NetBSD OS update/upgrade quick howto.

1. Fetch/Update the OS sources.
refs: NetBSD Docs (and NetBSD guide ; Fetching sources)

Fetch the source if you don't have it:
$ cd /usr
$ export CVS_RSH=ssh 
$ cvs -d co -r netbsd-5-0-2 -P src

Update the source if you already have it:
$ cd /usr/src
$ export CVS_RSH=ssh 
$ cvs update -dP

If you are fetching the sources from scratch use:
$ cd /usr
$ export CVS_RSH=ssh 
$ cvs -d co -r netbsd-5-1 -P src

Hint: If you are using 5-0 and want to update to 5-1, use
$ cvs update -r netbsd-5-1 -dP

2. Create obj dir and build the tools:
$ mkdir /usr/obj /usr/tools
$ cd /usr/src
$ ./ -O /usr/obj -T /usr/tools -U -u tools

3. Compile brand new userland:
NetBSD page says: Please always refer to -h and the files UPDATING and BUILDING for details - it's worth it, there are many options that can be set on the command line or in /etc/mk.conf.
$ cd /usr/src
$ ./ -O ../obj -T ../tools -U distribution

4. Compile brand New Kernel:
$ cd /usr/src
$ ./ -O ../obj -T ../tools kernel=

is a Kernel options file located in: /usr/src/sys/arch/amd64/conf/

I have XEN3_DOMU there that holds all my xen kernels compile options.
You can also find GENERIC and others there.

5. Install Kernel

Installing the new kernel (copy it in Dom0), rebooting (to ensure that the new kernel works) and installing the new userland are the final steps of the updating procedure:
$ cd /usr/obj/sys/arch/`uname -m`/compile/XEN3_DOMU/
$ scp netbsd Dom0 machine...

Go and change the kernel in the Dom0 to load the new one.
reboot the machine.

Or on native machines:
$ cd /usr/src
$ su
# mv /netbsd /netbsd.old
# mv /usr/obj/sys/arch/`uname -m`/compile/KERNEL/netbsd /
# shutdown -r now

6. Install new userland and reboot again to be sure it'll work. ;-)
Afrer we've rebooted we are sure all new calls in the new userland will be handled by the new kernel.
Now we'll install the new userland.
$ cd /usr/src
$ su
# ./ -O ../obj -T ../tools -U install=/ 

7. Build a complete release so we can copy it on all other machines and upgrade with sysinst.
$ ./ -O ../obj -T ../tools -U -u -x release
The resulting install sets will be in the /usr/obj/releasedir/ directory.

When you've tested on the package server. Install/update on all other machines.

1. Make a backup
2. Fetch a new kernel and the binary sets from the release dir and store them /some/where/
3. Install the kenrel (in XEN dom0)!
4. Install the sets except etc.tzg and xetc.tgz!!
   # cd /
   # pax -zrpef /some/where/set.tgz
   # ...
   # ...
5. Run etcupdate to merge important changes:
   # cd /
   # etcupdate -s /some/where/etc.tgz -s /some/where/xetc.tgz
6. Upgrade finished, time to reboot.

Backup xen lvm/image disks. xenBackup script.

Long time no write.

I'm trying to migrate all of my freebsds to xen+netbsd. (I gave up of this OS. You can't release STABLE that's not that stable. It's a long story but in shor, I've had a sleepless night after deploying a production. The problem - when it gets real world load it hangs with kernel panic and no auto reset about every 5-15mins. WTF? Devs asked me for a dump and told me that maybe they will find the problem. Sorry. That's sux and is not an option for a production used by thousands of people. Goodbye FreeBSD (for at least 5 years).

After successfully running xen for some time, it's time to think of automated backup, that cares for everything instead of writing short shells to do each xen backup.
I've made a quick search and found this xenBackup script that almost suits my needs.
I didn't like that it mounted lvm read only and didn't use snapshots.
The second thing I disliked was that is used to work only with lvms and I do have a sparse xen images (for small machines that don't need quick disk access and have only 1-2 services running in memory).

I've modified the script and now the xenBackup script supports:
- creating backup from lvm snapshots
- creating backup from disk.img file
- dynamic determination of the disk type and path ($hostname-disk for lvms and disk.img for sparse) (BE WARNED: only -disk and .disk will be backed up!)

I'm using tar, so I didn't tested with rsync and rdiff-backup.
I'm using snapshots. Never tested with readonly lvm mounted.

so, here is the code:

#   Copyright John Quinn, 2008
#   Copyright Anton Valqkoff, 2010
#   This program is free software: you can redistribute it and/or modify
#   it under the terms of the GNU General Public License as published by
#   the Free Software Foundation, either version 3 of the License, or
#   (at your option) any later version.
#   This program is distributed in the hope that it will be useful,
#   but WITHOUT ANY WARRANTY; without even the implied warranty of
#   GNU General Public License for more details.
#   You should have received a copy of the GNU General Public License
#   along with this program.  If not, see .

# xenBackup - Backup Xen Domains
#             Version:    1.0:     Created:  John D Quinn,
#             Version:    1.1:     Added file/lvm recognition. lvm snapshot:  Anton Valqkoff,

# initialize our variables
domains="null"                           # the list of domains to backup
allDomains="null"                        # backup all domains?
targetLocation="/root/backup/"                    # the default backup target directory
mountPoint="/mnt/xen"                    # the mount point to use to mount disk areas
shutdownDomains=false                    # don't shutdown domains by default
quiet=false                              # keep the chatter down
backupEngine=tar                         # the default backup engine
useSnapshot=true                        # create snampshot of the lvm and use it as backup mount.
rsyncExe=/usr/bin/rsync                  # rsync executable
rdiffbackupExe=/usr/bin/rdiff-backup     # rdiff-backup executable
tarExe=/bin/tar                      # tar executable
xmExe=/usr/sbin/xm                       # xm executable
purgeAge="null"                          # age at which to purge increments
globalBackupResult=0                     # success status of overall job
#valqk: xm list --long|grep -A 3 device|grep vbd -A 2|grep uname|grep -v swap|awk '{print $2}'

# settings for logging (syslog)
loggerArgs=""                            # what extra arguments to the logger to use
loggerTag="xenBackup"                    # the tag for our log statements
loggerFacility="local3"                  # the syslog facility to log to

# trap user exit and cleanup
trap 'cleanup;exit 1' 1 2

   ${logDebug} "Cleaning up"
   #check if file or lvm.if lvm and -snap remove it.
   mountType=`${mountExe}|${grepExe} ${mountPoint}|${awkExe} '{print $1}'`;
   [ -f ${mountType} ] && mountType="file";
   cd / ; ${umountExe} ${mountPoint}
   if [ "${mountType}" != "file" ] && [ "${useSnapshot}" = "true" ]; then
      #let's make sure we are removing snapshot!
      if [ `${mountExe}|${grepExe} -snap|wc -l` -gt 0 ]; then
         ${lvmExe} lvremove -f ${mountType}

   # restart the domain
   if test ${shutdownDomains} = "true"
      ${logDebug} "Restarting domain"
      ${xmExe} create ${domain}.cfg > /dev/null

# function to print a usage message and bail
usageAndBail() {
   cat << EOT
Usage: xenBackup [OPTION]...
Backup xen domains to a target area. different backup engines may be specified to
produce a tarfile, an exact mirror of the disk area or a mirror with incremental backup.

   -d      backup only the specified DOMAINs (comma seperated list)
   -t      target LOCATION for the backup e.g. /tmp or
           (not used for tar engine)
   -a      backup all domains
   -s      shutdown domains before backup (and restart them afterwards)
   -q      run in quiet mode, output still goes to syslog
   -e      backup ENGINE to use, either tar, rsync or rdiff
   -p      purge increments older than TIME_SPEC. this option only applies
           to rdiff, e.g. 3W for 3 weeks. see "man rdiff-backup" for
           more information

Example 1
   Backup all domains to the /tmp directgory
   $ xenBackup -a -t /tmp

Example 2
   Backup domain: "wiki" using rsync to directory /var/xenImages on machine backupServer,
   $ xenBackup -e rsync -d wiki -t root@backupServer:/var/xenImages

Example 3
   Backup domains "domainOne" and "domainTwo" using rdiff purging old increments older than 5 days
   $ xenBackup -e rdiff -d "domainOne, domainTwo" -p 5D


   exit 1;

# parse the command line arguments
while getopts p:e:qsad:t:h o
do     case "$o" in
        q)     quiet="true";;
        s)     shutdownDomains="true";;
        a)     allDomains="true";;
        d)     domains="$OPTARG";;
        t)     targetLocation="$OPTARG";;
        e)     backupEngine="$OPTARG";;
        p)     purgeAge="$OPTARG";;
        h)     usageAndBail;;
        [?])   usageAndBail

# if quiet don't output logging to standard error
if test ${quiet} = "false"

# setup logging subsystem. using syslog via logger
logCritical="logger -t ${loggerTag} ${loggerArgs} -p ${loggerFacility}.crit"
logWarning="logger -t ${loggerTag} ${loggerArgs} -p ${loggerFacility}.warning"
logDebug="logger -t ${loggerTag} ${loggerArgs} -p ${loggerFacility}.debug"

# make sure only root can run our script
test $(id -u) = 0 || { ${logCritical} "This script must be run as root"; exit 1; }

# make sure that the guest manager is available
test -x ${xmExe} || { ${logCritical} "xen guest manager (${xmExe}) not found"; exit 1; }

# assemble the list of domains to backup
if test ${allDomains} = "true"
   domainList=`${xmExe} list | cut -f1 -d" " | egrep -v "Name|Domain-0"`
   # make sure we've got some domains specified
   if test "${domains}" = "null"

   # create the domain list by mapping commas to spaces
   domainList=`echo ${domains} | tr -d " " | tr , " "`

# function to do a "rdiff-backup" of domain
backupDomainUsingrdiff() {
   test -x ${rdiffbackupExe} || { ${logCritical} "rdiff-backup executable (${rdiffbackupExe}) not found"; exit 1; }

   if test ${quiet} = "false"


   # make the targetSubDir if it doesn't already exist
   mkdir ${targetSubDir} > /dev/null 2>&1
   ${logDebug} "backing up domain ${domain} to ${targetSubDir} using rdiff-backup"

   # rdiff-backup to the target directory
   ${rdiffbackupExe} --verbosity ${verbosity} ${mountPoint}/ ${targetSubDir}

   # purge old increments
   if test ${purgeAge} != "null"
      # purge old increments
      ${logDebug} "purging increments older than ${purgeAge} from ${targetSubDir}"
      ${rdiffbackupExe} --verbosity ${verbosity} --force --remove-older-than ${purgeAge} ${targetSubDir}

   return ${backupResult}

# function to do a "rsync" backup of domain
backupDomainUsingrsync() {
   test -x ${rsyncExe} || { ${logCritical} "rsync executable (${rsyncExe}) not found"; exit 1; }


   # make the targetSubDir if it doesn't already exist
   mkdir ${targetSubDir} > /dev/null 2>&1
   ${logDebug} "backing up domain ${domain} to ${targetSubDir} using rsync"

   # rsync to the target directory
   ${rsyncExe} -essh -avz --delete ${mountPoint}/ ${targetSubDir}

   return ${backupResult}

# function to a "tar" backup of domain
backupDomainUsingtar ()

   # make sure we can write to the target directory
   test -w ${targetLocation} || { ${logCritical} "target directory (${targetLocation}) is not writeable"; exit 1; }

   targetFile=${targetLocation}/${domain}.`date '+%d.%m.%Y'`.$$.tar.gz
   ${logDebug} "backing up domain ${domain} to ${targetFile} using tar"

   # tar to the target directory
   cd ${mountPoint}

   ${tarExe} pcfz ${targetFile} * > /dev/null

   return ${backupResult}

# backup the specified domains
for domain in ${domainList}
   ${logDebug} "backing up domain: ${domain}"
   [ `${xmExe} list ${domain}|wc -l` -lt 1 ] && { echo "Fatal ERROR!!! ${domain} does not exists or not running! Exiting."; exit 1; }

   # make sure that the domain is shutdown if required
   if test ${shutdownDomains} = "true"
      ${logDebug} "shutting down domain ${domain}"
      ${xmExe} shutdown -w ${domain} > /dev/null

   # unmount mount point if already mounted
   umount ${mountPoint} > /dev/null 2>&1

   #inspect domain disks per domain. get only -disk or disk.img.
   #if file:// mount the xen disk read-only,umount sfter.
   #if lvm create a snapshot mount/umount/erase it.
   xenDiskStr=`${xmExe} list --long ${domain}|${grepExe} -A 3 device|${grepExe} vbd -A 2|${grepExe} uname|${grepExe} -v swap|${awkExe} '{print $2}'|${egrepExe} 'disk.img|-disk'`
   xenDiskType=`echo ${xenDiskStr}|${cutExe} -f1 -d:`;
   xenDiskDev=`echo ${xenDiskStr}|${cutExe} -f2 -d:|${cutExe} -f1 -d')'`;
   test -r ${xenDiskDev} || { ${logCritical} "xen disk area not readable. are you sure that the domain \"${domain}\" exists?"; exit 1; }
   #valqk: if the domain uses a file.img - mount ro (loop allows mount the file twice. wtf!?)
   if [ "${xenDiskType}" = "file" ]; then
      ${logDebug} "Mounting file://${xenDiskDev} read-only to ${mountPoint}"
      ${mountExe} -oloop ${xenDiskDev} ${mountPoint} || { ${logCritical} "mount failed, does mount point (${mountPoint}) exist?"; exit 1; }
      ${mountExe} -oremount,ro ${mountPoint} || { ${logCritical} "mount failed, does mount point (${mountPoint}) exist?"; exit 1; }
   if [ "${xenDiskType}" = "phy" ] ; then
      if [ "${useSnapshot}" = "true" ]; then
         vgName=`${lvmExe} lvdisplay -c |${grepExe} ${domain}-disk|${grepExe} disk|${cutExe} -f 2 -d:`;
         lvSize=`${lvmExe} lvdisplay ${xenDiskDev} -c|${cutExe} -f7 -d:`;
         lvSize=$((${lvSize}/2/100*15)); # 15% size of lvm in kilobytes
         ${lvmExe} lvcreate -s -n ${vgName}/${domain}-snap -L ${lvSize}k ${xenDiskDev} || { ${logCritical} "creation of snapshot for ${xenDiskDev} failed. exiting." exit 1; }
         ${mountExe} -r /dev/${vgName}/${domain}-snap ${mountPoint} || { ${logCritical} "mount failed, does mount point (${mountPoint}) exist?"; exit 1; }
         ${mountExe} -r ${xenDiskDev} ${mountPoint}

   # do the backup according to the chosen backup engine
   backupDomainUsing${backupEngine} ${domain}

   # make sure that the backup was successful
   if test $? -ne 0
      ${logCritical} "FAILURE: error backing up domain ${domain}"
      ${logDebug} "SUCCESS: domain ${domain} backed up"
   # clean up
if test ${globalBackupResult} -eq 0
   ${logDebug} "SUCCESS: backup of all domains completed successfully"
   ${logCritical} "FAILURE: backup completed with some failures"

exit ${globalBackupResult}

Setup SVN repositories only for specified users over ssh. OpenSSH limit only one command execution.

Just to blog this. I'll need it in future.
If you have svn repositories server and you are using svn+ssh for the checkout and all svn actions you will want users to have access to only predefined repos only and not to any shell or anything.
I've done this by doing symlinks in their homes and using ssh file that looks like this

command="svnserve -t --tunnel-user=user -r /home/user",no-port-forwarding,no-agent-forwarding,no-X11-forwarding,no-pty ssh-rsa AAAAB3Nz1...KEY HERE....

this way, you can lock them to use only svnserve and it will lock them to co only what's in their home dirs.

If you're not familiar with details - eg. how to generate keys, what is authorized_keys etc, I stole this from here: - read more there.

Of course you have to keep your snserve up to date and pray there are no vulns in it, otherwise users can hack you :-)
But hey, you know the owners of the keys, don't you? :-)
Got my pont? ;-)

Roundcube with plugins support!!! WOW! Writing a plugin - display custom template has bogus docs.

Today I've noticed Roundcube has released a new version that finally has plugins support!

As expected in there is a change password plugin (with drivers supports) and some other that are pretty cool!
A list of plugins here:

Of course I've had some custom patching for my hosting users and now it's not working.
I've configured my change password plugin (which was the main showstopper for not upgrating to new roundcube) and the the little tiny hack for domain notification left.
I've decided to write a plugin that will do the job for me, so I can easily upgade after that.

Writing plugin isn't that hard at all. Here you can read more:

also you can read plugins directory for more.

While creating my plugin I hit a problem and I've lost about 40 minutes searching for description and resolution.
The Resolutions was 5mins reading the class for temapltes but I thought I was wrong - no this is a mis-explanation in docs.
When you want to create a custom template you mkdir skins/default/templates and create/copy-modify html in it (I've copied login.html template).
Well all was fine while I've tried to show it.
Documentation is wrong.
When you call:


you must actually call:

so the tpl class can understand this is a plugin and show your template and not search for default tpl.

Hope that helps someone.
Going to change/report this in docs now.
Oh. Symptoms are:

[12.Nov.2009 17:57:27 +0200]: PHP Error: Error loading template for logininfo in /var/www/roundcube/program/include/rcube_template.php on line 372 (GET /)

in your error log.

Dojo: breaking in IE*

If your dojo based website breaks in IE browsers and not in others, with strange errors in dojo.js then you have to check VERY CAREFULLY for unclosed tags.

I've had this problem - didn't closed one (only one!) div inside a HTML markup node that used dojoType and viola - dojo threw a "NICE" js error in IE (you know how js is debuged in IE don't ya?) :-)

So be very very careful when closing tags and using IE+dojo :-)

IE8 and Opera 10 absolute positioning problems

IE8 and Opera 10 differs to ALL other browsers (FF3, Safari, Chrome, IE6, IE7) in positionin an absolute element inside a div.
If you have something like this:

If you don't put the right: 0px the element won't keep it's original position an will go to the left side of the div becase IE8 and Opera will put default left: 0px if nothing set.
All other browsers will keep a's original position (no left: 0px;)
hope that help to someone.
Keywords: IE8 Opera absolute positioning problem

Q&A for apache in debian

Q: Why does Apache Web server in Debian has 'It works!' page as it's default host?
A: Because after you have setupped a complex VirtualHost configuration for half an hour or more (yesh, there can be such), it's nice to see that 'It worked!'
--answered by valqk. :-D

Debian HP SmartArray RAID monitoring.

You need to install 2 utils to monitor and query your smart array:

apt-get install arrayprobe cpqarrayd
the one is a daemon that logs events from the controller - cpqarrayd (thanks velin)
arrayprobe is the cli tool.

More links on the topic:
source I've got this from.
driver and utils page.

if you have faulty drive

hope that helps.


In squeeze there is no cpqarrayd and arrayprobe is not that good.
You can use the hp tools provided in debian packages.
Simply add this source:

deb squeeze/current non-free

#> apt-get update && apt-get install hpacucli

This is the way this CLI is being used: hpacuclu usage
p.s. not yet figured out about monitoring. hp-health is something I've read but didn't tested yet.

Fun with JavaScript... I don't recommend this in your code! :-)

Facebook, FB.Connect - write nice js code and reuse code call...

I'm writing a poc code that calls some FB.Connect methods.
As a quick and nasty code reuse I've come up with this code:

A method that inits and makes the actual code:
function fbCall(code) {
    FB_RequireFeatures(["XFBML"], function(){
        FB.Facebook.init('ApiKey', '/xd_receiver.htm', null);
        FB.ensureInit(function () {
so far so good - it all seems ok.
Here comes the tricky part. I wanted to be able to call multiline variable with comments in it - a normal js code but encapsulated in somethind...
If you don't know in JS you can't have multiline variable, and if you have something like:
    var mycall = 'FB.Connect.showFeedDialog(
//here we put some data...
comment_data, '', "Awesome", null, 
FB.RequireConnect.promptConnect, function(){alert("Callback");}, fortune, user_message);';
you'll get error while parsing because of the new lines.
If you replace the new lines with ' ' you'll get the whole code after a comment - commented exept you don't use / /

The solution is this:
function fbCall(code) {
    FB_RequireFeatures(["XFBML"], function(){
        FB.Facebook.init('ApiKey', '/xd_receiver.htm', null);
        FB.ensureInit(function () {
function askPerms() {
    var c = function() {
Notice the difference between two fbCall functions - the second one calls code as a function - it do not evals it.
This way you can write up your code inside the c 'function' variable and call it after that.
It's a bit tricky while you get it how it works but the code looks more readable after that.

Protect yourself from accidentally halting a server.

In Short: use molly-guard (debian name)
While reading my 10 unix command line mistakes I've saw the wrong halting machine command.
It's nasty to halt some server instead of your local desktop.
I use molly-guard (on Debian servers - not avaliable in FreeBSD. dunno for other linuxes? any comments?) to protect myself from this kind of mistake.
It modifies the halt/shutdown script and asks you for the hostname of the server before shutdown if from ssh session.
#>apt-get install molly-guard

when installed if you try to shutdown or reboot:
storm:/home/valqk# halt
W: molly-guard: SSH session detected!
Please type in hostname of the machine to reboot: ^C
Good thing I asked; I won't halt storm ...
W: aborting reboot due to 30-query-hostname exiting with code 1.

molly-guard saved the world for me again! :-)
have a nice Friday evening!

How to enable new NetBSD ffs WAPBL feature? How to extend ffs size?

How to enable/use WAPBL in netbsd 5.0?

  1. you MUST have options WAPBL in your kernel (it's there in most archs)

  2. mount the desired filesystem with -o log (or add rw,log in /etc/fstab) - that's all. The log will be created automatically when this optiuon is in act.

(Source )

How to extend ffs size?

According to my research you can't do this at the moment.
Can anyone correct me and make me happy?

FreeBSD jails: how to login /jexec JID SHELL/ quickly in a jail by name (jlog command)?

Have you ever wondered why the heck you write jls then jexec JID /bin/csh?
I got sick of this few year ago and I wrote a tiny little script that makes my life easy every day.
Be warned, there are few cases when you'll have to see your jid but this works with jid too.
(for example when you have hang jail that won't shutdown /stop/ - happens to me pretty often and this is reported as non-critical bug for years...)

How it works?
Let's pretend that we have a jail named: ''. Then you simply type this to get in the mailserver:
#>jlog mail
Logging in to

It's that easy. Also you can add a preffered custom shell for this session after the jail (or part) name.

What looks like the script itself?
There it goes:
[ -z "$1" ] && echo "No jail specified." && exit 1;
[ -z "$2" ] && loginSHELL="/bin/tcsh" || loginSHELL="$2"
jID=`jls | grep $jName|awk '{print $1}'`
jRealName=`jls | grep $jName|awk '{print $3}'`
[ -z "$jID" ] && echo "No such jail name $jName!" && exit 1;
echo "Logging in to $jRealName"
jexec $jID $loginSHELL
please feel free to use, comment, improve this script! If you make any improvements, pls tell me!
I'll definitely add changes if I like them!!!

Serendipity and dpSyntaxHighlighter plugin with bash support.

I wanted a syntax highlighter for the previous post.
I've installed dpsyntaxhighlighter from serendipity plugins list. I wanted a JS code highlighter - not a php one (like GeSHi), so I've chosen this one. (it uses google syntaxhighlighter)
I've noticed that this nice lib don't have a bash syntax support which I needed now.
In the wiki of the project I've found this link to a script a guy wrote for bash syntax.
I've placed the js file in ROOT/plugins/serendipity_event_dpsyntaxhighlighter/dp.SyntaxHighlighter/Scripts/ and expected it to work.
No, it didn't.

It turned out that you have to add each language highliter js script file in ROOT/serendipity_event_dpsyntaxhighlighter.php

                case 'frontend_header':
                    echo '    <link rel="stylesheet" type="text/css" href="' . $pluginDir.  '/SyntaxHighlighter.css" />' . "\n";
                    return true;
                case 'frontend_footer':

there is a list with all js files.
Simply add the new language (on both frontend_footer and backend_preview cases) and there you go!

How to highlight your text?
It is used very simple.
You add code like this in your post:

    ... some code here ...

and you have nice formatted code.
More usage tips here.

That's all folks.
Hope that helps you.

Xen: firewall DomU from Dom0

Have you ever wondered how to force some firewall rules on a xen DomU and the DomU root won't be able to use some ports etc?
Well, the only proper way is to firewall DomU from the Dom0 machine.
Here is a way to do it.
This script is just an example. It should be made more universal and can apply to ALL of your DomU's for their protection :-) or logging specific traffic.
/sbin/iptables -N vps
#outbound traffic redirect to vps - a per DomU chain.
/sbin/iptables -I FORWARD -m physdev  --physdev-out peth0 --physdev-in $vifname -j vps
#log some of the traffic
/sbin/iptables -A "vps" -j LOG -m  tcp --dport 80,110,113 --log-level 4 --log-prefix '*DomUNameHere-shows-in-logs*'
#allow some ports
/sbin/iptables -A "vps" -p tcp -m tcp --dport 20 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 21 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 22 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 80 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 443 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 6666 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 6667 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 6668 -j RETURN
/sbin/iptables -A "vps" -p tcp -m tcp --dport 6669 -j RETURN
/sbin/iptables -A "vps" -p udp -m udp --dport 53 -j RETURN
#allow establieshed connections from inside the DomU to go back in
/sbin/iptables -A "vps" -p tcp -m state --state RELATED,ESTABLISHED -j RETURN
#drop all other traffic.
/sbin/iptables -A "vps" -p tcp -j DROP

Setting up GRUB to boot from both disks of mirrored RAID

copy/paste from:

Many people use mirrored RAID (also known as 'RAID 1') to protect themselves against data loss caused by hard disk failure. Sometimes, you even want GRUB to boot from the secondary hard disk in case the primary fails to keep the system up and running. This is however not as easy as one might think...

GRUB keeps track of the hard disks currently available on your system, on most distributions you can find this information in /boot/grub/ You might have a file like this:

hopper:~# cat /boot/grub/
(hd0) /dev/sda
(hd1) /dev/sdb

Of course you can install GRUB to /dev/sdb (which is hd1), but obviously GRUB will be confused if /dev/sda fails and hd1 becomes hd0. Most likely, it will complain about a failing hard disk at boot time:

GRUB Hard Disk Error

In this case, you want to install GRUB to /dev/sdb and have sdb also mapped to hd0:

hopper:~# cat /boot/grub/
(hd0) /dev/sda
(hd0) /dev/sdb
hopper:~# grub-install /dev/sdb
The drive (hd0) is defined multiple times in the device map /boot/grub/

GRUB doesn't accept this duplicate definition (which is indeed incorrect), so you need to configure things by hand:

hopper:~# grub
grub> device (hd0) /dev/sdb
grub> root (hd0,0)
Filesystem type is ext2fs, partition type 0xfd

grub> setup (hd0)
Checking if "/boot/grub/stage1" exists... yes
Checking if "/boot/grub/stage2" exists... yes
Checking if "/boot/grub/e2fs_stage1_5" exists... yes
Running "embed /boot/grub/e2fs_stage1_5 (hd0)"... 15 sectors are embedded.
Running "install /boot/grub/stage1 (hd0) (hd0)1+15 p (hd0,0)/boot/grub/stage2 /boot/grub/menu.lst"... succeeded

grub> quit

Now, /dev/sda and /dev/sdb are configured as hd0 and the system remains bootable if /dev/sda fails.

Assumptions about partitions

The above information only works if your boot filesystem can be found on both /dev/sda1 and /dev/sdb1. If you have /boot on e.g. /dev/sda5 and /dev/sdb5, you'll have to replace root (hd0,0) with something more applicable for your specific configuration.

Howto: Migrate linux (debian lenny) from one single disk to two mirrored/lvm-ed disks?

I've got a server (actually my desktop testing machine) with two brand new installed 2x1T disks.
I'm going to setup the disks like this:
3 partitions:
1 swap (we really don't need abstractions for just keeping swap)
2 boot partition in md raid1 (grub2 really sux, so no boot from lvm support in the old one...)
3 all othe space for md raid1 and lvm over it.
It's a good idea to use lvm because you can always add another disk and also can make snapshots... and in short - have more fun with space allocating.

1. partition the two disks identically:
#> fdisk -l /dev/sdb
Device Boot Start End Blocks Id System
/dev/sdb1 1 974 7823623+ 82 Linux swap / Solaris
/dev/sdb2 975 1461 3911827+ fd Linux raid autodetect
/dev/sdb3 1462 121601 965024550 fd Linux raid autodetect

(yes, I know the boot partition is quite big, but there is a lot of space and I prefer to have more space than to wonder wtf I've done... happened few timesof course :-D)

2. create raids.
#> mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sdb2 /dev/sdc2
and activate
#> mdadm --readwrite /dev/md0
be sure it's sync-ing
#> cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdc2[1] sdb2[0]
3911744 blocks [2/2] [UU]
[=>...................] resync = 5.2% (206848/3911744) finish=0.5min speed=103424K/sec

do the same for lvm raid partitions....

3. format boot partition (I'll use etx3) and copy boot files in there.
#> mkfs.ext3 /dev/md0
#> mount /dev/md0 /mnt/
#> cd /mnt/
#> cp -a /boot/
#> cd grub
WARNING: sda2 is supposed to be your current partition.
#> sed -i'' -e 's/\/boot\//\//g' -e 's/sda2/mapper\/1tb-root/g' menu.lst
#> cd /;unmount /mnt

4. create physical volume on md1.
#> pvcreate /dev/md1
Physical volume "/dev/md1" successfully created

5. create volume groups - my vg name 1tb
#> vgcreate -A y 1tb /dev/md1
Volume group "1tb" successfully created

6. then add root volume group. mkfx.ext3.mount and copy the currently running root system in new root partition...
#> lvcreate -A y -L 30G -Z y -n root 1tb
#> mkfs.ext3 /dev/1tb/root
#> mount /dev/1tb/root /mnt
#> cd /mnt
#> cp -a {/bin,/cdrom,/emul,/etc,/home,/initrd*,/lib,/lib32,/lib64,/media,/opt,/root,/sbin,/selinux,/srv,/tmp,/usr,/var,/vmlinuz*} .
#> mkdir dev proc sys mnt misc boot
#> cd etc
#> sed -i'' -e 's/sda2/mapper\/1tb-root/g' fstab
WARNING: There is a nasty bug with initramfs tools described here:
You MUST set root to /dev/mapper/VGNAME-LVNAME otherwise you won't get lvm support in your kernel.
#> echo "/dev/md0 /boot ext3 defaults 0 0" >> fstab
#> mount -o bind /dev /mnt/dev
#> mount -o bind /prov /mnt/proc
#> cp -a /boot/ /mnt/boot/
#> chroot /proc
#> update-initramfs -u -t -k `uname -r`
#> exit
reboot the machine, edit the grub menu by hand to boot from (hd0,0) as boot and load /dev/mapper/1tb-root as root.
login as root and make:
#> cd /boot; mkdir oldrootfs; mv
oldrootfs; mv -fr oldrootfs/boot/* .;
edit grub/menu.lst to have / instead ot /boot/ dirs.
#> grub-install --root-directory=/ "(hd0)" (described in /usr/share/doc/grub/README.Debian.gz
(root-directory is where the boor partition is on hd0, if you put /boot/ then in live system you'll have /boot(md0)/boot/)...
reboot again and it must be ok now.

7. Install grub on both disks as described here...

8. reboot and disable hdd from bios. the system should boot normally and you should see that you are using the lvm root partition.

Linux LVM - MD Raid vs LVM mirror. Snapshots.

I'm migrating a xen machine (my gallery - ) from one machine to another.
I wanted to research from quite a long time what should I use in Linux.
I've got a nice working setup but when you research and test you always makes things better.

So, this night I've researched on if I should do MD raid1 or LVM2 mirroring.
Short answer: MD0 raid1 and Physical Volume over it (as used to now). Sometimes LVM is faster but when using single file. You can have serious problems if you have power failure or disk failure/disconnection when using LVM mirroring. It relies on underneath layer to write caches. For further reading open the link below.
Long answer here: lvm2-mirrors-vs-md-raid-1

PureEdit (rocking ultra light CMS system) and utf8 / unicode support. Pagination.

I've found PureEdit today and decided to use it as my CMS backend for some of my simpler projects!
This is a GREAT and FAST way to setup a backend content editing.
Take a look at the videos on their site! I was fascinated!

I've had old database filled with correct Unicode (UTF-8) data (in Cyrillic).
I've installed and loaded pe-admin, updated my db accortind to PureEdit specs, added some fields type like (status 1||0).
When opened pe-admin I've come up with few Unicode problems.
(as described here: )
I've solved my problem exactly by putting: mysql_query('SET NAMES UTF8');

I've had to modify the connect function in pe-admin/databases/mysql.db.php like this:
function connect($host, $username, $password, $database)
$dbh = mysql_connect($host, $username, $password);
mysql_query('SET NAMES UTF8');
mysql_select_db($database, $dbh);
(it used to connect/select db in one line, but you MUST make SET NAMES BEFORE selecting db!)

The second problem that I've had as with utf8 again.
Utils class is not UTF-8 ready.
I've had to add to set utf8 encoding
public function __construct() {
and replace all str functions with mb_str.

After that I wanted to have pagination.
Here it's described how to set it up.

Now I have to setup host type filed (that will be automatically filled with $_SERVER['REMOTE_ADDR']) and some othe fancy stuff, but in about 40 minutes I've setup a GREAT and easy to use backend posting solution for my website!

Thanks PureEdit :-)

The PureEdit code is not from the greatest ones but when it works and if .htaccess protected - it simply works! :-)

Filesystems speed comparison: ext2, ext3, ext4dev, reiserfs and xfs.

Just a simple test for speed of ext2, ext3, ext4dev, reiserfs and xfs.
I've created a 5G lvm and did mkfs.* to it with all these systems.
lvm> lvcreate -L 5G -n speedtest tvg
mkfs (all default opts)
mount (no opts)
Then make dd if=/dev/zero of=a bs=G count=3
Then make dd if=a of=b

Here are the results:

FS type dd if=/dev/zero of=a speed dd if=a of=b speed
ext2 3221225472 bytes (3.2 GB) copied, 225.283 s, 14.3 MB/s 2047094784 bytes (2.0 GB) copied, 38.4914 s, 53.2 MB/s
ext3 3221225472 bytes (3.2 GB) copied, 248.433 s, 13.0 MB/s 1912479744 bytes (1.9 GB) copied, 52.1087 s, 36.7 MB/s
ext4 3221225472 bytes (3.2 GB) copied, 281.699 s, 11.4 MB/s 1918500864 bytes (1.9 GB) copied, 49.1343 s, 39.0 MB/s
reiserfs 3221225472 bytes (3.2 GB) copied, 248.827 s, 12.9 MB/s 2108379136 bytes (2.1 GB) copied, 62.5061 s, 33.7 MB/s
xfs 3221225472 bytes (3.2 GB) copied, 426.823 s, 7.5 MB/s 2132619264 bytes (2.1 GB) copied, 55.3356 s, 38.5 MB/s

It seems like ext3 is working best with big continious files currently.
Of course this is not a real life situation.
Maybe xfs and reiser will be better for a lot of small files? (reiser has it's glory about that).

Please let me know what you think!

Copy permissions of files from one dir to another or how to change permissions and ownership of identical dirs.

Have you ever been in the middle of deployment and noticed that the owner and permissions on dev package you have and production one are different?
Very oftern when deploing some sites I have two identical dirs with different owners and permissions - one from a dev web server and another from the production (where everything is tuned and working).
Here is a quick and durty way to change all ownership and permissions from prodution to development dir.
BE VERY CAREFULL. These permissions are not all masks. These are the one I needed!
Don't blind apply this script and run after that!!!
FIRST check that you don't have some permissions rules in the chmod!!!
If you have 'chmod rw-r---- FILENAME' you will finish with -------- (0000) permissions!!!
You were warned!
mysite.production is a copy of the production dir somewhere else in the tree (your home for example).

$> find mysite.production/ ! -type l -ls | awk '{print "chmod "$3" "$11 " && chown "$5":"$6" "$11}' | sed -e 's/\.production//g' -e 's/-r--r--r--/044/' -e 's/-rw-rw-rw/0666/' -e 's/-rwxrwxrwx/0777/' -e 's/-rwx------/0700/' -e 's/-rwxr-xr-x/0755/' -e 's/-rwxr-x---/0750/' -e 's/-rw-r--r--/0644/' -e 's/-r-x------/0500/' -e 's/drwxr-xr-x/755/' -e 's/drwxrwxrwx/777/' >

and try to compare some files that have specific perms.

that's all.
I'm SURE there is an easier way, and I'll be glad to share!!!

Xen + Desktop (3d) or what video card is ok if you want to use your quad desktop for xen machine

Short story:
If you have a machine that you want to have XEN+3D use ATI with radeonhd driver (check wich cards are supported).
Take a look here before try more than basic configuration of the driver:

Long story:
For more than 5 months I've had a machine at home that was always off.
But not until tonight. :-)
It is a Core 2 Quad @ 2.5 GHZ with 4 gigs ram - a killer for desktop, very descent for server.
You'll almost never need that kind of power (expect you are a mad scientist doing some kind of calculations or a crazy gamer - not me!).
You'll definitely don't need it if you only watch movies, play some stupid doom like games, browse, read pdfs and do all kind of daily stuff on it.
When I got it I told to myself that I'll never use it until I can both have it as my test deployment server and as my primary desktop (including a dual headed video with two monitors connected and using two different GNOMEs on each).

You'll tell to yourself: 'uh! easy stuff.' (at least that's what I thought)
Well here I am five months later, at last I was able to complete my setup.

I've started by installing Debian Lenny (current unstable - now stable) and started digging....
The easiest part was configuring the dual head card to use the two separate monitors and to have two totally separated GNOMEs running.
It took me about two days to figure out that. It turned out that it is impossible to do this with only two X servers.
That is because they can't get attached to the same hardware card (if you have two different hardware cards - no problems).
The trick was to start a X server with wide screen splited on the two monitors (so called Xinerama or dual-view-mode - you know when you can move a window from one monitor to another). Upon this server you get another two 'virtual servers' started each running a GDM.
I'll find the link describing this and put it here.

So far, so good - I've got the two separate X working perfectly and I was almost complete.
Last step was just to change the kernel with XEN enabled one and I'm done.
When I rebooted in the XEN kernel I was 'WOW---F**KING' ... the XEN kernel was not able to work with Nvidia proprietary drivers.
At that time the machine was with NVIDIA GForce 7600 card.
I dug few weeks just to find nothing new - there were two ways to use NVIDIA (drivers):

  • stop using XEN

  • patch your kernel and hope you get it running and hopefully you'll have no problems

I didn't wanted to change my vanilla debian kernel because this is going to be my test deployment server and it should be as close as it can to my production machines.
Until then NVIDIA is dropped from my favorite video cards list - sad but true, it was my leader and most recommended card to be used it with linux...
What can you think of after NVIDIA?
ATI of course!
I have one - Asus X1600 (ATI chipset: RV530 X1600).
I've plugged it in. At that time I knew that only proprietary drivers were available for newer ATI cards, otherwise you get VERY bad performance and no 3d. That's why I've installed fglrx driver simply to find out that they also can't work with XEN.
The situation seemed hopeless. What other card can you use? (matrox? they've reports to have troubles with xen too!) and have descent 3d in linux? (please If you are aware of any card, post into the comments!)
I've left the project for few months but from few days while talking with a friend of mine, I've started looking for alternative of NVIDIA and ATI.

Tonight I was very surprised when I saw someone mentioned radeonhd driver. I've never heard of it. I've started digging and in about an hour I've got my ATI running with radeonhd driver and showing about 1500 FPS in glxgearx (WOW!!!) with XEN ENABLED KERNEL!
What was my suprise to see this. My problem was solved!
Just lost another hour to figure out how to start my two monitors split ed out in two with one on the left of the other... Xinerama won't work with this driver.
To fix things as you want use xrandr command ( from #radeonhd hint)
You can read more here:
Be careful with the device names.
These commands did the job for me:

xrandr --output DVI-I_1/digital --off
xrandr --output DVI-I_1/digital --mode 1680x1050 --rate 75
xrandr --output VGA_1 --left-of DVI-I_1/digital

Now I'm happy opensource driver user with most debian vanilla system I can imagine. :-)

php mail function overloading or why we shouldn't use mb_send_mail as sending mail function.

As a programmer I have a nice cool helpers - classes I've written to help me do my daily work.
Such class is my Mailer class which I use almost daily.
It has all kind of fancy options and can do all kind of stuff - from sending simple mail to just setting an array with files which should be attached in the mail.
It used to work very good until today.
I was very surprised when I saw a broken email coming from it.
I've investigated a few minutes what has changed and I've found that the problem
I've set
mbstring.func_overload = 7
according to the description in file:
; For example, 7 for overload everything.
; 0: No overload
; 1: Overload mail() function
; 2: Overload str*() functions
; 4: Overload ereg*() functions
it turned out that when set to 7 instead of using mail() function, php starts using mb_send_mail.
mb_send_mail works only with base64 emails.
It automatically encodes the body and subject (but not the sender wtf?) and it breaks mail ugly!
If you manually add/set headers do not put \r\n at the very end of the string, mb_send_mail adds two headers after yours:
Mime-Version: 1.0
Content-Transfer-Encoding: BASE64
and if you put one new line(\r\n) to separate the headers from body then you get the email totally messed up, because the base64 transfer won't apply and you'll get encoded string as plain text or html or whatever content-type you set previously.
After I've lost another 20 minutes figuring out this, I've finally achieved receiving my email as it should be... well... almost.
When the mail contained only ASCII symbols everything was ok, but when I tried to send email in utf8 Bulgarian... omfg?!
The encoding was broke and for some reason the UTF8 symbols looked like you have page in UTF8 but your browser is set to WINDOWS-1251....
WTF?!!! I've read few posts that said they never got it successfully working fine with mb_send_mail.
It took me some time to find something very useful - there is an undocumented php function: mb_orig_mail (
When you use overloaded function you can fall back to the standart mail function by calling this one.
So here is what I did to solve my problem - I use exactly the same class that used to work until the overload get active.
But when it comes to sending I simply check if the mail function is overloaded and if yes, then call the mb_orig_mail.
Here is the actual code:

$mailFunct = 'mail';
if (ini_get('mbstring.func_overload') == 1 || ini_get('mbstring.func_overload') > 4) {
$mailFunct = 'mb_orig_mail';
if(! $mailFunct($tostr, $subject, $tmpBody, $tmpHeaders) ) {
echo 'sent';
} else {
echo 'error sending';

It's as simple as that.
Hope that helps to anyone.
Comments are welcome.

Test your sites with more than one browser or what web developers use for testing their websites...

In my daily work, I need to test sites I work on with a lot of browsers.

Another common problem is testing a website as a different users - anon users, registered users, admins etc.

This is the list of browsers I use in Linux and Windows.

in my Linux (Debian stable or Ubuntu depends on which machine):

  1. Firefox ( - my favorite! I use it on hourly basis :-). It's a lot of features, plugins and of course something that's a must for a web dev - firebug.

  2. Opera ( - my second choice - Awesome browser!!! Feature rich, perfect for casual user. It has integrated mail client, torrent client. Also has a developer toolbar, but It was hard for me to get used with.

  3. Galeon ( advanced gnome web browser. uses gecko engine. pretty thin and quick.

  4. Epiphany ( - the default gnome web browser. Light, solid, gecko based, won't eat up your memory as firefox does. :-) Of course it misses some features.

  5. Konqueror ( - Default KDE3 web (and file) browser (as all KDE apps - qt based). If you have KDE (and all qt as dependent) libs installed, you can use/install it, otherwise - no point installing so much bloat for nothing. This one uses khtml engine but I've faced a lot of
    annoying problems while using it. It was removed from kde4 [maybe I'm wrong?but it's not default filebrowser anymore? correct me if I'm wrong].

in Windows (XP Pro, yes I have a license coming with my laptop):

  1. Firefox

  2. Opera

  3. Chrome - very light. very fast. uses khtml engine for html/css representation. pretty cool, but not for me. I prefer opera if want a light experience. :-)

  4. Safari - apple browser, but for windows. Uses khtml engine. I don't like it very much, but it's pretty slick (as all mac programs). Used only for testing :-)

  5. Internet explorer7 (and 6, 8 in different VirtualBox machines. DUMB MicroS!!! They don't provide descent way of installing all three on one windows!!!) - maybe I shall keep silence :-) the link below describes it better.

more funny reading: If browsers were women.

I hope this article is helpfull. Please share your experience with me.