floyd's quick notes - life of a sysadmin

My Peavey combo amp

2009-12-26T02:56:00.003+02:00

A few pictures of my Peavey BACKSTAGE 2 guitar combo, with 'Transtube'. Sounds great, and at 10 Watts - I must turn the volume down to 10% or less to avoid being killed by the neighbors :)

Installing Slackware Linux on an EeePC 701 by PXE

2009-12-14T23:13:00.002+02:00

Just a few notes on this matter.
I've used my notebook (running slackware 13.0 too) as a pxe server to install slackware on my Asus EeePC 701.

- Install dhcpd

slackpkg install dhcp

- Sample dhcpd config:

floyd[~]$ cat /etc/dhcpd.conf 
# dhcpd.conf
#
# Configuration file for ISC dhcpd (see 'man dhcpd.conf')
#

option domain-name "eulinux.org";
authoritative;
default-lease-time 600;
max-lease-time 7200;
log-facility local7;
ddns-update-style none;

allow bootp;
option tftp-server-name "10.0.0.1";
next-server 10.0.0.1;

subnet 10.0.0.0 netmask 255.255.255.0 {
    option domain-name "eulinux.org";
    option broadcast-address 10.0.0.255;
    option subnet-mask 255.255.255.0;
    option domain-name-servers 192.168.2.1;
    option routers 10.0.0.1;
    range dynamic-bootp 10.0.0.50 10.0.0.100;
    use-host-decl-names on;
    if substring (option vendor-class-identifier, 0, 9) = "PXEClient" {
 filename "/slackware-13/pxelinux.0";
    }
}

- Mount the slackware 13.0 .iso to some folder, and copy a few files

root[floyd]# mount -o ro,loop -t iso9660 ~/iso/slackware-13.0-install-dvd.iso /mnt/slack/
root[floyd]# mkdir -p /tftpboot/slackware-13/
root[floyd]# cp /usr/lib/syslinux/pxelinux.0 /tftpboot/slackware-13/
root[floyd]# cp /mnt/slack/isolinux/message.txt /tftpboot/slackware-13/
root[floyd]# cp /mnt/slack/isolinux/f2.txt /tftpboot/slackware-13/
root[floyd]# cp -a /mnt/slack/kernels/ /tftpboot/slackware-13/
root[floyd]# cp -a /mnt/slack/usb-and-pxe-installers/pxelinux.cfg_default /tftpboot/slackware-13/pxelinux.cfg/default
root[floyd]# cp -a /mnt/slack/isolinux/initrd.img /tftpboot/slackware-13/

- Sample pxe loader config (/tftpboot/slackware-13/pxelinux.cfg/default):

default hugesmp.s
prompt 1
timeout 1200
display message.txt
F1 message.txt
F2 f2.txt
label huge.s
  kernel kernels/huge.s/bzImage
  append initrd=initrd.img load_ramdisk=1 prompt_ramdisk=0 rw SLACK_KERNEL=huge.s
label hugesmp.s
  kernel kernels/hugesmp.s/bzImage
  append initrd=initrd.img load_ramdisk=1 prompt_ramdisk=0 rw SLACK_KERNEL=hugesmp.s
label speakup.s
  kernel kernels/speakup.s/bzImage
  append initrd=initrd.img load_ramdisk=1 prompt_ramdisk=0 rw SLACK_KERNEL=speakup.s

- Also we must export /mnt/slack/ directory by nfs:

echo "/mnt/memory 192.168.2.0/24(ro,all_squash) 10.0.0.0/24(ro,all_squash)" >> /etc/exports

- Start up dhcpd, start your eeepc, press as the system boots, and choose the pxe boot option.
- The rest is fairly simple, usual slackware install, except that when selecting the source, you will choose "NFS" over "CD/DVD"

Have fun!

RWhoisd, indexer issue on Linux

2009-12-05T11:12:00.003+02:00

Getting `sort...` errors when running rwhois_indexer on Linux? (I was experiencing this on CentOS 5)

[root@server html]# /usr/local/rwhoisd/bin/rwhois_indexer -c /usr/local/rwhoisd/etc/rwhoisd.conf -v -A 203.23.23.0/22 -C network -s txt
sort: open failed: +4: No such file or directory
error: sort failed: Success
warning: could not delete temporary index file '': Success
warning: could not delete temporary index file '': Success
indexing failed

This is caused by the parameters used by the indexer when using the `sort` command.
In mkdb/index.c:

#ifdef NEW_STYLE_BIN_SORT
#define SORT_COMMAND "sort -o %s -k 5,5 -k 4,4n -t : %s"
#else
#define SORT_COMMAND "sort -o %s +4 +3 -t : %s "
#endif

Just add before those lines:

#define NEW_STYLE_BIN_SORT

Run a

make clean && make && make install

Nirvana - Seasons in the Sun - Kurt Cobain on drums! :)

2009-11-21T20:49:00.002+02:00

Easily converting your .flac files to .mp3 format

2009-10-17T09:11:00.005+03:00

flac2mp3 is a tool to convert potentially large sets of FLAC audio files into the MP3 audio format. It is implemented as a bash shell script, and should run on any Unix-like operating system.
You can get it at: bytemonkey.org
You'll have to install FLAC and LAME on your linux box, on debian/ubuntu this is easily done with

apt-get install lame flac

After you download flac2mp3 and untar it somewhere, run the script, feeding it with the source and destination directories:

/path/to/flac2mp3 /home/music/JimiHendrix_Experience/ /home/music/mp3/JimiHendrix_Experience/

Stresstesting IMAP clients

2009-10-07T00:48:00.003+03:00

Just found an interesting article about IMAP clients comparison:
http://www.hackvalue.nl/en/article/64/stresstesting_imap_clients

Personally I'm using claws-mail, with 4 mail accounts, one of which has more than 250000 messages.

Migrating to it from thunderbird was fairly easy, I've converted my thunderbird filters to claws' format, using http://www.claws-mail.org/tools/claws-mail-thunderbird-filters-convertor.tar.gz
(See http://www.claws-mail.org/tools.php)

my ~/.fluxbox/keys

2009-10-07T00:45:00.000+03:00

!mouse actions added by fluxbox-update_configs
OnTitlebar Mouse2 :StartTabbing

!mouse actions added by fluxbox-update_configs
OnTitlebar Double Mouse1 :Shade
OnTitlebar Mouse3 :WindowMenu

!mouse actions added by fluxbox-update_configs
OnWindow Mod1 Mouse1 :MacroCmd {Raise} {Focus} {StartMoving}
OnWindow Mod1 Mouse3 :MacroCmd {Raise} {Focus} {StartResizing BottomRight}

!mouse actions added by fluxbox-update_configs
OnToolbar Mouse4 :NextWorkspace
OnToolbar Mouse5 :PrevWorkspace

!mouse actions added by fluxbox-update_configs
OnDesktop Mouse1 :hideMenus
OnDesktop Mouse2 :workspaceMenu
OnDesktop Mouse3 :rootMenu
OnDesktop Mouse4 :nextWorkspace
OnDesktop Mouse5 :prevWorkspace

Mod1 Tab :NextWindow
Mod1 Shift Tab :PrevWindow
Control F1 :Workspace 1
Control F2 :Workspace 2
Control F3 :Workspace 3
Control F4 :Workspace 4
Control F5 :Workspace 5
Control F6 :Workspace 6
Control F7 :Workspace 7
Control F8 :Workspace 8
Control F9 :Workspace 9
Control F10 :Workspace 10
Control F11 :Workspace 11
Control F12 :Workspace 12
Mod1 F2 :ExecCommand fbrun
Mod4 m :RootMenu
Mod4 r :ExecCommand urxvt
Mod4 a :ExecCommand aterm
Mod4 x :ExecCommand mousepad
Mod4 t :ExecCommand konsole
Mod4 f :ExecCommand firefox
Mod4 m :ExecCommand mozilla
Mod4 g :ExecCommand gimp
Mod4 y :ExecCommand thunar
Mod1 F4 :Close
Mod4 F1 :ToggleDecor
Mod4 F2 :MaximizeWindow
Mod4 l :ExecCommand xlock

DIY: New saddle for your bike

2009-09-20T22:06:00.009+03:00

The "leather" on the bike saddle was all worn off, so i had two options:

buy a new saddle
"fix" the existing one

For this, I used two pieces of jeans fabric, and some glue
First, take the saddle off your bike

Take all the parts away

Remove the leather, it could be glued to the saddle, so you will have to use some scissors or a knife

Use a felt-tip pen to draw the approximate shape of the leather to the fabric you are going to use for the saddle,

Cut the fabric according to the needed shape, apply the glue on the seat's margins

Glue the fabric to the plastic saddle, and allow the glue to dry for half an hour (or more, depending on the glue you have used.
After that, you should cut the excess of fabric, and maybe apply some more glue at the sharp corners.

Mount the top part to the saddle, add all the parts back (nuts, bolts, etc)

Mount the seat back on the bike

That'd be all :)

exim + dovecot + clamav + postfixadmin

2009-09-14T23:20:00.004+03:00

Ok, here we go...

I needed to quickly configure a mailserver, initially - for a single domain (but probably will host more later).

I am using exim over the last two years, I don't like qmail because it's very old, and to add some new feature to it - requires patching it. BTW, did you ever try reading qmail's logs? :D
I don't like postfix, but this is probably because I didn't use it too often.

Also, as an Imap/POP3 server, I do like dovecot, easy to configure, yet very comfortable to use, and also quite stable.
Adding domains/users manually isn't too much fun, we need some sort of web interface too, to manage virtual domains, mailboxes, aliases.

Prerequisites:
- apache
- php
- mysql

I will leave this as an exercise to the reader :)

All this setup was done on a CentOS 5.3 system, so I'll give examples using yum to install the packages (on Fedora/RedHat you could use the same yum, on Debian/Ubuntu - use aptitude or apt-get)

First of all, create a MySQL database to store all the settings:

mysql> CREATE DATABASE vexim;
mysql> GRANT ALL ON vexim.* to vexim@localhost identified by 'kieHeed3ailuph8b';

Next, download the latest postfixadmin tarball from http://downloads.sourceforge.net/project/postfixadmin/postfixadmin/postfixadmin-2.2.1.1/postfixadmin-2.2.1.1.tar.gz, and unpack it to /var/www/html/mailadmin/

Adjust: /var/www/html/mailadmin/config.inc.php, more exactly, the following section:

...
// Database Config
// mysql = MySQL 3.23 and 4.0
// mysqli = MySQL 4.1
// pgsql = PostgreSQL
$CONF['database_type'] = 'mysql';
$CONF['database_host'] = 'localhost';
$CONF['database_user'] = 'vexim';
$CONF['database_password'] = 'kieHeed3ailuph8b';
$CONF['database_name'] = 'vexim';
$CONF['database_prefix'] = '';
...

We will need clamd installed, in order to enable virus checks in exim.
Debian/Ubuntu have this in the base repositories, CentOS - doesn't.
We can use the rpmforge packages, see http://dag.wieers.com/rpm/FAQ.php#B2 on how to add the rpmforge repositories to your system.

Install clamd:

[root@server #] yum install clamd

Adjust the following parameters in clamd settings:

#In /etc/clamd.conf:
...
User exim
...
#In /etc/freshclam.conf:
...
DatabaseOwner exim
...

Adjust permissions/ownership:

[root@server #] chown -R exim:exim /var/log/clamav/
[root@server #] chown -R exim:exim /var/clamav/

Now on to exim, dovecot and postfixadmin.
Install exim and dovecot from repositories,

[root@server #] yum install dovecot exim
[root@server #] mv /etc/exim/exim.conf /etc/exim/exim.conf-orig
[root@server #] mv /etc/dovecot.conf /etc/dovecot.conf-orig

Sample config used for exim, this doesn't include spam filtering, or rbl checks, but this can be easily added:
===============================

primary_hostname = mail.eulinux.org
hide mysql_servers = localhost/vexim/vexim/kieHeed3ailuph8b

domainlist local_domains = ${lookup mysql{SELECT `domain` \
                            FROM `domain` WHERE \
                            `domain`='${quote_mysql:$domain}' AND \
                            `active`='1'}}

domainlist relay_to_domains = ${lookup mysql{SELECT `domain` \
                            FROM `domain` WHERE \
                            `domain`='${quote_mysql:$domain}' AND \
                            `active`='1'}}

hostlist   relay_from_hosts = localhost:127.0.0.0/8:192.168.0.0/16  #add the hosts from which you allow relaying here

acl_smtp_rcpt = acl_check_rcpt
acl_smtp_data = acl_check_data

av_scanner = clamd:/var/run/clamav/clamd.sock

# spamd_address = 127.0.0.1 783

qualify_domain = mail.playlainc.com
qualify_recipient = mail.playlainc.com
allow_domain_literals = false
exim_user = exim
exim_group = exim
never_users = root
rfc1413_query_timeout = 0s

sender_unqualified_hosts = +relay_from_hosts
recipient_unqualified_hosts = +relay_from_hosts

ignore_bounce_errors_after = 45m
timeout_frozen_after = 15d
helo_accept_junk_hosts = 192.168.0.0/16
auto_thaw = 1h
smtp_banner = "$primary_hostname, ESMTP EXIM $version_number"
smtp_accept_max = 50
smtp_accept_max_per_connection = 25
smtp_connect_backlog = 30
smtp_accept_max_per_host = 20
split_spool_directory = true
remote_max_parallel = 15
return_size_limit = 70k
message_size_limit = 64M
helo_allow_chars = _
smtp_enforce_sync = true

log_selector = \
    +all_parents \
    +connection_reject \
    +incoming_interface \
    +lost_incoming_connection \
    +received_sender \
    +received_recipients \
    +smtp_confirmation \
    +smtp_syntax_error \
    +smtp_protocol_error \
    -queue_run

syslog_timestamp = no

begin acl

acl_check_rcpt:
  accept  hosts = :
  deny    message       = "incorrect symbol in address"
          domains       = +local_domains
          local_parts   = ^[.] : ^.*[@%!/|]

  deny    message       = "incorrect symbol in address"
          domains       = !+local_domains
          local_parts   = ^[./|] : ^.*[@%!] : ^.*/\\.\\./

  accept  local_parts   = postmaster
          domains       = +local_domains

  deny    message       = "HELO/EHLO required by SMTP RFC"
          condition     = ${if eq{$sender_helo_name}{}{yes}{no}}

  accept  authenticated = *

  deny    message       = "Your IP in HELO - access denied!"
          hosts         =  * : !+relay_from_hosts : !81-196.lissyara.su
          condition     = ${if eq{$sender_helo_name}\
    {$sender_host_address}{true}{false}}

  deny    condition     = ${if eq{$sender_helo_name}\
    {$interface_address}{yes}{no}}
          hosts         = !127.0.0.1 : !localhost : *
          message       = "My IP in your HELO! Access denied!"

  deny    condition     = ${if match{$sender_helo_name}\
    {\N^\d+$\N}{yes}{no}}
          hosts         = !127.0.0.1 : !localhost : *
          message       = "Incorrect HELO string"

# filter spammers from dynamic ips
  deny    message       = "your hostname is bad (adsl, poll, ppp & etc)."
          condition     = ${if match{$sender_host_name} \
                               {adsl|dialup|pool|peer|dhcp} \
                               {yes}{no}}

  warn
        set acl_m0 = 30s
  warn
        hosts = +relay_from_hosts:4.3.2.1/32:192.168.0.0/16 #disable waits for 'friendly' hosts
        set acl_m0 = 0s
  warn
        logwrite = Delay $acl_m0 for $sender_host_name \
[$sender_host_address] with HELO=$sender_helo_name. Mail \
from $sender_address to $local_part@$domain.
        delay = $acl_m0

                          
  accept  domains       = +local_domains
          endpass
          message       = "No such user"
          verify        = recipient

  accept  domains       = +relay_to_domains
          endpass
          message       = "i don't know how to relay to this address"
          verify        = recipient

#  deny    message       = "you in blacklist - $dnslist_domain \n $dnslist_text"
#          dnslists      = opm.blitzed.org : \
#                          cbl.abuseat.org : \
#                          bl.csma.biz 

  accept  hosts         = +relay_from_hosts

  deny    message       = "Homo hominus lupus est"

acl_check_data:

  # check for viruses
  deny malware = *
  message = "Your message contains viruses: $malware_name"

  # if needed - add spam filtering here

  # permit everything else
  accept

begin routers

dnslookup:
  driver = dnslookup
  domains = ! +local_domains
  transport = remote_smtp
  ignore_target_hosts = 0.0.0.0 : 127.0.0.0/8
  no_more

system_aliases:
    driver      = redirect
    allow_fail
    allow_defer
    data = ${lookup mysql{SELECT `goto` FROM `alias` WHERE \
            `address`='${quote_mysql:$local_part@$domain}' OR \
                `address`='${quote_mysql:@$domain}'}}

dovecot_user:
  driver = accept
  condition = ${lookup mysql{SELECT `goto` FROM \
  `alias` WHERE \
  `address`='${quote_mysql:$local_part@$domain}' OR \
  `address`='${quote_mysql:@$domain}'}{yes}{no}}
  transport = dovecot_delivery

begin transports

remote_smtp:
  driver = smtp
  interface = 4.3.2.1 #your public ip address, if it's the main server ip - you could skip this parameter. if it's an alias- it must be specified in clear


dovecot_delivery:
  driver = pipe
  command = /usr/libexec/dovecot/deliver -d $local_part@$domain
  message_prefix =
  message_suffix =
  delivery_date_add
  envelope_to_add
  return_path_add
  log_output
  user = exim

address_pipe:
  driver = pipe
  return_output

address_reply:
  driver = autoreply

begin retry

*                    *       F,2h,15m; G,16h,1h,1.5; F,4d,6h

begin rewrite

begin authenticators

auth_plain:
  driver = plaintext
  public_name = PLAIN
  server_condition = ${lookup mysql{SELECT `username` FROM \
                     `mailbox` WHERE `username` = \
                     '${quote_mysql:$auth2}' AND `password` = \
                     '${quote_mysql:$auth3}'}{yes}{no}}
  server_prompts = :
  server_set_id = $auth2

auth_login:
  driver = plaintext
  public_name = LOGIN
  server_condition = ${lookup mysql{SELECT `username` FROM \
                     `mailbox` WHERE `username` = \
                     '${quote_mysql:$auth1}' AND `password` = \
                     '${quote_mysql:$auth2}'}{yes}{no}}
  server_prompts = Username:: : Password::
  server_set_id = $auth1

auth_cram_md5:
  driver = cram_md5
  public_name = CRAM-MD5
  server_secret = ${lookup mysql{SELECT `password` FROM \
                        `mailbox` WHERE `username` \
                        = '${quote_mysql:$auth1}'}{$value}fail}
  server_set_id = $auth2

Now, on to dovecot:

#File: /etc/dovecot.conf
base_dir = /var/run/dovecot/
protocols = imap pop3

listen = *
disable_plaintext_auth = no
shutdown_clients = yes
log_timestamp = "%b %d %H:%M:%S "
syslog_facility = mail
ssl_disable = yes

login_dir = /var/run/dovecot/login
login_chroot = no

login_process_size = 64
login_process_per_connection = yes
login_processes_count = 3
login_max_processes_count = 128
login_max_connections = 256
login_greeting = Dovecot ready man.
login_log_format_elements = user=<%u> method=%m rip=%r lip=%l %c
login_log_format = %$: %s
mail_location = maildir:~/home/vmail/%d/%n

first_valid_uid=93
last_valid_uid=0

protocol imap {
  mail_plugins = quota imap_quota
  imap_client_workarounds = outlook-idle netscape-eoh tb-extra-mailbox-sep
}
 
protocol pop3 {
  pop3_uidl_format = %08Xu%08Xv
  pop3_client_workarounds = outlook-no-nuls oe-ns-eoh
}

protocol lda {
  postmaster_address = admin@eulinux.org
  auth_socket_path = /var/run/dovecot/auth-master
}

auth default {
  mechanisms = plain
  passdb pam {
  }

  passdb sql {
    args = /etc/dovecot-mysql.conf
  }

  userdb passwd {
  }

  userdb sql {
    args = /etc/dovecot-mysql.conf
  }
  user = root
  socket listen {
    master {
      path = /var/run/dovecot/auth-master
      mode = 0660
      user = exim
      group = exim
    }
  }
}

dict {
}

plugin {
}

#File: /etc/dovecot-mysql.conf
driver = mysql
connect = host=localhost dbname=vexim user=vexim password=kieHeed3ailuph8b
default_pass_scheme = CRYPT
password_query = SELECT `username` as `user`, `password` FROM  `mailbox` WHERE `username` = '%n@%d' AND `active`='1'
user_query = SELECT "" AS `home`, 93 AS `uid`, 93 AS `gid` FROM `mailbox` WHERE `username` = '%n@%d' AND `active`='1'

Now, start the services:

for srv in exim dovecot clamd; 
do
 service ${srv} start
done

Access your postfixadmin configuration file at:
http://hostname/mailadmin/setup.php

You should see a list of 'OK' messages.

The setup.php script will attempt to create the database structure
(or upgrade it if you're coming from a previous version).

Assuming everything is OK you can create the admin user using the form displayed.

Once you submit the form, all that's left to do is to delete "setup.php"

That seems to be all! ;)

If you notice any errors or have any questions on this - please comment

Iptraf with bonding support

2009-05-29T12:55:00.002+03:00

Default iptraf build that comes with RedHat or CentOS does not support bondX interfaces, or vlan (ethX.Y) interfaces,
however there's a version on bugzilla.redhat.com that does support it.
More details:
https://bugzilla.redhat.com/show_bug.cgi?id=441081
Also, for you the lazy ones out there, quick install procedure (done on CentOS5 x86_64) :

#remove current iptraf version
root@host ~# yum remove iptraf
root@host ~# wget "https://bugzilla.redhat.com/attachment.cgi?id=304745" -O iptraf-3.0.0-6.el5.src.rpm
root@host ~# rpmbuild --rebuild iptraf-3.0.0-6.el5.src.rpm
root@host ~# rpm -Uvh /usr/src/redhat/RPMS/x86_64/iptraf-3.0.0-6.x86_64.rpm

Ready to go.
If there's no `rpmbuild` command, install rpm-build, and also you'll have to add some packages like `ncurses-devel`

Detecting Conficker with nmap

2009-04-01T16:13:00.001+03:00

As the Conficker worm is vastly discussed and getting a lot of attention, latest nmap version can detect it, won't post here how, here's the direct link to the resource:

http://www.net-security.org/secworld.php?id=7252

Restoring the MBR on RHEL/CentOS servers

2009-03-16T13:08:00.003+02:00

It happens that the MBR is getting bad (after os updates/power failures, etc)
We can restore it in a few minutes if we have a RHEL/CentOS install CD at hand, or if we have a PXE server nearby.

Boot the system from the install cd,
at the boot prompt choose:

text rescue

After booting the rescue system, perform the following commands (actually only the chroot and grub-install may be needed, but sometimes the rescue system does not see all devices under the mounted system tree):

chroot /mnt/sysimage
mount -t proc proc /proc
mount -t sysfs sysfs /sys
mount -t tmpfs tmpfs /dev
mkdir dev/pts
mount -t devpts devpts /dev/pts
/sbin/udevstart
/sbin/grub-install hd0
quit
quit

The system will reboot, and hopefully all will work right away.
If it doesn't - you will need to dig further :)
P.S - Don't sue me if these commands do some nasty things to your server ;)

Quick shell script to convert ifcfg-ethX-rangeY to ifcfg-ethX:Y format

2009-03-04T16:47:00.001+02:00

Quick shell script to convert ifcfg-ethX-rangeY to ifcfg-ethX:Y format:

for i in `ls ifcfg-eth1-range*`; 
do
. ./$i
D1=`echo $IPADDR_START|cut -d. -f4`;
D2=`echo $IPADDR_END|cut -d. -f4`;
D=`expr $D2 - $D1`;
echo $D;
S=`echo $IPADDR_START|cut -d. -f1,2,3`;
for j in `seq 0 $D`;
do
echo "DEVICE=eth1:`expr $j + $CLONENUM_START`
BOOTPROTO=static
ONBOOT=yes
IPADDR=$S.`expr $D1 + $j`
NETMASK=$NETMASK" > ifcfg-eth1:`expr $j + $CLONENUM_START`;
done
done

Freebsd php-fpm fastcgi server init script

2009-01-29T12:54:00.000+02:00


#!/bin/sh

# PROVIDE: phpfpm
# KEYWORD: shutdown

#
# Add the following lines to /etc/rc.conf to enable php-fpm:
#
#phpfpm_enable="YES"
#
#

. /etc/rc.subr

name=phpfpm
rcvar=`set_rcvar`

command="/usr/local/sbin/php-fpm"
command_args=""
pidfile=/usr/local/logs/php-fpm.pid
required_files=/usr/local/etc/php-fpm.conf

# set defaults

phpfpm_enable=${phpfpm_enable:-"NO"}

start_cmd="phpfpm_start"
stop_cmd="phpfpm_stop"
restart_cmd="phpfpm_restart"

phpfpm_start()
{
 if [ -f ${pidfile} ]
 then
  echo "php-fpm already running!"
  exit 1;
 fi
 echo "Starting php-fpm server..."
 /usr/local/bin/php-cgi --fpm
}
phpfpm_stop()
{
 if [ ! -f ${pidfile} ]
 then
  echo "${name} is not running or pidfile not found!";
  exit 1;
 fi
 echo "stopping ${name}..."
 kill -SIGQUIT `cat ${pidfile}`
} 
phpfpm_restart()
{
        if [ ! -f ${pidfile} ]
        then
                echo "${name} is not running or pidfile not found!";
                exit 1;
        fi
 echo "sending SIGUSR2 to php-fpm master"
        kill -SIGUSR2 `cat ${pidfile}`
}

load_rc_config ${name}
run_rc_command "$1"

How to check SpamAssassin version

2009-01-21T12:51:00.001+02:00

Quick one-liner for checking spamassassin version:

[root@server ~]# perl -MMail::SpamAssassin -e 'print $Mail::SpamAssassin::VERSION;'
3.002004

OpenSUSE developers announced csync

2009-01-13T05:15:00.002+02:00

Andreas Schneider has unleashed the first public release candidate of csync 0.42, which is now available as source from the csync Web site, or via one-click for openSUSE 11.0 and 11.1.

Project homepage: http://www.csync.org/

This tool looks like a promising alternative to unison, which was written in ocaml, while csync is written in C

Did not have time to test it yet, but as soon as it proves that is working ok - i plan to replace unison with it.

RDP through a ssh tunnel

2009-01-13T04:54:00.005+02:00

How to access RDP if firewall is closed down, and access to server is restricted to a few ips:

ssh user@allowed.server.com -L 3389:windows.server.com:3389
rdesktop -uadministrator -pp455w0rd localhost

That's it! Secure encrypted rdesktop session :)

Parallels decided to drop qmail?

2009-01-01T20:44:00.003+02:00

While installing latest Parallels Plesk version (9.0), noticed an unusual option :) :

  Different mailservers
16. (*) Qmail mailserver
17. ( ) Postfix mailserver

Could this be true, that Parallels finally decided to drop Qmail?
Now don't get me wrong, but qmail wasn't updated for 10 years already, and any additional feature requires a patch...
And.. did you ever try to read qmail's logs? :)

2009 will be a second longer | Happy new year!

2009-01-01T17:14:00.004+02:00

Found a confirmation of this in my system logs:

Jan  1 01:59:59 (none) kernel: Clock: inserting leap second 23:59:60 UTC

:)

Btw,


 _________________
( Happy new Year! )
 -----------------
     \   ^__^
      \  (oo)\_______
         (__)\       )\/\
             ||----w |
             ||     ||

Icewarp Merak mail server not working behind a smtp relay

2008-12-24T09:24:00.005+02:00

Just migrated an email server to a dedicated server at GoDaddy, and according to their policies, 25 TCP port is blocked "to prevent spam", and they force us to use their smtp relay for mail delivery.

After properly configuring all delivery related settings, server still could not send emails.

A quick look in smtp logs:

SYSTEM          [087C] 23:03:41 Client session ***   1 569 00:00:00 INCOMPLETE-SESSION ESZ92941
SYSTEM          [0BE8] 23:07:55 Client session ***   1 569 00:00:00 INCOMPLETE-SESSION ESZ92941

Strange.
Turned on debugging:

SYSTEM          [04C4] 23:42:26 Client session Message id ESO58526 item 200812232342260002.tm$
SYSTEM          [04C4] 23:42:26 Client session Connecting to 'smtprelay.server.net'
SYSTEM          [04C4] 23:44:26 Client session Could not connect to 'smtprelay.server.net(0.0.0.0)'

This "0.0.0.0" looked strange to me, it seems there was some issues with dns.
After a little digging in Merak's settings, it appears that actually the mail server does not use system default resolvers, and uses it's own.

Just adjusted dns settings in System>Internet Connection>Dns, it started to work like a charm

Installing SVN server on FreeBSD

2008-12-18T00:46:00.003+02:00

Quick installation of SVN on FreeBSD, access by svn:// protocol

portinstall subversion
mkdir /home/svn-repo
pw user add -n svn -u 3690 -d /home/svn-repo
chown svn:svn /home/svn-repo/
chsh -s /bin/false svn

Add to /etc/rc.conf:

svnserve_enable="YES"
svnserve_data="/home/svn-repo"
svnserve_user="svn"
svnserve_group="svn"

Start the server itself:

/usr/local/etc/rc.d/svnserve start

Also, remember to open port 3690 in firewall, if you are using one

vsftpd - chroot users to their homedir

2008-11-30T17:54:00.003+02:00

Open vsftpd configuration file - /etc/vsftpd/vsftpd.conf

nano /etc/vsftpd/vsftpd.conf

Make sure following line exists (and uncommented):

chroot_local_user=YES

Save and close the file. Restart vsftpd.

/etc/init.d/vsftpd restart

ScribeFire

2008-11-30T03:16:00.000+02:00

Installed ScribeFire, a nice blog editor running as a firefox add-on, pretty impressed, now it's easier to post dirrectly from my browser, no need to login to blogger for this ;)

Using remote logging on a dd-wrt enabled Linksys wrt54g

2008-09-16T01:34:00.007+03:00

On the machine you want to log to (i have a slackware one):
root[floyd]# nano /etc/rc.d/rc.syslog
replace:

echo -n "/usr/sbin/syslogd"
/usr/sbin/syslogd

with:

echo -n "/usr/sbin/syslogd -r "
/usr/sbin/syslogd -r

root[floyd]# /etc/rc.d/rc.syslog restart
Starting sysklogd daemons:  /usr/sbin/syslogd -r /usr/sbin/klogd -c 3 -x
root[floyd]#

After that, all you have to do is:
Login to dd-wrt (e.g. http://192.168.2.1/ - please replace with actual IP fof your wireless router )
Go to Services -> System Log -> Syslogd -> Enable
Also specify the IP address of the logging server.

On the logging server itself, remember to put some kind of filtering, to prevent any other device to log to it.
I'd go for iptables.

MySQL master-master replication

2008-09-03T10:38:00.003+03:00

A quick howto for setting up mysql master-master replication (of course IP adresses and passwords are fake :)
http://dev.mysql.com/doc/refman/5.0/en/server-system-variables.html

Used variables:
- server-id: The server ID. This value is set by the --server-id option. It is used for replication to enable master and slave servers to identify themselves uniquely.

- auto_increment_increment and auto_increment_offset are intended for use with master-to-master replication, and can be used to control the operation of AUTO_INCREMENT columns. Both variables can be set globally or locally, and each can assume an integer value between 1 and 65,535 inclusive. Setting the value of either of these two variables to 0 causes its value to be set to 1 instead. Attempting to set the value of either of these two variables to an integer greater than 65,535 or less than 0 causes its value to be set to 65,535 instead. Attempting to set the value of auto_increment_increment or auto_increment_offset to a non-integer value gives rise to an error, and the actual value of the variable remains unchanged.
These two variables affect AUTO_INCREMENT column behavior as follows:

One thing that is missing from the above, is how to deal with auto-increments so that you do not get collisions if you add data to both masters at the same time. There is an easy way to deal with this if you are using MySQL 5.0.2 or above.
In the my.cnf file on server1 add:

auto_increment_increment=2
auto_increment_offset=1

In the my.cnf file on server2 add:

auto_increment_increment=2
auto_increment_offset=2

This will make the auto-increment on server1 go, 1,3,5,7,9,etc... and on server2 go, 2,4,6,8,etc... thereby preventing collisions.

server1:

mysql> GRANT REPLICATION SLAVE ON *.* TO 'server2'@'67.15.239.12' IDENTIFIED BY 'Sho6ra5aedoh';
Query OK, 0 rows affected (0.00 sec)

mysql> FLUSH PRIVILEGES;
Query OK, 0 rows affected (0.00 sec)mysql> CREATE DATABASE testdb;
Query OK, 1 row affected (0.00 sec)

server2:

mysql> GRANT REPLICATION SLAVE ON *.* TO 'server1'@'67.15.239.11' IDENTIFIED BY 'Sho6ra5aedoh';
Query OK, 0 rows affected (0.00 sec)

mysql> FLUSH PRIVILEGES;
Query OK, 0 rows affected (0.00 sec)
mysql> CREATE DATABASE testdb;
Query OK, 1 row affected (0.00 sec)

server1:/etc/my.cnf

[mysqld]
server-id = 1
replicate-same-server-id = 0
auto-increment-increment = 2
auto-increment-offset = 1

master-host = 67.15.239.12
master-user = server1
master-password = Sho6ra5aedoh
master-connect-retry = 60
replicate-do-db = testdb

log-bin = /var/lib/mysql/mysql-bin.log
binlog-do-db = testdb

relay-log = /var/lib/mysql/slave-relay.log
relay-log-index = /var/lib/mysql/slave-relay-log.index

expire_logs_days        = 10
max_binlog_size         = 500M

server2:/etc/my.cnf

[mysqld]
server-id = 2
replicate-same-server-id = 0
auto-increment-increment = 2
auto-increment-offset = 2

master-host = 67.15.239.11
master-user = server2
master-password = Sho6ra5aedoh
master-connect-retry = 60
replicate-do-db = testdb

log-bin = /var/lib/mysql/mysql-bin.log
binlog-do-db = testdb

relay-log = /var/lib/mysql/slave-relay.log
relay-log-index = /var/lib/mysql/slave-relay-log.index

expire_logs_days        = 10
max_binlog_size         = 500M

Run:

server2:

mysql> stop slave;
mysql> CHANGE MASTER TO MASTER_HOST='67.15.239.11' , MASTER_USER='server2', MASTER_PASSWORD='Sho6ra5aedoh',MASTER_LOG_FILE='mysql-bin.000001', MASTER_LOG_POS=98;
mysql> start slave;
mysql> show slave status \G
mysql> show master status;

server1:


mysql> STOP SLAVE;
mysql> CHANGE MASTER TO MASTER_HOST='67.15.239.12' , MASTER_USER='server1', MASTER_PASSWORD='Sho6ra5aedoh',MASTER_LOG_FILE='mysql-bin.000001', MASTER_LOG_POS=98;
mysql> start slave;
mysql> show master status;
mysql> show slave status \G