Disable iptables in a whole cluster

for node in mynode{03,04,05,06,07,08,09,10,11,12}; do ssh root@$node '/etc/init.d/iptables stop; chkconfig iptables off'; done

Scala console

Start with 'sbt console'

scala> import sys.process._
import sys.process._

scala> "ls -al" !
total 64
drwxr-xr-x  10 Al  staff   340 May 18 18:00 .
drwxr-xr-x   3 Al  staff   102 Apr  4 17:58 ..
-rw-r--r--   1 Al  staff   118 May 17 08:34 Foo.sh
-rw-r--r--   1 Al  staff  2727 May 17 08:34 Foo.sh.jar
res0: Int = 0

Interactive debugging in Scala Play

Start with:

sbt console. At prompt, enter:

scala>

import java.io.File


import play.api._

import play.api.Play


val application = new DefaultApplication(new File("."), this.getClass.getClassLoader, None, Mode.Dev)

Play.start(application)

import play.api.Play.current


Play context becomes available on the console:

scala> conf.Global.esClient.search("mazda")
EmbeddedESServer.search: Setting future
res6: String = {"took":218,"timed_out":false,"_shards":{"total":5,"successful":5,"failed":0},"hits":{"total":246,"max_score":null,"hits":[{"_shard":0,"_node":"qffjZhZWTqObiZaTqQB7WA","_index":"twitter","_type":"tweet","_id":"523156744353370112","_score":null,"_source":{"filter_level":"medium","retweeted":false,"in_reply_to_screen_name":null,"possibly_sensitive":false,"truncated":false,"lang":"en","in_reply_to_status_id_str":null,"id":523156744353370112,"in_reply_to_user_id_str":null,"timestamp_ms":"1413565262298","in_reply_to_status_id":null,"created_at":"Fri Oct 17 17:01:02 +0000 2014","favorite_count":0,"place":null,"coordinates":null,"text":"New Mazda MX-5 - exclusive studio pictures http://t.co/mqUhp8Wf2w","contributors":null,"geo":null,"entities":{"trends":[],"symbols":[],"urls":[{

Also may use:

new play.core.StaticApplication(new java.io.File("."))

Passwordless SSH

# on client:
user=root
[[ ! -f ~/.ssh/id_rsa.pub ]] && ssh-keygen
for n in $CLUSTER_NODES; do ssh-copy-id -i ~/.ssh/id_rsa.pub $root@$n; done

The code above is roughly equivalent to the code below, which is included only to explain roughly what ssh-copy-id does under the covers. The first approach is strongly recommended.  

CLUSTER_NODES="39.96.64.13 39.96.64.15 39.96.64.17"

for node in $CLUSTER_NODES
do
     cat ~/.ssh/id_rsa.pub |
          ssh $user@${node} 'cat >> ~/.ssh/authorized_keys;
               chmod go-w ~/;
               chmod 700 ~/.ssh;
               chmod 600 ~/.ssh/authorized_keys ~/.ssh/id_rsa;
               chmod 644 ~/.ssh/id_rsa.pub' # ~/.ssh/known_hosts
done

# on server:
chmod go-w ~/
chmod 700 ~/.ssh
chmod 600 ~/.ssh/authorized_keys ~/.ssh/id_rsa
chmod 644 ~/.ssh/id_rsa.pub ~/.ssh/known_hosts

Set up SSH tunnel

Tunnel firewalled Elasticsearch port (9200) through SSH

Tunnel service out with access to server host:

ssh -N -f remoteuser@remotehost -L 19200:localhost:9200

Options:

• -N: Do not execute command (optional, mostly for security)
• -f: Go into background (otherwise remote prompt comes into terminal)
• -L: Local port mapping (local port 19200 will tunnel into remote 9200)

SOCKS proxy

ssh -D 8080 -f -C -q -N myuser@remote_ssh_server
-D 8080: Dynamic SOCKS server
-f : Fork into background
-C : Turn on compression
-q : Quiet mode
-N : No commands (required by -f)

Test host name resolution (python)

python -c 'import socket; print socket.getfqdn(), socket.gethostbyname(socket.

getfqdn())'

 

VirtualBox, CentOS eth0: transmit timed out; status 0073, resetting.

Error seemed due to unused virtual adapters having 'Cable Connected' checked:

1. Clear 'Cable Connected' in all unused adapters

2. Set 'Attached to:' to 'Not attached' in all unused adapters

3. service network restart

Hello Neo4j

service neo4j-service stop
service neo4j-service start


// all Page nodes named Automobile
match (n:Page{name:'Automobile'}) return count(*);
// = 24 // should be 1

match (n:Page) return n.id, n.name, count(*) limit 10;

n.id	n.name	count(*)
null	European_Car_of_the_Year	1
null	Hamm_AG	1
null	Ford_Motors	1

Set up static IP address

vi /etc/sysconfig/network/ifcfg-eth0 (SLES)

vi /etc/sysconfig/network-scripts/ifcfg-eth0 (CentOS)

 

vi /etc/sysconfig/networking/devices/ifcfg-eth0 (CentOS?), set values to:

BOOTPROTO=static
IPADDR=10.1.1.100
NETMASK=255.255.255.0
BROADCAST=10.1.0.255
NETWORK=10.1.0.0

Also, set up DNS server:

vi /etc/resolv.conf
search home
nameserver 10.1.1.1

May need to :

route add -net 0.0.0.0 netmask 0.0.0.0 gw 10.1.1.1

and

1. rm -f /etc/udev/rules.d/70-persistent-net.rules

2. Edit ifcfg-eth0 and remove the UUID and MACADDR (HWADDR):
vi /etc/sysconfig/networking/devices/ifcfg-eth0

3. Reboot (init 6)

Centos error on VirtualBox (Device eth0 does not seem to be present, delaying initialization)

1. rm -f /etc/udev/rules.d/70-persistent-net.rules

2. Edit ifcfg-eth0 and remove the UUID and MACADDR (HWADDR):
vim /etc/sysconfig/networking/devices/ifcfg-eth0

3. Reboot (init 6)

Install Elasticsearch on CentOS

Install RPM

sudo rpm -i https://download.elasticsearch.org/elasticsearch/elasticsearch/elasticsearch-1.3.4.noarch.rpm

ES is installed under /usr/share/elasticsearch
Config at /etc/elasticsearch/elasticsearch.yml

Start automatically
sudo /sbin/chkconfig --add elasticsearch

Start now
sudo service elasticsearch start

Disable IPv6 on CentOS

Method #1 (recommended)

Permanent: Edit /etc/sysctl.conf to:
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1

Runtime only:
sudo sysctl -w net.ipv6.conf.all.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.default.disable_ipv6=1

Method #2 (discouraged: it may interfere with SELinux and other components)

Add the following to /etc/sysconfig/network-scripts/ for the device
IPV6INIT=no

SCRIPT (caution!):

echo -e "net.ipv6.conf.all.disable_ipv6 = 1\nnet.ipv6.conf.default.disable_ipv6 = 1" | while read f; do if [ $(grep -c "$f" /etc/sysctl.conf) -eq 0 ]; then echo -e "\n$f" >> /etc/sysctl.conf; fi; done

Mount block device

How to mount a new block device (CentOS)

# Check current status: shows no partition info and disk ID=0x0

sudo fdisk -l

Disk /dev/xvdc: 214.7 GB, 214748364800 bytes
255 heads, 63 sectors/track, 26108 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000

# Create a partition. Answer prompts as indicated
sudo fdisk /dev/xvdc
n # new partition
p # primary
1 # partition no.
<enter> # accept start cyclinder default
<enter> # accept end cyclinder default (take whole device!)
w # write partition

# Verify. fdisk should show new partition info
sudo fdisk -l

Disk /dev/xvdc: 214.7 GB, 214748364800 bytes
255 heads, 63 sectors/track, 26108 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xb7719266

    Device Boot      Start         End      Blocks   Id  System
/dev/xvdc1               1       26108   209712478+  83  Linux

# Format new device
sudo mkfs -t ext3 /dev/xvdc1

# Mount it
[ ! -d /data ] && sudo mkdir /data
sudo mount /dev/xvdc1 /data

# Make permanent
sudo nano /etc/fstab
# edit to add:
/dev/xvdc1            /data                    ext3    defaults,noatime,barrier=0      1 1

Generate Elasticsearch test documents

for i in $(seq 1 100); do echo curl -XPUT \''http://roadrunner2:9200/test-001/doc/'${i}''\' -d \''{"sentence":"'$(head -c 56 < /dev/urandom | base64)'"}'\'; done | bash