How to install BPPM client on Linux server

Dear Friendz,

Thanks and appreciate your views and visits. A bit of work i did today on BPPM installation. steps and work done is given in detailed below.

What is BPPM :- it is a monitoring tool, which is to monitor the incidents on on any kind of server. It has a patroling agent which monitors the parameter on the server/,


1st Step :- we need to create an image on any linux box. and that same image will be copied to other servers through scp or winsc or using any tool which you use. I copied the same through

#scp -rp /tmp/image/ 192.168.2.23:/tmp/

this image is tranferred then we need to perform below steps.

2nd :- for creation of Image, we need to download the gzip file from BPPM site with specific version of OS like 6.2 incase of mine.
3. gunzip or unzip the same.
4. that will extract some files named as setup.sh.
5. chnage the permission of the file using
chmod 777 setup.sh
6. now run the same and that will open a window and will ask you to open the same through web browser.
you need to click and enter some details about root and its password and patrol user credentials too.
7. some more window will open, you can go with default option in image installation window.
8. once image is created successfully.
9. create patrol user account with

uid=201(patrol) gid=190(patrol)
10.

Steps for installation of BPPM on the Linux server.

 

 

1.       Copy the image folder from /tmp folder using scp or winscp. Then

 

Mount the DVD  of 6.2 64 bit and then install below packages.

 

First step :-  #yum clean all

                        #yum update all

 

Install below packages on the server.

 

#[root@ /]# yum install libstdc++.so.5* compat-libstdc++-33* -y

 

Loaded plugins: product-id, refresh-packagekit, security, subscription-manager

Updating certificate-based repositories.

Setting up Install Process

Resolving Dependencies

--> Running transaction check

---> Package compat-libstdc++-33.i686 0:3.2.3-69.el6 will be installed

--> Finished Dependency Resolution

 

Dependencies Resolved

 

==============================================================================================================================

Package                               Arch                   Version                          Repository                Size

==============================================================================================================================

Installing:

compat-libstdc++-33                   i686                   3.2.3-69.el6                     Server                   189 k

 

Transaction Summary

==============================================================================================================================

Install       1 Package(s)

 

Total download size: 189 k

Installed size: 716 k

Is this ok [y/N]: y

 

 

·         Create patrol user if not created already with below details.

#useradd –m –d /home/patrol –s /bin/bash

#passwd patrol

Password should be :- xxxxxxxxxxx

 

 

·         After this.  We have to run the ./install.sh from /image folder in /tmp location. After successful installation. We have to do below steps.

 

 

After installation, below are the post installation steps.

 

Check the shell of Patrol user, if correct then leave else.

 

If found shell other than /bin/bash for patrol user then Change it to:-

 

#usermod –s /bin/bash patrol :----- to get into proper directory of patrol user or login into it.

 

 1)  cd /opt/bmc/Patrol3

 2)  ./agent_configure.sh -d  (run as root only)

 3)  ./b1config9000.sh   (run as root ONLY)

 4)   chmod 777 /usr/adm    (run as root)

 5)   su - patrol  (switch to patrol user)

 6)  ./PatrolAgent -p 3181  (run as patrol)

 

cd /opt/bmc/Patrol3

./agent_configure.sh -d 

./b1config9000.sh  

chmod 777 /usr/adm

su - patrol    

./PatrolAgent -p 3181

 

That is it.

 

[patrol@Patrol3]$ ps -ef |grep -i 3181

patrol   22681     1  0 21:56 ?        00:00:00 PatrolAgent -p 3181

patrol   22713 22628  0 21:58 pts/1    00:00:00 grep -i 3181

[patrol@ Patrol3]$

 

 

Booting to single user mode HP UX

Hi All,

Thanks for your visits and comments. It always feels good to see you on my blog.

Question :- How to boot into single user mode or maintenance mode in HP UX?

To boot the server into single user mode,

1st.  We need to reboot the server and halt in after EFI boot and on HP UX boot.
2nd  As we halt the system into HP UX. we fell into HP UX > i.e. hpux prompt.
3rd  There we have to write

hpux> boot -is

4th. the server will boot the kernel.
5th. you will be fallen to # prompt withour any password/authentication.
6th. At # prompt, now you can start your maintenance work.
7th. For doing some activies like :-

we can get into single user mode.

a. To reset the password of root user, if forgotten or missed or if lost.
b. To edit fstab entry.
c. to edit files like password or shadow file. but for these we have to mount the root file system with below command.

#mount -o remount,rw /  :- this is to mount root file system with read and write access.

# mount -o remount,r /  :- to mount into read only mode.

d. for editing some of the corruption, etc. which is not allowing us to go to OS.

8. after editing or reseting root password, just type reboot or shutdown -ry from # prompt and system will boot to OS.

======================================================================
======================================================================

Booting to single user mode in LINUX

#init 1 :- from root prompt will bring to single user mode.

the if needs to done for reseting password from single user mode. then here we also need a DVD and get into maintenance mode. as the DVD boots up and ask for installation.

I amy not know other method, may be wrong here. if yes then please corrrect me. it will be very help ful for me too.

======================================================================


Thank you so much.

Love all, love sharing knowledge..
Amit Chopra

 

HP UX interview questions Part 1

Hi Friends,

Thanks for the visits. Glad to see your visits always and appreciate your responses. I learnt some questions today about HP UX, which can be asked in an interview.

Question 1. What is the difference between cp command and cpio command.
Answer :- the cpio command copies link files also with it. the same thing is also performed by fbackup tool. but the cp and scp command doesn't copy the link files. due to which application may or may not start due to unavailabilty of links or link files.
but the limit of cpio command is that, it can only be used to copy within the limit of 2 GB. after that it will fail or not do. where cp command leads in this and has no limit to copy data of any size.
  so as per the requirement, we can use this command.

Question 2. What is the meaning of below terms and their work, explain?

1. lock lun
2. lock disk
3. quorum server.
4. Arbitraror node.

how they differ.

Question 3. What are the conditions of Sevricegaurd Cluster failover?

Question 4. What are the important files while creating service gaurd cluster?

Question 5. please mention some commands for service gaurd cluster creation and setup?

Question 6. What is the work of cluster manager and work load manager?

Question 7. At what level or condition we cannot use Lock disk and Lock lun for servicegaurd?

Question 8. What is the command to start a cluster at both nodes and on a single node?

Question 9. what is tie breaking in cluster?

Question 10. What are the things required for creating a cluster?





Thanks and Best regards
Amit Chopra






 

Query for Strcit allocation Policy on LV

Hi Friends,

This was the error which i got after trying to extend the size of LV. i am not aware of this strict policy on LV or PG. but this is blocking me to extend the size of an LV. where i am having around 70 GB space in my vg00.

I came to know first time today that this policy, we can apply with below command and at the time of installation of server on root disk.

#lvchange -s ";;;;;;;;;"

Error :-

#
# lvextend -L 35000m /dev/vg00/11203
Warning: rounding up logical volume size to extent boundary at size "35008" MB.
lvextend: Not enough free physical extents available.
Logical volume "/dev/vg00/xxxx" could not be extended.
Failure possibly caused by strict allocation policy
#

I am trying to read some docs and information on it, will update once get some information.

Friends, you also please try to make me understand about this. how it works and how we have to apply it.

Thanks for the comments and visit.

Love all and Share Knowledge.
Amit Chopra

fsadm got stuck + HP UX

fsadm got stuck and did not give output for long duration. The fsadm command is run to take effect of extension or reduction of a file system.

EXTENSION of a File System

Question :- What is the to command to extend a File system HP UX?
Answer :- we have to use 2 commands :-
first :- lvextend -L "desired total size of LV in mb"  "name of the lv in /dev/vg00/lvolXX"
           fsadm -b  "desired total size of LV in mb" /name of the mount point


if command comes out properly or successfully then just run bdf. the extended size will be in front of you.

REDUCTION of a File system

Question :- What are the command to reduce the size of a file system.
Answer :- first of all be very much carefull while reducing the size of a file system. ensure you have a backup with you, using fbackup or any other too.

first :- fsadm -b "the desired reduced size which you want to keep" /mount point

next :- lvreduce -L "desired reduced size" /dev/vg00/lvolXX

if command comes out properly or successfully then just run bdf. the reduced size will be in front of you.
some times while extension and reduction, the fsadm command takes so much time that it seems it got hanged or the server become unresponsive. let it take its own time. i waited for 25 minute when i reduced 50 GB of spae from a file system. the fsadm didn't response for that duration. i was about to kill the porcess ID, but waited for some more time. it successfully come out and then i ran lvreduce.

if it is like 1 hour or more for waiting then you can kill the fsadm process id, with kill -9 also. and still if the processs is not getting killed then reboot is the option left. this I know, i may be lacking in information or knowledge. but would like to learn more for it.

ps -ef |grep -i fsadm

is the command to get the process ID or parent process ID.