Fsk141

# # /etc/sudoers -- visudo # Cmnd_Alias SUSPEND = /usr/sbin/pm-suspend Cmnd_Alias INTERNET = /usr/bin/netcfg, /sbin/ifconfig, /usr/sbin/iwconfig, /usr/sbin/iwlist, /usr/sbin/dhcpcd Cmnd_Alias SAVEPOWER = /home/fsk141/.scripts/autopower root ALL=(ALL) ALL %wheel ALL=(ALL) ALL, NOPASSWD: SUSPEND, NOPASSWD: INTERNET, NOPASSWD: SAVEPOWER

[fsk141@oHai ~]$ ssh-keygen Generating public/private rsa key pair. Enter file in which to save the key (/home/fsk141/.ssh/id_rsa): .meh Enter passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved in .ssh/id_rsa. Your public key has been saved in .ssh/id_rsa.pub. The key fingerprint is: a4:fb:92:4b:33:a5:f6:5b:b8:ce:f7:a1:78:73:4a:f0 fsk141@oHai The key's randomart image is: +--[ RSA 2048]----+ | | | | | . | | o | | . S | | + + | | B.. E . | | oo* =+... | | .+O+o=. | +-----------------+

## if you have ssh-copy-id ## ssh-copy-id myserver.com ## if you don't have ssh-copy-id ## ## you can run the ssh "commands" on you server ## I just made it so everything ## could be done from connecting localhost ssh me@myserver.com 'if [ ! -d ~/.ssh ]; then mkdir ~/.ssh; fi' scp .ssh/id_rsa.pub me@myserver.com:.ssh/authorized_keys ssh me@myserver.com 'chmod 600 ~/.ssh/authorized_keys'

# Digest Squid Auth -- better method # /etc/squid/squid.conf # http_port 3129 auth_param digest program /usr/lib/squid/digest_pw_auth -c /etc/squid/digest_passwd # Make /etc/squid/digest_password this way: ## First get a script... ## wget http://dl.dropbox.com/u/52078/digest_passwd.sh ## Execute something similar to the following ### sh ./digest_passwd.sh your_username_here your_password_here 'Squid proxy-caching web server' > /etc/squid/digest_passwd # need to execute as root/sudo ## This will give you a happy digest_passwd file auth_param digest children 5 auth_param digest realm Squid proxy-caching web server auth_param digest nonce_garbage_interval 5 minutes auth_param digest nonce_max_duration 30 minutes auth_param digest nonce_max_count 50 acl authenticated proxy_auth REQUIRED http_access allow authenticated

# Make /etc/squid/digest_password this way: ## First get a script... ## wget http://dl.dropbox.com/u/52078/digest_passwd.sh (contents below) ------ #!/bin/sh user=$1 pass=$2 realm=$3 if [ -z "$1" -o -z "$2" -o -z "$3" ] ; then echo "Usage: $0 user password 'realm'"; exit 1 fi ha1=$(echo -n "$user:$realm:$pass"|md5sum |cut -f1 -d' ') echo "$user:$realm:$ha1" ------ ## Execute something similar to the following ### sh ./digest_passwd.sh your_username_here your_password_here 'Squid proxy-caching web server' > /etc/squid/digest_passwd # need to execute as root/sudo ## This will give you a happy digest_passwd file

# Plaintext Authentication Squid Setup # /etc/squid/squid.conf # http_port 3129 #default port to connect with auth_param basic program /usr/lib/squid/ncsa_auth /etc/squid/passwd # might need to change paths dependant on distro #make /etc/squid/passwd with the following: ## htpasswd /etc/squid/passwd your_username_here # execute as root/sudo auth_param basic children 5 auth_param basic realm Squid proxy-caching web server auth_param basic credentialsttl 2 hours auth_param basic casesensitive off # acl allow rules acl authenticated proxy_auth REQUIRED http_access allow authenticated

# On your HN run: # Download the debian 5 template sudo wget -c http://download.openvz.org/template/precreated/debian-5.0-x86_64.tar.gz -o /vz/template/cache/debian-5.0-x86_64.tar.gz # Create debian VE sudo vzctl create 1 --ostemplate debian-5.0-x86_64 # Hooray your VE is setup, now move onto configuration >>

# Configure debian VE (naibed-zen is debian backwards with a lil zen added) # You can enter this all on one line, I just spread it out to make it easier to read sudo vzctl set 1 --hostname naibed-zen --ipadd <ipaddress> --searchdomain <yoursearchdomain> --nameserver "<your nameserver/nameservers (separated by spaces)> --vmguarpages $((256 * 4096)) --privvmpages $((256 * 6144)) --swappages $((256 * 1024)) --meminfo none --onboot yes --save #--vmguarpages $((256 * 4096)) # guaranteed memory (4GB) #--privvmpages $((256 * 6144)) # burst memory (6GB) #--swappages $((256 * 1024)) # swap memory #--meminfo none # I was having memory listing issues, and this fixed it

[jgerold@jg-iMac ~]$ sudo vzctl enter 2 entered into CT 2 naibed:/# echo "deb http://dev.zenoss.org/deb main stable" >> /etc/apt/sources.list naibed:/# apt-get update ... updating pkg-database ... naibed:/# apt-cache search zenoss-stack # just verifying that the repo is correct zenoss-stack - Zenoss Stack with all requirements. naibed:/# apt-get install zenoss-stack Reading package lists... Done Building dependency tree... Done The following NEW packages will be installed: zenoss-stack 0 upgraded, 1 newly installed, 0 to remove and 14 not upgraded. Need to get 110MB of archives. After this operation, 386MB of additional disk space will be used. WARNING: The following packages cannot be authenticated! zenoss-stack Install these packages without verification [y/N]? Y Get:1 http://dev.zenoss.org main/stable zenoss-stack 3.0.2-0 [110MB] Fetched 110MB in 11s (9756kB/s) Selecting previously deselected package zenoss-stack. (Reading database ... 23039 files and directories currently installed.) Unpacking zenoss-stack (from .../zenoss-stack_3.0.2-0_amd64.deb) ... Setting up zenoss-stack (3.0.2-0) ... naibed:/# /etc/init.d/zenoss-stack start nohup: redirecting stderr to stdout Starting mysqld.bin daemon with databases from /usr/local/zenoss/mysql/data /usr/local/zenoss/mysql/scripts/ctl.sh : mysql started at port 3307 Daemon: zeoctl . daemon process started, pid=2050 Daemon: zopectl . daemon process started, pid=2061 Daemon: zenhub starting... Daemon: zenjobs starting... Daemon: zenping starting... Daemon: zensyslog starting... Daemon: zenstatus starting... Daemon: zenactions starting... Daemon: zentrap starting... Daemon: zenmodeler starting... Daemon: zenperfsnmp starting... Daemon: zencommand starting... Daemon: zenprocess starting... Daemon: zenwin starting... Daemon: zeneventlog starting... naibed:/# # You're done, move onto testing & pat yourself on the back

[Tutorial] Add more RAM to an OpenVZ container

October 7th, 2010 by fsk141 It seems like a simple idea, but adding ram to an OpenVZ container is a little bit tricky. Lets say we would like to have 512MB dedicated & 1024MB Burst In OpenVZ l33t Sp3ak it would look something like this: [code language="bash"] vzctl set <vpsid> --vmguarpages 131072 --save # 512MB Dedicated (Guaranteed) vzctl set <vpsid> --privvmpages 262144 --save # 1024 Burst (Granted) [/code] For the longest time it was a pain in the ass to grasp this concept, and I would use someone else's config file & just copy those properties when I knew they allocated 512MB ram, or the like. Anywho I don't know why the OpenVZ wiki can't be simpler, or their 'vzctl' program for that matter. Why can't you just specify the KB/MB/GB value that you want??? Well they don't make it easy, and here's "The Easy Way" [code language="bash"] vzctl set <vpsid> --vmguarpages $((256 * 512)) --save # 512MB Dedicated (Guaranteed) // 512MB * <whatever_MB_you_want> vzctl set <vpsid> --privvmpages $((256 * 1024)) --save # 1024 Burst (Granted) // 1024MB * <whatever_MB_you_want> [/code] Q: Why & how does this sorcery work? A: Lets pick apart the commands declaration (--vmguarpages 131072) 131072 (number of pages)*4096(how big a page is)/1024 (knock up to KB)/1024 (knock up to MB) [code language="plain"]((131072*4096)/1024)/1024[/code] Oh, ok great, that really explains things? (you might say)? Well lemme go a little deeper & give you a little example to execute on your Host Node (HN) [code language="cpp"] // Find the pagesize of your machine (x86/x86_64 should be 4096) #include <unistd.h> #include <stdio.h> int main(int argc, char* argv[]) { long sz = sysconf(_SC_PAGESIZE); printf("Memory pagesize on this box : %i Bytesn", sz); return 0; } // compile & execute // copy to a file (pages.cpp) g++ -o pages pages.cpp ; chmod +x ./pages ; ./pages // you "should" get 4096 Bytes (unless you're on ia64 hardware) [/code] Now that you've verified your page size, lets move onto how 512MB computes to 131072 pages: 4096 Bytes = 1 page = 4 Kilobytes Since there are 1024 KB in a MB lets calculate for pages which is 1024/4 (where 1024 is a MB we are calculating for one page (4 part groups of 1024) == 256 What is 256? 256 is the number of pages in a MB, so if we multiply 256 * <number of MB's RAM we want> that will give us the number of pages to correctly give OpenVZ the amount of pages it needs. [code language="plain"] // some examples 1GB (1024 * 256) = 262144 vzctl set <vpsid> --vmguarpages 262144 --save // if you pre-calculate vzctl set <vpsid> --vmguarpages $((256 * 1024)) --save // this just calculates with bash 16GB (18432 * 256) = 4718592 vzctl set <vpsid> --vmguarpages 4718592 --save vzctl set <vpsid> --vmguarpages $((256 * 18432)) --save [/code] Well I hope this helped someone like me that couldn't for the life of me understand what the hell they were talking about on the OpenVZ wiki. It wasn't until I hit a tiny tidbit of happiness from the linux.com article (where they did 256 * 256), then it took me a little longer reading the Setting_UBC_parameters to figure out page size, and after I knew pages were 4KB, I thought it would be nice to outline my findings. Enjoy :) Sources: http://wiki.openvz.org/Setting_UBC_parameters http://wiki.openvz.org/UBC_primary_parameters http://wiki.openvz.org/UBC_secondary_parameters http://www.linux.com/archive/articles/114214

extra ram

It seems like a simple idea, but adding ram to an OpenVZ container is a little bit tricky.

Lets say we would like to have 512MB dedicated & 1024MB Burst

In OpenVZ l33t Sp3ak it would look something like this:

vzctl set <vpsid> --vmguarpages 131072 --save # 512MB Dedicated (Guaranteed)

vzctl set <vpsid> --privvmpages 262144 --save # 1024 Burst (Granted)

For the longest time it was a pain in the ass to grasp this concept, and I would use someone else’s config file & just copy those properties when I knew they allocated 512MB ram, or the like. Anywho I don’t know why the OpenVZ wiki can’t be simpler, or their ‘vzctl’ program for that matter. Why can’t you just specify the KB/MB/GB value that you want??? Well they don’t make it easy, and here’s “The Easy Way”

vzctl set <vpsid> --vmguarpages $((256 * 512)) --save #  512MB Dedicated (Guaranteed)
// 512MB * <whatever_MB_you_want>

vzctl set <vpsid> --privvmpages $((256 * 1024)) --save # 1024 Burst (Granted)
// 1024MB * <whatever_MB_you_want>

Why & how does this sorcery work?

Lets pick apart the commands declaration (--vmguarpages 131072)

131072 (number of pages)*4096(how big a page is)/1024 (knock up to KB)/1024 (knock up to MB)

((131072*4096)/1024)/1024

Oh, ok great, that really explains things? (you might say)? Well lemme go a little deeper & give you a little example to execute on your Host Node (HN)

// Find the pagesize of your machine (x86/x86_64 should be 4096)

#include <unistd.h>
#include <stdio.h>

int main(int argc, char* argv[])
{
long sz = sysconf(_SC_PAGESIZE);
printf("Memory pagesize on this box : %i Bytesn", sz);
return 0;
}

// compile & execute
// copy to a file (pages.cpp) g++ -o pages pages.cpp ; chmod +x ./pages ; ./pages
// you "should" get 4096 Bytes (unless you're on ia64 hardware)

Now that you’ve verified your page size, lets move onto how 512MB computes to 131072 pages:

4096 Bytes = 1 page = 4 Kilobytes
Since there are 1024 KB in a MB lets calculate for pages which is 1024/4 (where 1024 is a MB we are calculating for one page (4 part groups of 1024) == 256

What is 256? 256 is the number of pages in a MB, so if we multiply 256 * <number of MB’s RAM we want> that will give us the number of pages to correctly give OpenVZ the amount of pages it needs.

// some examples
1GB (1024 * 256) = 262144
vzctl set <vpsid> --vmguarpages 262144 --save // if you pre-calculate
vzctl set <vpsid> --vmguarpages $((256 * 1024)) --save // this just calculates with bash

16GB (18432 * 256) = 4718592
vzctl set <vpsid> --vmguarpages 4718592 --save
vzctl set <vpsid> --vmguarpages $((256 * 18432)) --save

Well I hope this helped someone like me that couldn’t for the life of me understand what the hell they were talking about on the OpenVZ wiki. It wasn’t until I hit a tiny tidbit of happiness from the linux.com article (where they did 256 * 256), then it took me a little longer reading the Setting_UBC_parameters to figure out page size, and after I knew pages were 4KB, I thought it would be nice to outline my findings. Enjoy

Sources:

http://wiki.openvz.org/Setting_UBC_parameters
http://wiki.openvz.org/UBC_primary_parameters
http://wiki.openvz.org/UBC_secondary_parameters
http://www.linux.com/archive/articles/114214

/me elsewhere

Last.fm

Categories

sudo (brief history — & installation) — Part one of multipart series

[tutorial] Disable ssh password login; Enable ssh key auth

[tutorial] Squid Proxy [with digest authentication]

[tutorial] Setup your very own squid proxy [with basic authentication]

[Tutorial] How to unencrypt kindle books

[Tutorial] OpenVZ – Debian – Zenoss (monitor your world)

[Tutorial] Add more RAM to an OpenVZ container

[Tutorial] SSL IIS > Apache (Export Windows Server SSL Cert to Apache)