Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

when cloning CentOS (minimal install + nginx, tomcat, postgres all not running) in a VMware virtual environment what unique identifiers are present on the cloned machine?

The information I have found so far is that when cloning, the new MAC address causes an a new interface to be added, so it is necessary to follow instructions here: That is, remove the 70-persistent-net.rules and update the MAC address in /etc/sysconfig/networking-scripts/ifcfg-eth*

But, is there any other unique properties which may cause problems, or even indicate that the machine is cloned? Is there some sort of System ID created by centos minimal during install.

The ones I can think of are:

  • If ssh server creates server creates keys during install these would be duplicated and that could be a security issue.
  • Any static IP set up
  • The MAC address set up

Would there be any others, even if they would not effect the usage of the cloned machine.

Thank you in advance!

-----edit-- I just found a page it seems to have a fairly complete list. Would any of these actually cause problems if they were on a shared network, or even on a cluster?

share|improve this question

2 Answers 2

up vote 3 down vote accepted

I'm currently running cloned CentOS Xen images, and only changing mac addresses and server keys between them. IPs and Macs are the only things that would cause networking issues.

The other thing to worry about is mountpoints - if you are mounting e.g. /var/www off shared storage, you (perhaps) want to make sure that all the cloned servers are each mounting their own /var/www directory.

share|improve this answer
    
Thank you very much! would you mind if I asked what architecture you are running, is it a load balanced cluster type of thing, or is it separate images hosting separate services? I know it doesn't have a huge impact, but it would be good to know what the standard practices are! –  Rqomey May 1 '12 at 15:28
    
Many of my images are running separate/independent services, but a couple pairs are behind the F5 load balancer, and a couple are running software that clusters itself, like RabbitMQ. I don't think that there are necessarily "best practices" independent of knowing anything about your use. One nice thing about VMs is that you can size them up rather than needing to cluster them for load, but then of course you lose the redundancy insurance of having a cluster. –  Nada May 1 '12 at 15:34
    
Thanks, we will be testing anyway, but it gives me hope that you are running both types of systems! –  Rqomey May 1 '12 at 15:53

I think both the links pretty much cover everything. Just one thing missing, fixing the disk UUID just incase.

Following are the steps to resolve your disk UUID issue

  • Run 'blkid' to get the new disk UUID /dev/sda1: UUID="5057334e-065e-40e2-9da0-16a282fbb0de" TYPE="ext4" /dev/sda2: UUID="bb5141fc-c43c-400a-bd3e-6e90483c8cd1" TYPE="ext4"

    • Update your /etc/fstab with the new UUID.
share|improve this answer
    
That's a great point, I will have to check how the storage is presented to the cloned machines (they may actually have the same uuid's) –  Rqomey May 1 '12 at 15:58

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.