Running Gauss On A Cluster

I created 3 test machine and made them into cluster.
On Virtual
"Virtual Machine Manager 0.9.0"
Machines them selves are CentOS 5.11.

Code:

<cluster name="mycluster" config_version="3">
   <clusternodes>
     <clusternode name="one" nodeid="1">
         <fence>
         </fence>
     </clusternode>
     <clusternode name="two" nodeid="2">
         <fence>
         </fence>
     </clusternode>
     <clusternode name="three" nodeid="3">
         <fence>
         </fence>
     </clusternode>
   </clusternodes>
   <fencedevices>
   </fencedevices>
   <rm>
   </rm>
</cluster>

When I start cman service. I see get this:

Code:

[root@one ~]# time service cman restart
Stopping cluster: 
   Stopping fencing... done
   Stopping cman... done
   Stopping ccsd... done
   Unmounting configfs... done
                                                           [  OK  ]
Starting cluster: 
   Loading modules... done
   Mounting configfs... done
   Starting ccsd... done
   Starting cman... done
   Starting daemons... done
   Starting fencing... failed

                                                           [FAILED]

real	5m7.353s
user	0m0.063s
sys	0m0.095s
[root@one ~]#

clustat:

Code:

[root@one ~]# clustat 
Cluster Status for mycluster @ Tue Jun  2 11:10:28 2015
Member Status: Inquorate

 Member Name                                            ID   Status
 ------ ----                                            ---- ------
 one                                                        1 Online, Local
 two                                                        2 Offline
 three                                                      3 Offline

[root@one ~]#

Thank you.

Hy All,

I have two nodes running RHEL 6.4 which are in cluster. I have clustered httpd service. Now I have certain application suppose X which reside in SAN partition. Now my job is to only automate this script to run by analyzing httpd status. I have successfully created the script apptrigger.sh whose content is given below,

-bash-4.1$ cat apptrigger.sh
a=0
b=0
c=0
d=0
a="service httpd status"
b=`eval $a`
##### CHECKING WHETHER HTTPD SERVICE IS RUNNING OR NOT #####
if [ "$b" != "httpd is stopped" ]; then
c="bash /switchapp/switch/ezlinkenterprise/bin/ezlink-cluster.sh status"
d=`eval $c`
##### CHECKING WHETHER EZ IS RUNNING OR NOT #####
if [ "$d" != "EzTaskMgr is Running" ]; then
ezstart
fi
fi

I have created this script in say test user. Now when I put this script to automate or run every minute through test user and defined its value in crontab -e of test user whose content is as shown below
$crontab -l
* * * * * /bin/bash /home/test/apptrigger.sh

but this script is running as root user so I cannot get the o/p
I think some environment variable need to be set. Anyone can guide me with this step? Thanks for your views and suggestion.

I have a hostlist and there are several hosts. I want to use ssh to connect to them. I want to see whether the host is available or not by the time length of ssh takes. If it take long than 5s (which means it is not available). Then stop it, and ssh to next host. Once a host is available, then output the name of host. I previously use nmap, but the IT security told it is not allowed to install it on the desktop on campus.
Code:

host_list="/home/campus27/zwang10/Desktop/cluster/program/hostlist"
HOSTS=`cat $host_list`
for line in $HOSTS
do 
timeout -5s `ssh $line`
done

The above script is all I can do.

Hy,
I am new to linuxquestion.org, however i am mid-level linux administrator.

As per my scenario, I want to cluster python and java services in two nodes running RHEL 6.4. Till now by watching tutorial I have seen application level clustering of linux for eg: webserver, mysql database and so on. However I haven't found any with python & java clustering.
I am already familiar with windows clustering having clustered ip of the two nodes. In my scenario, we have both the options open either going with active-active or active-passive cluster. So I want to some sort of idea to achieve linux os clustering and hence obtained clustered ip.
Further any idea on how this service will float from one node to another. Further, I cannot get to the fencing mechanism in Linux. Any idea for this will be helpful to understand.

Hope to get positive respond ASAP. Thanks in advance for those who help me with this.

I have serveral nodes related files under /var/lib/ directory as shown below:

Code:

machine001.cluster.org 
machine002.cluster.org
machine003.cluster.org
machine004.cluster.org

I am trying to write a shell script which first finds out how many number of files are present in this directory. Then it should fetch the first name like machine001, machine002, machine003 etc. Then it should fetches ipaddress, hostname and netmask under machine001 first, then store it in some variable. Repeat this for all the nodes. How will the script look like?

Hi, I'm pretty new to Linux.

I'm currently working on a project in my Linux Administration class, but have run into a bit of a bump.

I'm tasked with booting Linux (currently using CentOS 7) through the GRUB prompt. I've loaded the kernel, but I'm having an issue with the initrd.

I've tried both of these commands:
"linux /boot/initrd-plymouth.img"
"linux /boot/initramfs-3.10.0-123.el7.x86_64.img"

Neither of these commands have worked. It only returns the message "invalid magic number" for both of the commands. Any ideas on how I can fix this?

Thanks for any help in advance. It should help me understand Linux better!

I have many hosts as following. But those hosts are dual operating system (Linux and Windows). I always run program background. If someone is using Linux, it is fine. But if someone is using windows or the host is offline. Then, I cannot use ssh. The way I do preiviously is first ssh one by one. And then find the hosts which are offline or windows. And write them down one by one. And then ssh to hosts except them. Let us assume the number of programs is less than the total available hosts. Can someone write a shell script to output the all available hosts to a file like "host_available"?
Here is the host file.
https://www.dropbox.com/s/vbz6w864y3...tlist.txt?dl=0
I am using ssh to connect the computers on campus. If the computer I am trying to connect is offline or using windows, the ssh will take long time, and finally it will failed. I write a shell script to generate the hostlist
Code:

#!/bin/bash
for i in `seq -w 1 28`
do
echo "c15-0330-$i.ad.mtu.edu"
# I would like to add a command here to see whether ssh c15-0330-$i.ad.mtu.edu it succeeds or not. And then output a #file which contains the all available host.
done
for i in `seq -w 1 20`
do
echo "c28-0112-$i.ad.mtu.edu"
done
for i in `seq -w 1 20`
do
echo "c28-112a-$i.ad.mtu.edu"
done

I do not how to set a certain time to see whether connection is successful or not (see the comment in the shell script).

I am using the following two commands to output CPU and RAM usage on a Linux machine.

Code:

/bin/grep 'cpu ' /proc/stat | awk '{usage=($2+$4)*100/($2+$4+$5)} END {print usage "%"}'

/usr/bin/free | grep Mem | awk '{print $3/$2 * 100.0 "%"}'

My problem is that the output is like this

Quote:
5.33672%
13.9723%
Is there a way to output a single number? For example

Quote:
5
13
Thank you

Ubuntu 14.10 runs great on my laptop but I think I have dependency problems because EVERY time I install or remove software I get the following error message. However everything seems to run just fine...

"Package operation failed"
"The installation or removal of a software package failed."

...and details shows a massive list of gobbledygook with a lot of:

"insserv: Starting vpnagentd_init depends on rc.local and therefore on system facility `$all' which can not be true!"

I can post the whole message but I'll wait for feedback first.

Hi, I know that it is possible to get the last background pid using the $! command, however when I use it in my QT application I got a wrong result. Below follows an example where this procedure works
Code:

[linux@localhost ~]$ (sleep 6 && echo "A test") &
[1] 3312
[linux@localhost ~]$ echo $!
3312
[linux@localhost ~]$ A test

Now my Qt application, where I got a wrong result:
Code:

./dbsync &>/dev/null &
[1] 5047
[linux@localhost comm5]$ echo $!
5047

but using the ps command it is possible to verify that the pid related to the right process is not the 5047 but 5050.
Generally it is the process that comes with the command Code:

echo $!

added of two.

I use QCoreApplication in my program. I do not know if it has something to do with this error.

Any advice is welcome