create an entry in your DNS server for the samba machine. Test if winbind is working properly with

wbinfo -t

If you get

checking the trust secret via RPC calls failed Could not check secret

Then restart winbind:

/etc/init.d/winbind restart

allow AD accounts to logon to the machine:

/etc/pam.d/common-account

account sufficient pam_winbind.so
account required pam_unix.so

/etc/pam.d/common-auth

auth sufficient pam_winbind.so
auth required pam_unix.so
use_first_pass nullok_secure

/etc/pam.d/common-session

session required pam_mkhomedir.so skel=/etc/skel/ umask=0066
session sufficient pam_winbind.so
session required pam_unix.so

It is critical that "obey pam restrictions" is set to "yes". in your smb.conf for these pam settings to take effect.There are some other changes to these Pam.d settings i decided to make upon putting the server into production. I will write about that in another page [link].

Now, give your admin group from AD root access to the box

aptitude install sudo visudo

add this line:

%(domain)\\(admin's group) ALL =(ALL) ALL

When creating shares, to give an AD user access to a share, use:(More info in the future)

valid users = (domain)\username

to give an AD group access to a share, use:

valid users = @(domain)\groupname

Sometimes, I have to reboot my windows client machine to make it work... fyi. Sources:

O'reilly's Samba book 3rd Edition.
http://www.ccs.neu.edu/home/battista/documentation/winbind/index.html
...Countless articles, mailing lists, forums.

How to configure openDNS on pfsense firewall

2009-04-26T10:24:23-07:00

Create your openDNS.com account. Add your Network to their system, etc.
Go to the "System" -> "General setup" page.
Set the DNS address to the OpenDNS':
```
208.67.222.222
208.67.220.220
```

Now just make sure that "Allow DNS server list to be overridden by DHCP/PPP on WAN" checkbox is NOT checked.

Rip and convert to MP3 files in Debian Lenny

2009-03-23T21:53:45-07:00

To create an MP3 audio file in Lenny, you need to install the proper package to encode them. I suggest you add the "debian-multimedia" repository while you are at it. Edit your /etc/apt/sources.list and add:

deb http://www.debian-multimedia.org lenny main

run

aptitude update

then install the gstreamer lame encoder:

aptitude install gstreamer0.10-lame

Then you are done. You can launch soundconverter and convert formats to MP3 using VBR and other technologies. You can also use sound juicer to Rip CD's

How to update my sparc FreeBSD install

2009-03-18T10:43:07-07:00

So the spanking new freebsd-update doesn't work for sparc systems... boo. Have to do everything manually. Here's what I remember. Let me know if I forgot something, or its wrong.(very likely)

Install cvsup:

# cd /usr/ports/net/cvsup-without-gui/
# make install

install fastest cvsup:

# cd /usr/ports/sysutils/fastest_cvsup/
# make install
# fastest_cvsup -c us

Replace "us" with your country if it's different. copy the default supfile to our own before editing.

# cp /usr/share/examples/cvsup/stable-supfile /etc/stable-supfile
# cp /usr/share/examples/cvsup/ports-supfile /etc/ports-supfile

edit both of them and change the

*default host=

to the fastest cvsup server from the "fastest cvsup"

then update the ports

# cvsup -L2 -g /etc/ports-supfile
# cvsup -L2 -g /etc/stable-supfile

Then i guess we try to build the world. see if something breaky.

# cd /usr/src
# make buildworld

Recompiled kernel. I have my own kernel called "MYSPARC2". Yours might just be "GENERIC"

# cd /usr/src
# make buildkernel KERNCONF=MYSPARC2
# make installkernel KERNCONF=MYSPARC2

rebooted into single user mode then:

# mount -u /
# mount -a -t ufs
# swapon -a

Now you are ready to install the new world.

# cd /usr/src
# make installworld

After that was done, mergemaster to update your configs:

mergemaster -v

Then used portmaster to upgrade all of the ports.

# portmaster -a

Basically everything i know about BSD comes from or stems from something I've read at freebsdmadeeasy.com

source: http://www.freebsdmadeeasy.com/tutorials/freebsd/updating-freebsd-with-cvsup.php

Automysqlbackup.sh.2.5 on FreeBSD

2009-03-13T13:38:27-07:00

Automysqlbackup is a great tool. It's setup and configuration can be found at many other sites. When you try to run it out of the box, you will get something like this:

./automysqlbackup.sh.2.5: Command not found.

To run it in FreeBSD, you need to edit a single line. Automysqlbackup.sh.2.5 is a bash script, and it references the path '/bin/sh'. In freebsd, bash is installed at '/usr/local/bin/bash.' If you haven't already, install bash:

pkg_add -r bash

Then all you have to do is edit automysqlbackup.sh.2.5. Look for

#!/bin/bash foo

and change it to

#!/usr/local/bin/bash

Now you should be able to launch it properly by running './automysqlbackup.sh.2.5'. creating a symlink would probably work, but i like this better.

Install the proprietary Closed source version of Virtualbox on Debian Lenny

2009-02-17T11:21:28-08:00

I like the closed source version of virtualbox. I just do. This is how to install it quickly and simply:

add this to sources.list

deb http://download.virtualbox.org/virtualbox/debian lenny non-free

run this to get the public key and install it:

wget -q http://download.virtualbox.org/virtualbox/<span class="searchword0">deb</span>ian/sun_vbox.asc -O- | sudo apt-key add -

then just install it

aptitude install virtualbox-2.1

this will automatically create the group "vboxusers" and add the user who installed it to the group. If you need to add others:

sudo adduser `[username]` vboxusers

Install Debian Lenny Nvidia Drivers using binary packages

2009-02-16T21:30:40-08:00

As usual, the official Debian wiki has a great writeup on getting the Drivers installed. http://wiki.debian.org/NvidiaGraphicsDrivers. Do yourself a favor and read the document. It will give you the full story and not just the regurgitated short version below. I use debian because I like the packaging system. I think it works well. why add the agony of recompiling drivers whenever you update kernels and whatnot. Since X was broken, The key was to switch to a console session BEFORE logging in using GDM. As soon as the login page comes up, press CTRL+ALT+F2. This is what I had to be able to log in and use my X session:

I used the netinst disk and had to add the "non-free" and "contrib" repositories. to do this, edit /etc/apt/sources.list.

nano /etc/apt/sources.list

Find the line line that has something like

deb http://mirror.cc.columbia.edu/pub/linux/debian/debian/ lenny main
deb-src http://mirror.cc.columbia.edu/pub/linux/debian/debian/ lenny main

add "contrib non-free" to the end of both of these lines and then do a "aptitude update"

# apt-get install module-assistant nvidia-kernel-common
# m-a auto-install nvidia-kernel${VERSION}-source
# apt-get install nvidia-glx${VERSION}

In the device section, add

Driver "nvidia"

Restart GDM (and subsequently X)

# invoke-rc.d gdm restart

Install The nvidia settings interface thingy if you want. This can also verify if the driver is actually running.

aptitude install nvidia-settings

In debian, the package doesn't install a shortcut for some reason, so just launch it by typing "nvidia-settings" into the console.

Add a Single User mode entry in Grub bootloader.

2009-02-15T17:20:58-08:00

All you need to do is add 'single' to the end of any bootloader entry to make it single user. So what you need to do is edit the /boot/grub/menu.lst, copy one of the entries, and all 'single' to the end.

gksudo kate /boot/grub/menu.lst

take the 'kernel' line and add 'single' to the end. Like this:

kernel      /boot/vmlinuz-2.6.27-11-generic root=UUID=3699a36a-dd96-4186-8028-43776214c01e ro quiet splash single

Set Windows Environment variables through registry.

2009-02-11T13:35:58-08:00

The system variables are in there, so you can change and add to them permanently.

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Environment

Securing and configuring HP switches

2009-01-27T13:20:28-08:00

We got some Procurve 2810 Gigabit switches. they play some critical roles, so I wanted to keep others from playing with them. The exact code below would enable https for web management, enable ssh for console management, change the port ssh runs on, and disable telnet and unencrypted web management.

config
crypto key generate ssh
show crypto host-public-key
crypto key generate cert 512
crypto host-cert generate self-signed 01/01/2009 01/01/2020 [ip address] [department] [company] Chicago IL us
ip ssh
no telnet-server
no web-management plaintext
ip ssh port [port num]
web management ssl
write memory

OpenOffice Remove mis-spelled words

2009-01-17T11:26:04-08:00

beleive or believe? If you mistakenly add a word to Firefox's custom dictionary, the word is added to

~/.mozilla/firefox/[unique-alphanumeric-string].default/persdict.dat

just edit the file and remove your mistaken 'custom' word.

Remotely change power options in XP

2009-01-12T16:35:21-08:00

First of all, get the pstools and set them up. then run

psexec \\[workstation name] powercfg /query
Field Description          Value
-----------------          -----
Name                       Home/Office Desk
Numerical ID               0
Turn off monitor (AC)      After 20 mins
Turn off monitor (DC)      After 5 mins
Turn off hard disks (AC)   After 60 mins
Turn off hard disks (DC)   After 10 mins
System standby (AC)        Never
System standby (DC)        After 5 mins
System hibernates (AC)     Never
System hibernates (DC)     After 20 mins
Processor Throttle (AC)    Not Supported
Processor Throttle (DC)    Not Supported

that will tell you what the current settings are. After that, you can set the power scheme by doing this:

psexec \\[workstation name] powercfg /setactive "[scheme name]"

It will exit with "error code 0", but no worries. Just check your work using the query command above after your changes.

Ubuntu recovery console with no root password

2009-01-05T21:48:51-08:00

So your Ext3 root partition is corrupted. No matter how you try to log into your system, your system asks you for root password. But you are a good, obedient ubuntu user. And you do not have a root password set. You can press CTRL+D, but that just reboots. Are you boned?

No. problem. press ctrl+alt+delete. You will get a messy screen about having trouble launching your X session, but you will get a login towards the bottom of the screen. So just use your normal user login and password. after that, I use sudo su

Logitech webcam drivers

2008-12-30T15:26:04-08:00

Logitech drivers are packed in with an awful, bloated program. Just want the drivers and nothing else? I did this with a Quickcam 4000 Pro. The same basic steps may apply to other cams:

Download the Program from logitech.com
use 7-zip, winRAR, or some other decent compression program to extract the contents of the .exe to a folder on you computer.
Open up Device Manager. right-click on the unknown USB device, and select "Update Driver".
Point it at the folders you extracted. it should be under a folder appropriately named "drivers". mine was in "pro".
happy happy. you web cam works without the extra bloat.

Command line scp from linux(+BSD) to windows.

2008-12-20T17:37:00-08:00

i traveling and wanted to copy over my logs to find out what was happening on my server. didn't want to install a client like filezilla or another GUI. Using the command line SCP utility, you can grab remote files securely over SSH. You can grab the executable from http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html (pscp.exe is what you want) then, in the command line on the windows client machine, type something like this:

pscp -r -P 22 jimmy@johnscomputer.com:/home/jimmy/stuff "c:\documents and settings\jimmy\Desktop"

The "-r" makes it grab the whole folder, and everything under. the "-P" specifies a port (mostly useful if you aren't using 22. make sure it is capital).

running msc's as admin account

2008-12-16T11:51:11-08:00

Got tired of right-clicking and "run as" and putting in my uname/password. I just updated the shortcuts to run something similar to this. This example is for exchange 2007 server management console.

runas /user:DOMAIN\ADMIN_ACCOUNT "cmd /c \"c:\program files\microsoft\exchange server\bin\exchange management console.msc\""

Conference rooms with outlook and exchange 2007

2008-12-11T09:18:14-08:00

Exchange Server 2007 allows administrators to create objects such as Mailbox, Contacts, Mail User, and Distribution Groups. Here are the steps needed to create a room mailbox object:

Open the Exchange Management Console.
Expand Recipient Configuration.
Click on Mailbox.
In the Mailbox pane, click on New Mailbox...
For our purposes, we will pick "room mailbox" This is a mailbox to be assigned specifically to Meeting Rooms. Its associated user account will be disabled in Active Directory.
In the User Type window, we can choose either to create a new user or to assign an existent user to the new resource mailbox. If we choose to assign it to an existent user, we will have to check if the account does not already have a mailbox associated to it. Click Next to continue.
In the User Information window, we should fill out the user's personal information and select the Organization Unit where it will be created. After that click on Next
On the Mailbox Settings page, we can define the mailbox information such as Alias, Mailbox Server, and Mailbox Store where the new resource mailbox will be located. The policies for Mailbox and ActiveSync can also be defined in this step. We can choose which fields we are going to fill out and then click Next to continue.
In the New Mailbox window, we will get a summary of all the information that we have selected in the previous steps. These parameters will be used by the cmdlet New-Mailbox for the creation of this resource mailbox object. To create the resource mailbox, click on New.
In the Completion window, we will see the cmdlet New-mailbox and the parameters that we used in the creation process of this new resource mailbox.
Grant yourself and users who will manage the mailbox "full access". Right-click on the mailbox you just created and select manage full access permission.
Sign into OWA.
Click the dropdown next to your username on the upper-right corner of the screen and open up the room mailbox you just created.
go to Options, then Resource settings
go through these.
Add the room mailbox to outlook.
go to the calendar, right-click on the room's calendar, go to properties.
Under the permissions tab, set the default permission to "reviewer". This will let everyone see what is going on in the room when they are scheduling

props to: http://www.msexchange.org/articles_tutorials/exchange-server-2007/management-administration/managing-resource-mailboxes-exchange-server-2007-part1.html

Windows XP's /etc/hosts file

2008-12-05T16:48:20-08:00

To my surprise, XP has an equivalent to unix' "/etc/hosts" file. you can find it here: c:\WINDOWS\system32\drivers\etc\hosts'

script to update wordpress

2008-11-26T15:02:37-08:00

Why? Because i'm lazy. I'll update this to see if it works

#!/bin/sh
#your current site will be backed up to your home folder with the date in the name
#change this to whatever you want
BACKUPNAME="wordpress_site_b4_update"
BACKUPDATE=`date +%y%m%d`
#this is the path to your wordpress site's root
SITEROOT="/usr/local/www/"
#here she goez...
cd ~
tar -pczf "$BACKUPNAME""$BACKUPDATE".tar.gz "$SITEROOT"
cd /usr/local
fetch http://wordpress.org/latest.tar.gz
tar -xzf latest.tar.gz
rm latest.tar.gz
cp -R wordpress/* "$SITEROOT"/
rm -R wordpress
chown -R www "$SITEROOT"

FreeBSD time updates with ntpdate and ntpd

2008-11-18T00:53:33-08:00

So FreeBSD's time is out of wack. Two options:

ntpdate. This is manual. To run a time sync manually, just run

ntpdate -b pool.ntp.org

You can replace pool.ntp.org with whatever you want. To tell it to run whenever the system is booted, add this to /etc/rc.conf:

ntpdate_enable="YES"
ntpdate_hosts="[your ntp server]"

This works, but if you don't reboot often, could be unreliable.

ntdp. ntpd just runs all the time and keeps you nsync with internet time servers. You would want to use this for servers that you don't reboot all the time. for this:

create an /etc/ntpd.conf file:

server pool.ntp.org
driftfile /etc/ntp.drift
logfile /var/log/ntpd.log

you can start ntpd manually:

ntpd

and/or add a startup entry for it in /etc/rc.conf:

ntpd_enable="YES"

I recommend setting up both of these on servers. Ntpdate will snap your system to a good point before you begin synchronizing with ntpd. ntpd by itself will take a lot longer to get you synced up.

webalizer installation on freebsd 7

2008-11-13T10:05:54-08:00

cd /usr/ports/www/webalizer
make
make install
make clean
rehash
mkdir /usr/local/www/stats
webalizer -o /usr/local/www/stats /var/log/httpd-access_log

After that, just added it to the root crontab

crontab -e

like this:

0 * * * * /usr/local/bin/webalizer -o /usr/local/www/stats /var/log/httpd-access_log > /var/log/webalizer-hourly

That runs every hour, so that the stats are kept up-to-date.

Debian Samba share. no password. read only

2008-10-08T13:28:24-07:00

Just install samba server:

apt-get install samba

edit the config:

nano /etc/samba/smb.conf

set security to "share"

security = share

and guest account to nobody

guest account = nobody

then you just need to create your share like this:

[guest share]
comment = a guest share
path = /path/to/files
browseable = yes
read only = yes
guest ok = yes

MySQL table storage engine (switching from MyIsam to InnoDB)

2008-10-02T23:14:56-07:00

I know I did this the hard way. there is a command that you can issue that will convert them on the fly. what I did was mysqldump the database, reinstall mysql(because of other reasons), and then edit the dump so that the line that specified the engine reads "ENGINE=INNODB" instead of "ENGINE=MyISAM".Then I imported the dump.

I was getting annoyed that when listing my tables in phpmyadmin, it would show the list of tables in InnoDB (which is what i wanted), but then, when showing the summary at the bottom of the page, said "MyIsam". I easily got rid of it by adding "default-storage-engine=InnoDB" to my "my.cnf" file.

MySQL server --purge and reinstall

2008-10-02T23:06:03-07:00

Screwed around for a bit. messed up MySQL. I decided to remove --purge, and start fresh. Unfortunately, the reinstall wouldn't work. Long story short, i had to remove "mysql-server, mysql-server-5.0, and mysql-common". when i ran this, it also pulled out other dependencies, but i just reinstalled them all. voila! worky.

apt gives a "Segmentation faulty tree"

2008-10-01T20:59:08-07:00

when running an update:

sudo aptitude
Ouch!  Got SIGSEGV, dying..

apparently, just remove some files:

sudo rm /var/cache/apt/*.bin

now everything is back to normal.

Zabbix 1.6 Proxy with Debian

2008-10-01T14:10:21-07:00

I set up a Zabbix Proxy today on a little VM. looks to work nicely. Intructions will be kept up to date on this page.

Staying on time in Debian

2008-09-30T12:41:54-07:00

To stay on time, you should really use ntpd. It is a daemon that runs all the time and makes tiny changes to the system clock. All you need to do is install ntp:

aptitude install ntp

after it's installed, I encourage you to go over to http://www.pool.ntp.org/zone/@ and pick a pool in your continental zone. Debian has some defaults, and they are fine, i just like using the ntp pool. If you want to change where you are syncing to, open up /etc/ntp.conf and change the "server" lines.

nano /etc/ntp.conf

to manually sync Debian's time, just install "ntpdate"

apt-get install ntpdate

run this:

ntupdate-debian [a ntp server]

Uninstall MySQL 5 from Debian

2008-09-27T09:12:32-07:00

I completely. completely messed up MySQL. Badly. I wiped out the data directory (/var/lib/mysql in Debian) and the went to remove mysql ... apt-get remove --purge mysql-server That only deleted something like 86K. dissapointing. It looks like "mysql-server" is a shortcut pointing to the latest version of the software. The real software is found under the name "mysql-server-50". So what made it happen was: apt-get remove --purge mysql-server* That'll wipe out whatever mysql install is in place.

savecore? reboot after panic?

2008-09-27T08:56:07-07:00

scary. I updated freebsd, and i can only think that i messed it up when i tried to "shutdown -h" to get it shutdown so i could reboot. anyways, when i rebooted, i saw this message. apparently there was a kernel dump on my swap drive. But problem is that i didn't have enough room on /var/crash to make it happen. i redirected the commands ot the /usr/home/ program.

http://www.freebsd.org/doc/en/books/developers-handbook/kerneldebug.html

fsck -p
mount -a -t ufs
make sure /var/crash is writable
savecore /var/crash /dev/ad0b
exit
exit to multi-user

FreeBSD MYSQL configuration

2008-09-26T18:27:49-07:00

FreeBSD does not automatically include a default my.cnf for MySQL. Instead, a set of sample configuration files can be found under /usr/local/share/mysql. These configuration files include my-small.cnf, my-medium.cnf, my-large.cnf, my-innodb-heavy-4G, and my-huge.cnf. The appropriate file can be copied and used as a template configuration by performing the following actions:

cd /usr/local/share/mysql/
cp /usr/local/share/mysq/mysql-large.cnf /etc/my.cnf
/usr/local/etc/rc.d/mysql-server restart

thanks to http://www.barik.net/archive/2008/05/26/114616/ for the tip.

Debian 4.0r4 NIC bonding

2008-09-17T13:26:21-07:00

Install package ifenslave-2.6.

apt-get install ifenslave

make sure the real NICs kernel modules are loaded automatically

modprobe --list | grep -i bonding

edit /etc/network/interfaces to look like this:

auto bond0
iface bond0 inet static
address 192.168.1.59
netmask 255.255.255.0
gateway 192.168.1.1
up /sbin/ifenslave bond0 eth0 eth1
down /sbin/ifenslave -d bond0 eth0 eth1
broadcast 192.168.1.255

make sure to remove ALL of the lines refering to old eth0 or eth1 settings.

add the following lines to your /etc/modprobe.d/arch/i386:

alias bond0 bonding
options bonding mode=1 miimon=100 downdelay=200 updelay=200

what not to do in MySQL... and how to pick up the pieces

2008-09-16T22:31:13-07:00

note to self... do NOT kill mysql processes from phpmyadmin's control panel and force mysql to quit. Bad things happen. In this case, mysql wouldnt start again. InnoDB's indexes, binary logs, etc. were all out of sync. The 20 page log explaining all the ways the DB couldn't start is located in the DB's data dir, which in my case was in /dbdisks/.err.

To tell mysql to pick up the messy pieces and throw up whats left of your data, edit or create /etc/my.cnf and add this little line to the "[mysqld]" section:

innodb_force_recovery = 3

The command is talked aboot here the number is btw 1 and 6. bigger the number, the more drastic and desperate it is. I tried 4, and it pulled the tables with no data. 1 gave me nothing. but 3 let me start and mysqldump my database.

mysqldump -u root -p <dbname > /path/to/.sql/file

so once the dump was completed, i moved the db data directory, commented out the "innodb_force_recovery" line, and restarted the computer. At restart the db data dir was re-created by mysql. fresh and clean. this deletes EVERYTHING. even users. re-import the data with

mysql -u root <dbname> < /path/to/.sql/file

go ahead and create the users needed, and set root password again. There you go.